Home

User Guide - Scientific Computing Associates

1. E Ec wave ser_wave c CC ser wave c o wav lm Parallel Version Linda Concurrent Wave Equation C Example FILE clinda_wave cl ER FILES draw_wave c make clinda wave cl O H d E ESCRIPTION This program implements the concurrent wave equation described in Chapter 5 of Fox et al 1988 Solving Problems on Concurrent Processors vol 1 A vibrating string is decomposed into points Each processor is responsible for updating the amplitude of a number of points over time At each iteration each processor exchanges boundary points with nearest neighbors This version demonstrates domain decomposition Note use compile option 1m Explanation of constants values 0 1001 values at time t oGldvaltO l001 values at time t dt newval 0 1001 values at time t dt LI NP EE ND IS MOM II I I 1 E 11 121 1 0 1 2 01 0 M E ei ee include lt stdio h gt include lt math h gt define MAXPOINTS 1000 define MAXSTEPS 10000 define NWORKERS 4 define PI 3 14159265 yx Routine for creating the X graph of the wave extern void draw wave double values MAXPOINTS 2 newval MAXPOINTS 2 oldval MAXPOINTS 2 Linda User Guide 145 PEE real main ag int tpoints total points on the line int nsteps number of tim
2. Linda Simple Array Program Makefile FILE make clinda array cl RIPTION see clinda array cl make f make clinda array cl To compile using tuple scope us linda tuple scope compiler option CFLAGS linda tuple scope FEFE HE TE FE FE TE FE FE HE FE FE HE TE FE HE TE FE EE EE HEH k HE HE EE EE HE EE EE EE EE HEH EH EE EE EE HE E E E E E E H ele elinda arrav el C S CFLAGS clinda array cl o array Linda User Guide 131 pi Calculation Serial Version A Serial pi Calculation C Version FILE ser pi calc c OTHER FILES make pi_calo e DESCRIPTION pi calculation example program C Version This program calculates pi using a dartboard algorithm See M Fox et al 1988 Solving Problems on Concurrent Processors vol 1 page 207 All processes contribute to the calculation with the master averaging the values for pi include lt stdlib h gt include lt stdio h gt void srandom int seed double dboard int darts define DARTS 5000 number of throws at dartboard define ROUNDS 10 number of times darts is iterated main double pi average of pi after darts is thrown double avepi average pi value for all iterations int i ni srandom 5 avepi 0 for i 05 i lt ROUNDS i 4 Perform pi
3. Note To compile using tuple scope us linda tuple scope compiler flags CCFLAGS linda tuple scope k HE FE FE EE ERE HE TE FE FE FE FE FE HE FE HE E TE FE HE EE EE HEH EH EH FE FE EE EE E TE FE E TE FE EE EE EE EE EE HE HE HE EE EH i E ele CFLAGS mm clinda_mm cl CC S CFLAGS clinda mm cl o mm Linda User Guide 141 Concurrent Wave Eguation Serial Version fe SERIAL Concurrent Wave Equation C Version FILE ser_wave c OTHER FILES make wave c DESCRIPTION 5 This program implements the concurrent wave equation described in Chapter 5 of Fox et al 1988 Solving Problems on Concurrent m Processors vol 1 s A vibrating string is decomposed into points In the parallel version each processor is responsible for updating the amplitude of a number of points over time M At each iteration each processor exchanges boundary points with nearest neighbors This version uses low level sends and receives to exchange boundary points include lt stdlib h gt include lt math h gt include lt stdio h gt include lt time h gt define MAXPOINTS 1000 define MAXSTEPS 10000 define MINPOINTS 20 define PI 3 14159265 void init param void void init line void void update void int nsteps number of time steps tpoints total points along string rcode generic return
4. This program calculates pi using a dartboard algorithm See Fox et al 1988 Solving Problems on Concurrent Processors vol 1 page 207 All processes contribute to the calculation with the master averaging the values for pi i incl incl lude lt stdlib h gt lude lt stdio h gt void srandom int seed doub le dboard int darts define MAXWORKERS 20 define DARTS 5000 number of throws at dartboard define ROUNDS 10 number of times darts is iterated real_main local variables double homepi value of pi calculated by current task pi average of pi after darts is thrown avepi average pi value for all iterations pi recv pi received from worker pisum sum of workers pi values int nworkers number of workers round nbr wrker i int worker get number of workers nworkers 0 while nworkers lt 1 nworkers gt 20 printf Enter number of workers between 1 and 20 n scanf d amp nworkers start up the workers for i 1 i lt nworkers i eval workers worker i printf starting worker d n i Start collecting the results for each round avepi 0 srandom 0 for round nbr 0 ro nd nbr lt ROUNDS round _nbr 134 Sample Programs pisum 0 pi Calculation master does pi calc too homepi dboard DARTS get result
5. float EX float cy int nts parms 0 1 0 1 50 real_main float u NXPROB NYPROB int nwrkers ntile ixmin ixmax idum i ix iz it worker void inidat prtdat zs Initialize grid ui printf Grid size X d Y d Time steps din NXPROB NYPROB parms nts printf Initializing grid and writing initial dat file in inidat NXPROB NYPROB u p edat NXPROB NYPROB u initial dat Compute tile extents putting grid data into Tuplespace Create workers for all but the right most tile ui out struct par s parms nwrkers 1 ntile 0 for ix 1 ix lt NXPROB 2 ix NXTILE ixmin ix ixmax ix NXTILE 1 ntile 1f ixmax lt NXPROB 2 start up the workers printf starting worker d n nwrkers eval worker worker ixmin ixmax nwrkers put out the initial data into tuple space out initial data ixmin amp u ixmin 0 NXTILE NYPROB Linda User Guide 153 Put grid boundary data into Tuplespace f out left amp u 0 0 NYPROB cut rigbt amp u NXPROB 1 0 lt NYPROB P Call a worker to compute values for right most tile idum worker ixmin ixmax Gather results from T plespace ui for 1 i lt ntile i i n result id ixmin Tixmax n result rxmin kulixminj 1013 printf Writing final d
6. When subroutine f is invoked using the Ca11 statement the variable val will have the value 1 since that value was assigned just prior to the call However when the subroutine is invoked using the eval statement the variable will have the value 5 since that was its initial value at the inception of the program eval Function Restrictions evaled functions may have a maximum of 16 parameters and both their parameters and return values must be of one of the following types C Linda int long short char float double Fortran Linda integer real double precision logical The first four C types can be optionally preceded by unsigned the Fortran types may include a length specifier e g real 4 Note that no arrays structures pointers or unions are allowed as function arguments Of course data of these types can always be passed to a process through tuple space The other fields in an eval are also limited to the data types in the preceding list plus string constants 24 Using the Linda Operations Linda Operations Under Fortran Linda subprograms appearing in an eval statement can be either subroutines or functions Subroutines are treated as if they were integer functions always returning the value 0 Functions must return values of type integer logical or double precision Intrinsic functions may not be used in eval statements Predicate Operation Forms inp and rdp inp and rdp are predicate forms of in and
7. inda User Guide September 2005 SCIENTIFIC COMPUTING ASSOCIATES One Century Tower 265 Church Street New Haven CT 06510 7010 USA 203 777 7442 fax 203 776 4074 lsupport LindaSpaces com The software described in this manual is distributed under license from Scientific Computing Associ ates Inc SCAI Your license agreement specifies the permitted and prohibited uses of this software Any unauthorized duplication or use of Linda TCP Linda ot the Linda CDS in whole or in part in print or in any other storage or retrieval system is prohibited The Linda software is provided as is without warranty of any kind either expressed or implied re garding the software package its merchantability or its fitness for any particular purpose While the in formation in this manual has been carefully checked SCIENTIFIC cannot be responsible for errors and reserves the right to change the specifications of the software without notice The Linda software is Copyright 1988 2005 SCIENTIFIC Computing Associates Inc This manual is Copyright 1989 2005 SCIENTIFIC Computing Associates Inc All rights reserved Linda is a registered trademark SCIENTIFIC Computing Associates Inc Tuplescope is a trademark of Yale University UNIX is a registered trademark of AT amp T The X Window System is a trademark of MIT All other trademarks and registered trademarks are property of their respective holders Manu
8. The only changes here are the two loops which create and terminate the worker processes The first for loop consists of num workers eval operations Each worker is passed a unique integer as its argument its worker ID number so to speak It will use this value to retrieve its own task from tuple space The final for loop creates one wakeup task per worker with its second field set to 1 This is a poison pill telling the worker to exit The loop also retrieves the passive data tuple the worker emits as it dies before generating the next poison pill This ensures that the master process will not terminate until all the workers have If the speed of the shutdown process were important then the in operations could be placed in their own loop so that all of the poison pills would be generated essentially at once Then the workers die in parallel with the master collecting the worker data tuples only after generating all of the poison pills In this program there is no master routine per se rather the master functions are split among three routines real main which creates and terminates the workers nb setup energy which places global i e non iteration specific data into tuple space and controls the parallel special charges calculation Linda User Guide 89 90 Case Studies nb energy which controls the nonbonded interaction energy calculation for each iteration Hereis nb energy setup void nb en
9. will match it and assign the value 512 to i assuming that i is an integer Similarly if j is an integer variable equal to 8 then the statement C Form Fortran Form in oube j Pa in coube T Ti will match it again if i is an integer assign the value 512 to i and remove the tuple from tuple space If more than one matching tuple for a template is present in tuple space one of the matching tuples will be used Which one is non deterministic it will not necessarily be the oldest the most recent or a tuple specified by any other criteria Programs must be prepared to accept any matching tuple and to receive equivalent tuples in any order Similarly repeated rd operations will often yield the same tuple each time if the tuples in tuple space remain unchanged If no matching tuple is present in tuple space then both rd and in will wait until one appears this is called b ocking The routine that called them will pause waiting for them to return Note that the direction of data movement for the in and out operations is from the point of view of the process calling them and vot from the point of view of tuple space Thus an out places data into tuple space and an in retrieves data from it This is similar to the general use of input and output in conventional programming languages Al data operations to and from tuple space occur in this way Data is placed into tuple space as tuples and data is read or retrieved from t
10. How ntsnet Finds Executables ntsnet locates the local executable the executable file that will run on the local node which is the node where the ntsnet command is executed in the following manner If a full pathname is specified on the command line that path specifies the exact location of the local executable For example the following command executes the network program hello worldin tmp ntsnet tmp hello world If only an executable name is specified then ntsnet uses the TSNET PATH environment variable to locate the executable file The environment variable s value should be a colon separated list of directories which are searched in order for required executables working just like the UNIX PATH vatiable If TSNET PATH is unset it defaults to usr bin linda This means that first the directory usr bin linda is searched followed by the current directory Linda User Guide 51 The location of the local executable can play a large part in determining the locations of the executable files to be run on remote nodes using ntsne s map translation feature described below These remote directory locations are also explicitly specifiable using the rexecdir resource application and node specific Here is an example Tsnet Appl Node rexecdir usr local bin Tsnet Appl moliere rexecdir usr linda bin Tsnet hello world Node rexecdir usr bin Tsnet hello world moliere rexecdir usr linda bin t
11. Linda User Guide 45 46 Using TCP Linda the command line contained a full or partial directory specification that location is searched for this configuration file If only an executable filename was given then the directories in the TSNET_PATH environment variable are searched in turn The optional local configuration file and map translation file are named tsnet config and tsnet map respectively If used these must be located in the user s home directory The global files tsnet config and tsnet map are located in the common 1ib subdirectory of the Linda tree for example in usr sca linda7 1 common 1ib if the Linda tree begins at usr sca linda7 1 Settings in the local files always take precedence over those in the global files and settings in the application specific file take precedence over the local file The global files may also be ignored entirely if desired Note that precedence operates on a setting by setting basis and not by an entire file The ntsnet configuration files contain entries which specify various execution parameters and desired characteristics Depending on the parameter to which they apply entries can vaty significantly in their ultimate scope they can affect The execution of any TCP Linda program The execution of a specific program on every node it uses The execution of any program on a specific node e The execution of a specific program only on a specific node The configurat
12. 0 for i 0 i nb n 1 i force update i amp force update il force update i gt x 0 0 force update i gt y 0 0 force update i gt z 0 0 in wakeup wakeflag workerid Keep in mind that this code executes at the same time as the master routine for this part of the calculation nb_energy_setup which itself waits for the workers to place their partial sums into tuple space If we return our focus to the master program thread once the initialization phase is complete real main enters its main loop The routine verl step begins each iteration eventually control passes to the routine nb energy The sequential version of this routine computes the nonbonded interactions the parallel version of nb energy performs the master functions for this part of the calculation nb energy begins by initializing some variables and then sending out the wakeup tuple for each worker along with the current coordinates of the atoms in the molecule void nb energy str coor force evdw elec esup elec evdw esup count 0 0 wake up workers for workerid 0 workerid lt num workers workerid out wakeup workerid workerid send out current coordinates on each round out coor coor o str n 1 The worker who has been waiting to receive the wakeup tuple retrieves it and checks the value in its second field If this value is not 1 then the next time step commences T
13. Arrays match other arrays whose elements are of the same type Thus an array of integers will match only other arrays of integers and not arrays of characters Similarly real arrays will not match integer arrays even when they contain the same length in bytes e Scalar types don t match aggregate types For example if a is an array of integers then a 1 won t match an integer but a 2 in C and a 2 in Fortran will Similarly in C if p is a pointer to an integer p and p 1 do not match the n array notation is discussed later in this chapter The corresponding fields in the tuple contain the same values as the actuals in the template Scalars must have exactly the same values Care must be taken when using floating point values as actuals to avoid inequality due to round off or truncation Aggregate actuals such as arrays which otherwise match must agree in both the number of elements and the values of all corresponding elements The following sections contain many illustrations of these matching rules Linda User Guide 27 Scalars Arrays The following C operations all place a tuple with an integer second field into the tuple space int i p ar20 i p amp i out integer 3 constant integer out integer i integer variable out integer p dereferenced ptr to int out integer a 5 element of an int array Out integer f 1 function returns an int out
14. Begin program execution Pause program execution Resume execution of a paused program Create edit and or compile a TDL program Save the current contents of tuple space to a file not available when Dynamic Tuple Fetch mode is in effect The file is named program N dump where program is the application name and N is a integer incremented each successive save operation End Tuplescope session Controls whether single step mode is in effect or not default is off Display Aggregates Controls whether the contents of aggregates are displayed in tuple displays off by default If Display Aggregates is not in effect aggregates in tuple displays appear as the word Block The format for an aggregate display is the one that was in effect when its tuple icon was Linda User Guide 123 opened if Dynamic Tuple Fetch is in effect or when it entered tuple space if Dynamic Tuple Fetch is not in effect Dynamic Tuple Fetch When in effect tuple contents are copied to Tuplescope only when requested This mode may speed up execution somewhat but it has the side effect that not all tuples are always continuously available for inspection as they are under the normal mode Reverse Execution Available in postmortem mode only Causes execution to run in reverse when in effect Exit Modes Menu Close the Modes menu The Debug Menu Edit program debug Edit a TDL program named for the current application The editor specified by the
15. Here are five independent matrix multiplication operations Certainly it would be nice to produce a parallel version of this program but the question is where to parallelize We could replace each matrix multiplication call with a parallel version ot execute all five calls simultaneously In some circumstances each of these solutions will be the right one Here however we d also like to execute some of those exp calls in parallel as well which would tend to favor the second approach creating a parallel version of gaus3 that might very well use a sequential matrix multiplication routine There is a third possibility as well gaus3 is itself called a large number of times Do 10 I 1 npts call gaus3 x m 10 Continue and again the calls are independent of one another It might be possible to execute some of these calls in parallel leaving gaus3 essentially untouched Whether this is a good idea or not depends on the likely value of the loop s upper limit npt s If npt s is typically say 8 then parallelizing at this point will limit the number of processor which the program could take advantage of to 8 and so it is probably better to parallelize gaus3 itself If on the other hand npt s is typically 500000 then this is a perfectly good place to focus attention and the job will be much simpler as well Linda User Guide 81 Database Searching 82 Case Studies This case study looks at a generalized database search program What
16. oj woddns sa day Command Options Many of these options have the same meanings as they do with other UNIX compilers C Suppress linking and only produces 10 object files for each Linda source file Dname definition Define a C preprocessor symbol clc only g Produce additional symbol table information for debuggers help Display a help message Ipathname Add the specified directory to the include list clc only linda option arguments Linda specific compiler directives Values for option are compile args s Pass the string s on to the native compiler when it is used to compile source files c args s Pass the string s on to the C compiler fle only info Print out the pathname of the Linda directory and the default size of tuple space link args s Pass the string s on to the compiler when used to link the executable main file Use specified file in place of cmain o or fmain o ts N Initialize tuple space to N 200 byte blocks This option is applicable in Code Development System code tuple_scope Prepare the object files and or executable to be run with the Tuplescope debugger This option can be abbreviated as t_scope Ix Link in the library libx a Ldirectory Add directory to the object file library search list used by the linker o outputfile Name the executable file as indicated v Display subcommands for each step of the compilation 110 Linda Usage and Syntax Summary w72 The
17. 4 0 double score double darts return pi HEE HEH GRA EH FE HE HH HE FE HE FE FE FE TE FE FE EH FE E TE FE HE EH FE HE FE FE HE FE HE TE FE E TE FE HE EE AA erial pi calc Makefile ILE make ser pi calc c ESCRIPTION See ser pi calc c USE make f make ser pi calc c LAST REVISED 4 18 94 Blaise Barney CC EAE AE E E AE AE AE FE FE FE AE AE AE RENI cc pi cales ser pl calce SqQC ser p calc c o pi calec 136 Sample Programs Matrix Multiply Matrix Multiply Serial Version KKK KK oko oko k oko k k kok ok k kok ok kok kok RRA RR RRA kok k kok k kok k Koko kok I I He I I arrays are stored passed C arrays are row major order but Fortran arrays are column major order Ck CK Ck Ck CK Ck CK Ck AA A X AA A X A A X AXA AX X A AX X AA X X A A X A A X ZA Z X A A XZ A A X A Z Z A A X Z A Z Z A A Z X KA HH KKK ko A KH include lt stdio h gt Serial Matrix Multiply C Version FILES Ser mm c OTHER FILES make mm c DESCRIPTION Sequential version of a matrix multiply To make this a parallel processing progam this program would be divided into T two parts the master and the worker section The master task would distributes a matrix multiply operation to numtasks 1 worker tasks NOTE1 C and Fortran versions
18. Bi return rd row_index row row for col index 0 col index lt N col index clump rd col index columns columnst tor 0 r lt clump r 4 result ptr result VE col index M for e 0 e lt clump t dot 0 rp rows r M ep columns M for i G7 1 lt M 42 p ttep dot rp icp result potrete dot Put a block of the result matrix into tuple space out result rows index result M clump end while end worker The worker begins by retrieving a task and exiting if all tasks are already done It reads the corresponding tuple from tuple space containing clump rows of A It then reads the columns of B in chunks of clump columns at a time as they have been partitioned into tuples For each chunk all corresponding elements of the result matrix are computed when all of the chunks of the columns array holding B have been read and processed the worker sends the completed chunk of the result matrix to tuple space and begins the next task The variable clump can get its value in any number of ways from a preprocessor variable from a command line argument as some function of the sizes of the matrices and so on It allows you to adjust program execution in a number of ways For example if local memory for worker processes is limited clump could be chosen so that the portions of A and B a worker
19. Linda User Guide 5 To put it another way a computer program doesn t merely solve a particular problem but rather it embodies a particular approach to solving its specific problem Making a parallel version potentially can involve changes to the program structure or the algorithms it implements or both The examples in this manual include instances of all three possibilities Once the work has been divided into pieces yielding a parallel program another factor comes into play the inherent cost associated with parallelism specifically the additional effort of constructing and coordinating the separate program parts This overhead often is dominated by the communication between discrete program processes and it increases naturally with the number of chunks the program is divided into eventually reaching a point of diminishing return in which the cost of creating and maintaining the separate execution threads overshadows the performance gains realized from their parallel execution An efficient parallel program will maximize the ratio of work proceeding in parallel to the overhead associated with its parallel execution This ratio of computation to communication is referred to as granularity Think of dividing a rock into roughly equal sized parts There are many ways to do it in fact there is a continuum of possibilities ranging from fine grains of sand at one end to two rocks half the size of the original at the other A parallel program
20. The value in the second field is i s current value Linda provides four operations for accessing tuple space plus two variant forms desctibed in Predicate Operation Forms inp and rdp on page 25 Operation Action out Places a data tuple in tuple space eval Creates a live tuple usually starting new process es in Removes a tuple from tuple space rd Reads the values in a tuple in tuple space leaving the tuple there For example this out operation places a data tuple with one string and two integer fields into tuple space C Form Fortran Form out eube 4 64 out cube 4 64 There are two kinds of tuples data tuples also called passive tuples like those we ve looked at so fat which contain static data and process tuples also known as ive tuples which are under active evaluation An eval operation creates a process tuple consisting of the fields specified as its argument and then returns This process tuple implicitly creates a process to evaluate each argument Actual implementations create a process only for arguments consisting of a simple function call and fulfilling some other conditions which we ll note later all other fields in the eval are evaluated sequentially While the processes run the eval s tuple is referred to as a live tuple as each process completes its return value is placed into the corresponding field and once all fields are filled all processes have completed the resulting da
21. Using TCP Linda This chapter describes the special considerations involved when executing Linda programs on a network of computers The network version of Linda is called TCP Linda and that terminology is used here as well Release notes for particular platforms often contain additional information relevant to the use of TCP Linda Running parallel programs on networks is complicated by issues such as process scheduling executable location remote execution facilities and the like ntsnet is a powerful flexible utility for executing Linda programs on a network designed to help you manage this complexity This section discusses the simplest possible case and is designed to enable you to get started running programs right away The remainder of the chapter covers more complex scenarios and the ntsnet features provided to handle them Normally TCP Linda runs the real_main process on the local system and evaled worker processes run on remote hosts This requires that it is possible to successfully execute an rsh command from the local host to each remote host without being required to enter a password Consult the man page for rsh or your system administrator if this condition does not hold true for your site In the simplest case the current working directory is a commonly mounted directory accessible by the same pathname from every host that you want to use it can be a permanent mount point or be auto mounted Given these assumpti
22. gogol flaubert u home net aurora home A Sample Network These concepts will become clear once we look at several examples of map translation We ll use the sample network shown in the preceding illustration which shows the location of users home directories for each node in the network for our initial examples On aurora users home directories are located in home the mount point for one of its local disks This same disk is normally statically mounted via NFS to three other systems blake chaucer and flaubert On blake it is also mounted at home on the other two systems it is mounted at a different directory location at u on chaucer and at u home on flaubert Home directories on erasmus are found in the local directory m On the node gogo the home directories from aurora are automounted as necessary at net aurora home Finally users don t generally have their own home directories on node degas when they run remote worker processes on this node they use tmp as their working directory which is what is listed in the diagram How ntsnet Finds Executables Consider the following command run by user chavez from the work subdirectory of her home directory on flaubert Le u home chavez work ntsnet test24g For this command to work properly and successfully execute on all of the nodes in the sample network we need to construct rules that tell ntsnet how to translate this working directory for eac
23. integer a dereferenced ptr to int Note that single elements from an array are scalars of the type of the array The constructs amp i and p are not included among these examples because each of them is a pointer to an integer they are treated as arrays not as scalars see the next section Thus the following tuple and template will not match even though p points to i an integer integer p 1 integer i Here are some example Fortran Linda operations involving scalars Real a 20 x Integer i Character 10 name out integer i out real x out real a 6 out character name Note that Fortran Character vatiables are scalars Atray handling within tuples and templates is vety easy Here are some examples of tuples and templates involving arrays char ap20 int len out arrayl ans in arrayl Pas cut t array2 ini arrayz 28 Using the Linda Operations asilo a len Fortran Specifying Tuples and Basic Tuple Matching Rules Dimension a 20 Integer len out arravyl a in arrayl a Qut array2 azl0 in array2 a len The general format for an array field in a tuple is name length where name is the array name and length is the length of the array number of elements As the out operations in the examples indicate the length is often optional When it is omitted the entire length of the array is assumed In C you must still include t
24. maxprocspernode Maximum number of 1 maxprocspernode n application specific mp n Scope getload Whether or not to use current system load averages when scheduling workers on nodes true getload getload application specific loadperiod The period in minutes over which to compute load averages when they are used for scheduling loadperiod mins m mins application specific threshold Maximum load allowed on a node if the normalized load exceeds this value then no wotker will be started 20 node specific speedfactor A number indicating relative CPU capacity compared to other nodes Larger values indicate increased ability to run multiple workers Used in computing the adjusted load average 1 0 node specific masterload Load that the master process puts on its node the local node Used in computing the adjusted load average masterload n application specific workerload Load that a worker process puts on its node Used in computing the adjusted load average workerload n application specific fallbackload Value to use if ntsnet is unable to obtain the current system load average Setting this resource to a large value will ensure that nodes that are down will be excluded 0 99 fallbackload 7 application specific available Whether a node is available or not useful for temporarily disabling a node without rem
25. the following command prepares the program test 24 for use with gprof cc o test24 pg test24 o Then you run the resulting executable in the normal manner Doing so will create a file named mon out prof or gmon out gprof in the directory from which the program was executed These files contain the profiling data obtained during the run You then run prof or gprof on the output files There can be a lot of output from both of these commands Among the most useful are the breakdown of time spent the number of times each routine was called and the call graph information where each routine was called from Here is an example of the first Stime seconds cum cum sec procedure file 29 2 235 9100 2 2 235 91 gaus3 gaus3 f 24 6 198 5800 53 8 434 49 dgemm mm dgemm mm s 13 0 105 1600 66 8 3539 65 f nc3 fune3 t 9 1 73 2500 75 8 612 90 tria tria tE 8 0 64 8500 83 9 6717 75 exp exp s Tad 98 5500 91 1 736 30 intar intarc t This display shows the total amount and percentage of CPU time used by each routine in decreasing order In this program 90 of the total execution time is spent in just 6 routines one of which is a matrix multiply library call About 8 of the time is spent in calls to the exponential function The following display is an example of a call freguency table polls calls cum bytes procedure file 20547111 68 53 68 53 480 exp exp s 768 0 00 68 54 17072 gaus3 gaus3 f 14 Overview of Parall
26. 0005 seconds So a work task that invovled 20000 bytes total of input and output data should run at least 3milliseconds in order to justify the communication time while a task involving a few bytes of input and output data should run at least 1 millisecond As new network interconnect technology is developed the granularity effects must be re examined The use of certain high speed switches for example may give networks performance characteristics almost identical to distributed memory parallel computers Forcing an eval to a Specific Node or System Type 68 Using TCP Linda While there is no way within an eval operation itself to force its execution on any particular node or type of system adding an additional function layer in front of the target routine can accomplish this Here is an example master for i 0 i lt NWORKERS i eval worker do worker do_worker get the hostname or architecture type via standard system call if strcmp host moliere worker 1 elseif strcmp host goethe worker 2 elseif strcmp arch p4 p4 worker and so on The eval operation calls a generic worker function do worker which determines the hostname or architecture type and then calls the appropriate real worker function Debugging TCP Linda Programs Debugging TCP Linda Programs There are two ways of debugging a TCP Linda program Use the ntsnet debug option Manually start program
27. 1 e when kaon is true kaon This option turns on the keep alive mechanism This is the default kaon This option turns off the keep alive mechanism loadperiod minutes This option specifies the number of minutes over which the machine load is averaged Typical values for loadperiod are 1 5 and 10 The default is 5 masterload oad This option specifies the load that the master real main process is considered to put on the node The value specified can be any real number gt 0 The default is 1 Typically 1 or some smaller fraction is used If the master process uses much less CPU time than the workers the master load should be set smaller than the worker load maxprocspernode number This option specifies the maximum number of Linda processes started on any given node the application is running on On the local node maxprocspernode includes the master The default value is 1 mp number Synonym for maxprocspernode n minworkers maxworkers This option specifies the acceptable range of the number of workers that the application can run with If maxworkers is omitted it is set to the same value as minworkers ntsnet initially starts up the number of workers equal to the maximum of the minworkers and maxworkers resoutce values The master then waits as specified in the minwait and maxwait resources for the workers to join the execution group If at least minworkers join before the maxwait interval has elapsed execution
28. 119 message passing 7 8 messages increasing number 67 minwait resource 62 119 minworkers resource 62 119 molecular dynamics 86 multidimensional arrays 32 37 N named common blocks 33 native compiler passing options to 110 nice resource 51 119 node names in ntsnet configuration files 48 node set 61 nodefile 50 nodefile resource 51 119 nodelist resource 50 61 119 nodefile value 50 ntsnet command 21 bcast options 111 cleanup options 59 111 d options 59 112 distribute options 59 112 getload options 65 112 th options 51 112 thigh options 51 112 tkaon options 112 redirect options 67 114 Tsuffix options 60 114 translate options 58 114 tuseglobalconfig options 67 114 tuseglobalmap options 67 114 tv options 67 115 verbose options 67 115 tveryverbose options 67 115 tvv options 67 115 appl option 48 111 bcastcache option 111 configuration file format 116 configuration files 45 configuration information sources 45 debug option 69 112 delay option 67 disabling global configuration files 67 fallbackload option 65 112 help options 112 kainterval option 112 loadpetiod option 65 113 m option 65 113 masterload option 65 113 maxprocspernode option 65 113 mp option 65 n option 62 113 nodefile option 51 113 nodelist option 51 113 opt option 50 114 options vs resources 48 p option 52 61 114 process scheduling 61 return value 106 searching for exe
29. 36 image rendering 73 in 11 12 21 105 initialization workers and 75 inp 25 105 interprocessor communication 6 K kainterval resource 118 kaon resource 117 118 L language definition 105 length of array 28 lexit 37 106 lhalt 37 106 libraries 110 linda eval 26 linda in 26 linda inp 26 linda out 26 linda rd 26 linda rdp 26 Linda Code Development System 95 Linda directory tree 46 Linda model 10 linda prefix 105 LINDA CC environment vatiable 122 LINDA CC LINK environment variable 122 LINDA CLC environment vatiable 95 122 LINDA PATH environment vatiable 122 LINDA PROCS environment variable 107 123 linda rcp shell scripts 118 linda rsh shell script 118 lindarcparg resource 118 lindarsharg resource 118 linking 110 passing switches to 110 linking with clc 20 lintoff 106 linton 106 live tuple 11 load balancing 10 loadperiod resource 63 65 118 loffexit 107 lonexit 107 Iprocs 107 M map translation 52 121 configuration files 53 disabling 120 master becoming worker 38 39 distributed 89 waiting until workers finish 19 master worker paradigm 8 73 masterload resource 64 65 118 matching rules 26 37 matrix multiplication extended example 77 under distributed data structures 9 under message passing 7 8 maximum fields in tuple 26 maxnodes resource 118 maxprocspernode resource 65 66 118 maxwait resource 62 70 117 119 maxworkers resource 62
30. Assignment C Version ILE clinda array cl OTHER FILES make clinda array cl D ESCRIPTION In this simpl xample the master task initiates a number of tasks indicated by user input It then distributes portion of the array into tuple space to be worked on by the workers Each worker task retrieves data indicating what portion of the array the worker should update and performs a simple value assignment to each of its elements The value assigned to each element is simply that element s index in the array l Each worker task then puts its array back into tuple space and will continue retrieving work assignments until it is notified to stop As the master receives back each portion of the array selected elements are displayed When it has received all portions of the array It will put out a poison pill to notify the workers to stop work EOKCKCKCKCKCkCkCK Ck kCk ck kCk Ck kCk RR k kok RR RARA RRA CK Ck kCk Ck kCKCk KCk Ck KCk k kCk Ck Ck kc k Ck kc k k ck kok ck kok ck kok I e ke include stdio h define ARRAYSIZE 1000 define MAXWORKERS 20 real main int argc char argv Local Variables 128 Sample Programs Array Assignment int numworkers index i nt data ARRAYSIZE result ARRAYSIZE int extrachunk id TE data_index int chunksize nbr_chunks ime worker KKK RIA RARA RARA RR KK K initial lizatiohs KKKKKKKKKKKKKKKKKKKKKKKKK
31. DESCRIPTION Example Heat Egaution Domain Decomposition C Version l make clinda_heat2d cl This example is based on a simplified two dimensional heat equation domain decomposition The initial temperature is computed to le of the domain and zero at the boundaries The at zero throughout the simulation During the an array containing two domains is used these domains alternate between ol The paral advanced worker al in time by LSO kk hk A 0X 0 006 006 06 09 00 0X 0 152 Sample Programs llel version decomposes the domain into tiles note that the code assumes that NXTILE quotient is the number of parallel processes Id data and new data and each tile is a separate process The master process performs as a NXPROB NYPROB and NXTILE may be varied to study performance E evenly divides NXPROB 2 This In the C version the X 2D Heat Equation tiles are actually strips This is because C Linda does not have the rich Fortran 90 array section notation that is available in Fortran Linda Two data files are produced an initial data set and a final data set An X graphic of these two states displays after all calculations have completed eK k ok k I I kCk Ck kCk Ck kCkCk KCkCk kCkCK Ck kCKCk kCkCk kCkCk kCkCk kok k kok k Ck kc k Ck kck ck kok I I ke ke include lt stdio h gt define NXPROB 11 define NYPROB 11 define NXTILE 3 struct Farms
32. Index SCIENTIFIC COMPUTING ASSOCIATES One Century Tower 265 Church Street New Haven CT 06510 7010 USA 203 777 7442 fax 203 776 4074 lsupport LindaSpaces com
33. Integer len out array2 b 20 in array2 d len In both cases the first twenty elements of array d are set to the value of the corresponding element in array b and the variable len is set to 20 This requirement makes sense since these constructs can be used with arrays of different sizes within the same program The following out operations yield identical tuples C int p al201 p a cut array al out array 20 7 out array pi20 37 Fortran Integer a 20 out array a out array gr out array ai20 The length is required in the third C example pointers must always specify an explicit length You could create a tuple containing the first ten elements of a with any of these operations C ink p ap20 BL20 10 l n pu out ten elements a 10 out ten elements p 10 30 Using the Linda Operations Fortran Fortran Fortran C Fortran Specifying Tuples and Basic Tuple Matching Rules Integer a 20 b c 10 len out ten elements a 10 and retrieve it with any of these operations Len elements a len Len elements p len ten elements b in inti in in ten elements c in ten elements a len in ten elements a All of the operations will retrieve ten integers and len will be assigned the value 10 for those operations including it Note that omitting the length variable is allowed and
34. It is useful for protecting time consuming system calls from being interrupted Interrupts should not be disabled for long periods linton and lintoff calls can be nested For many implementations this is a no op Restores the interrupts associated with tuple space handling see lintoff below For many implementations this is a no op The clc and flc Commands flloffexit hd Deletes a specific handler from the list of handlers set up by lonexit where hd is the handler descriptor returned by lonexit Returns 0 on success and 1 on failure descriptor out of range or not referring to an active termination handler lonexit p a Names a routine to be called after a Linda process calls lexit Ihalt or P returns normally The routine p is called as p status a where status is the argument with which return lexit or Ihalt was called and a is typically the address of an argument vector although it also can be an integer value Multiple calls can be made to lonexit specifying up to 16 termination handlers which are called in reverse chronological order i e the last specified routine is called first lonexit returns a unique non negative termination handler descriptor upon success or 1 if the termination handler could no be stored f lprocs Returns the total number of processes that have joined the computation including the master process In the Code Development System this function is not meaningful and it return
35. Linda is easy to use Conceptually Linda implements parallelism via a logically global memory virtual shared memory called zup space and a small number of simple but powerful operations on it Tuple space and the operations that act on it are easy to understand and quickly mastered In addition the C Linda and Fortran Linda compilers support all of the usual program development features including compile time error checking and runtime debugging and visualization How to Use this Manual If you are new to parallel programming Chapter 1 Overview of Parallel Programming and the Linda Model provides a general introduction to the subject and to Linda It also introduces tuple space the Linda operations and other essential concepts Linda User Guide 1 For a more in depth look at parallel programming and algorithm development SCIENTIFIC recommends the book How to Write Parallel Programs A First Course by Nicholas Carriero and David Gelernter which is available in PDF format at no cost on the SCAI Website at www lindaspaces com See the Bibliography for the complete citation for this book and related works If you are using Linda to create parallel programs Chapter 2 Using the Linda Operations describes the Linda operations in detail It explains the program compilation and execution process provides a couple of simple example programs and includes an extended discussion of tuple matching rules and restrictions If y
36. Strinps idees eere RE yee ease eens ee KK ERG NU RETE ca 35 Anonynious Fotrmals 12452 dpa dae eae reed sodn Qao Qa nde tacite 36 Fixed Aggregates IN C esee sese cies kee epee el cn Re R ooo vd ea e Rus 36 Termination of Linda Programs sss 0 0 eee eee eee 37 Example Freewake 24 cues kode ce oe He a a HERR oe eie n 38 Linda User Guide i 3 USING TCP Linda x ose y Zea dates hail REDUCES 43 Quick Start d EM PM LIP TM 43 Mhacntsnet Do6esu cse coe iisa la gane NS 44 Using the ntsnet Command 00000 k 45 Customizing Network Execution eee eee ee 45 ntsnet Confiouration Piles s uro tes sieves e ete ds a Peeters 45 Resource PES vue ine rte teer snes BO Ba Ep SMS 49 Determining Which Nodes a Program Will Run On 50 Specifying Execution Priority rears a aa ey aa eee eee eee es 51 How ntsnet Finds Executables 44 4 4444 51 About Map Translation ees ee eh eee l ese mr n 52 The Map Translation File 53 Map Translation Entry Wildcards 0 ee eee eee 58 Distributing Executables 0 0 4444040000000 k k 59 Architecture Specific Suffixes socrus e iai e een 59 Specifying the Working Directory for Each Node lt 61 Permissions and Security Issues s oona nai p a cee etter ees 61 ntsnet Worker Process Scheduling 4444444400000 61 Forming The Execution Group 44444444 eee 61 Selecting Nodes for Workers ce
37. The default is 0 99 The value specified can be any real number gt 0 If failure of the RPC call indicates that the node is down this option can be used to set fallbackload to a very large value effectively making the node unavailable to ntsnet This option indicates that ntsnet should use load average information when scheduling workers on the nodes in the network This is the default This option indicates that ntsnet should not use load average information This can be used to make worker scheduling consistent between different runs of ntsnet It also makes sense if the rstatd daemon is not available on the network Synonymous with high high This option causes ntsnet to display the usage message This option causes all workers to be run at normal priority and causes Linda internodal communication to run at full speed This is the default This option causes all workers to run at a nice ed priority unless specifically overridden on a per node per application basis using the nice resource It also causes Linda internodal communication to be throttled to avoid flooding the network kainterval seconds 112 Linda Usage and Syntax Summary Specifies how often in seconds each Linda process sends out a keep alive message The default is 100 seconds The range of legal values is 100 to 31536000 one year The range is silently enforced This The ntsnet Command resource is only useful when the keep alive mechanism is used
38. This method is used to study the very large molecules of commercial and biological interest typically containing thousands of atoms The atoms in these molecules are constantly in motion this movement results in changes in the overall molecular structure which can in turn affect the molecule s properties A molecular dynamics simulation calculates the changing structure of a molecule over time in an effort to understand and predict its properties These calculations are carried out iteratively solving for the total molecular energy and the forces on and positions of each atom in the molecule for each time step In general an atom s position depends on the positions of every other atom in the molecule making molecular dynamics calculations require significant computational resources This case study illustrates the following techniques Per iteration worker wakeup Cleaning up tuple space Distributed master functions Since the original program for this case study is very long we ll only look at the central portions as we examine how it was parallelized with C Linda This program required changes to several key routines and illustrates using C Linda to parallelize the calculation setup work as well as the computation core The diagram below presents a schematic representation of the sequential version of the computation After performing some initial setup steps in which it reads in and stores the data for the calculation and
39. We ve now introduced all the pieces of the ntsnet configuration file The following sections will introduce many of the specific resources in the context of TCP Linda execution scenarios Determining Which Nodes a Program Will Run On 50 Using TCP Linda Two resources control what nodes a given application will run on First the nodelist resource which takes a list of nodes as its value specifies a node list for a given application Here are some examples Tsnet Appl nodelist chem gauss newton descartes Tsnet hello world nodelist gauss moliere dalton avogadro The first line specifies the default set of execution nodes for TCP Linda programs in addition to the local node The second line specifies a different set for the application hello world which overrides the default value set in the first line Duplicates are automatically removed from node lists Variant name forms for the same node the full name and the nickname for example are also discarded the first one is kept In such cases a warning message is printed The nodelist resource can also take the special value nodefile This indicates that the contents of the file specified in the nodefile resource contains the list of nodes to be used one name per line If nodefile has not been given a value then the file tsnet nodes in the current directory is used The value for nodelist defaults to nodefile so ignoring both of these resources will result in th
40. a far better programming practice to use a counter or semaphore tuple in situations where inp or rdp seems called for Consider this C Linda example if rdp globals x y z 0 do globalsix yvy z Linda User Guide 25 If the globals tuple is not available in tuple space then rdp returns 0 and the process computes the globals itself Simply doing a rd would result in blocking if the globals tuple weren t available and recomputing them can be faster than waiting for them although slower than reading them The same effect can be accomplished via a tuple that can take on one of two values for example globals ready 0 or 1 The master process outs this tuple with value 0 at the beginning of the program and the process that computes the globals and sends that tuple to tuple space also ins this tuple and outs it again changing its second field to 1 Then the preceding code can be replaced by rd globals ready 1 if 1 pai globals x y T else do globalsix v z C Linda Alternate Operation Names The alternate names linda in linda rd linda out linda eval linda inp and linda rdp are provided for cases where the shorter names conflict with other program symbols Each alternate operation name begins with two underscore characters Specifying Tuples and Basic Tuple Matching Rules This section discusses tuples and tuple matching rules in more detail and includes examples that use a variety
41. an optimization designed to improve performance by detecting patterns in tuple usage and attempting to place tuples on those nodes where they will eventually be used If successful this optimization produces significant communications savings This feature can be disabled by setting the value of the redirect resource application specific to false its default value is true The value of this resource may also be set with the redirect redirect command line options Disabling Global Configuration Files The useglobalconfig and useglobalmap resources both application specific specify whether to use the entries in the global configuration file and global map file in addition to the local files In any case local file entries take precedence over those in the global files The default value for both resources in true Command line options are available for both resources useglobalconfig useglobalconfig and useglobalmap useglobalmap Generating Additional Status Messages ntsnet can optionally display informational messages as program execution proceeds Whether and how frequently messages are displayed are controlled by the verbose and veryverbose resources both application specific Both are mainly useful for debugging configuration and map files and both default to false The command line options verbose verbose abbreviable to v v and veryverbose veryverbose or vv vv can also be used to specify these resources Process Ini
42. array 28 literal strings in 11 live 11 matching 12 13 matching rules 26 37 max of fields 10 maximum fields 26 multidimensional arrays in 32 pronunciation 10 redirection 67 scalars in 28 strings in 35 structures in 33 varying length structures in 35 tuple classes 13 tuple space 10 cleaning up 94 discarding data 36 operations 11 results when full 22 size under TCP Linda 22 Tuplescope Break button 99 buttons 123 clc compiler options 110 Continue button 99 control panel 96 Debug button 101 Debug menu 124 display 96 dynamic tuple fetch 99 exiting from 96 99 icons 98 invoking 95 menus 123 Modes menu 98 123 program preparation 95 Ouit button 96 99 reguirements 95 Run button 99 run modes 99 single step mode 99 TDL 101 using dbx with 100 varying execution speed 96 viewing ageregates 98 viewing tuples 98 X Toolkit options and 95 Tuplescope Debugging Language 101 syntax 101 124 types allowed in tuples 26 typographic conventions 3 U useglobalconfig resource 67 120 useglobalmap resource 67 120 user resource 61 121 username specifying an alternate 61 V varying length structures 35 verbose resource 67 121 veryverbose resource 67 121 virtual shared memory 8 Ww watermarking 84 worker compated to task 9 initiated by eval 11 repeating setup code 75 wakeup technique 92 workerload resource 65 121 workerwait resource 70 117 121 X X Windows 95 Linda User Guide v vi
43. be easily modified for different environments Changing the granularity level then becomes as simple as changing a few parameter definitions This technique complements the preceding one and both can be used in the same program These are the major issues facing any parallel programmer In the next section we ll look at three different approaches to creating parallel programs and indicate how Linda is situated with respect to each of them 6 Overview of Parallel Programming and the Linda Model Approaches to Parallel Programming Approaches to Parallel Programming There are two main challenges facing any parallel programmer How to divide the work among the available processors Where to store the data and how to get it to processors that need it Two radically different approaches to these problems have emerged as the dominant parallel processing paradigms They are message passing and distributed data structures implemented in virtual shared memory Message Passing Message passing focuses on the separate processes used to complete the overall computation In this scheme many concurrent processes are created and all of the data involved in the calculation is distributed among them in some way There is no shared data When a process needs data held by another one the second process must send it to the first one For example let s again consider a matrix multiplication calculation A B 2C A message passing version might cr
44. calculates the sums of special charges and some other quantities for the molecule the program enters its main loop For each time step one loop iteration the program must calculate the bonded and nonbonded interactions among all of the atoms in the molecule The bonded interactions occur between atoms that are directly connected together by a chemical bond and the nonbonded interactions are the effects of atoms that are not bonded upon one anothet s position The latter take up the bulk of the time in any molecular dynamics calculations because they are both more numerous and more complex than the bonded interactions Molecular Dynamics Read data from disk H Setup H c u alloc memory N E all gt H build matrix C Se sum changes Repeat NSteps C C Calculate bonded interactions times Calculate nonbonded interactions For this molecule the oxygen atom s only bonded interaction is with the carbon atom it is connected to A molecular dynamics calculation will also compute the effects of its nonbonded interactions with every other atom in the molecule Pm Sequential Version of the Molec Dyn Program amp Bonded vs Nonbonded Interactions Calculate new atomic coordinates Here is a simplified version of the original main routine main argc argv T 300 0 process_args argc argv Read data Initialize parameters amp data structures verl init str coo
45. calculation on serial processor pi dboard DARTS avepi avepi i pi i 1 prints After 3d throws average value of pi 10 8f n DARTS i 1 avepi KKK k kok k oko k kok ko k kok k oko k A A I A A kok k kok I I e ke k k dboard KOKCKCKCKCKCKCKCk kCk Ck KCkCk kok k Ck kCK Ck kCkCk kCkCk KCkCk kok k Ck kCK kok k k kCKCk kok kok k Ck k Ck k kc k Ck kck ck k kk k kk ckck kc kk ke e kx x fdefine sqr x x x long random void double dboard int darts double x coord x coordinate between 1 and 1 y_coord y coordinate between 1 and 1 pi je pi ES random number between 0 and 1 int score number of darts that hit circle n unsigned long cconst 132 Sample Programs used to convert integer random number between 0 and 2 31 to double random number pi Calculation between 0 and 1 cconst 2 lt lt 31 1 score 0 KOR KKK KK KR KK KK AK I A A I I IK kk for Throw darts at board Done by generating random numbers between 0 and 1 and converting them to values for x and y coordinates and then testing to see if they land in the circle If so score is incremented After throwing the specified number of darts pi is calculated The computed value of pi is returned as the value of this function dboard Note the seed value for rand is set in pi calc KCKCKCKCKCKCKCKCk kCk Ck kCkCk kCkCK Ck kCK Ck kCk Ck kCkCk kok k
46. chavez work which will in turn be translated to the appropriate directory when ntsnet starts remote Worker processes If we want to run the ntsnet command on node gogo however we must create an additional rule Home directories on gogo are automounted from aurora on demand When referred to in the context of a process starting from a remote system their location can be written as in the first rule thus when a remote process is initiated on gogo from flanbert the current working directory for the remote node is correctly translated to net aurora home chavez work Linda User Guide 55 56 Using TCP Linda However if the ntsnet command is run instead on gogo from this same directory the literal directory location what is returned by pwd or the get cwd system call is what is used for map translation in this case using the actual automounter mount point tmp mnt net aurora home chavez work Thus the translation from generic to remote directories is handled correctly by the first rule and what is needed is a rule for translating this local directory to a generic directory This is the function of mapto entries The following entry maps the local directory on gogo to the same generic directory we ve been using mapto home gogol tmp_mnt net aurora home With this rule in place running ntsnet from gogo will be successful and the remote working directories we ve considered so far will be set appropriately Sinc
47. code double values MAXPOINTS 2 values at time t oldval MAXPOINTS 2 values at time t dt newval MAXPOINTS 2 values at time t dt fe Obtains input values from user void init param void char tohar s set number of points number of iterations tpoints 0 142 Sample Programs Concurrent Wave Eguation nsteps 0 while tpoints lt MINPOINTS tpoints gt MAXPOINTS printf Enter number of points along vibrating string n Scant ss tehar tpoints atoi tchar if tp ints lt MINPOINTS tpoints gt MAXPOINTS printf enter value between d and d n MINPOINTS MAXPOINTS while nsteps lt 1 nsteps gt MAXSTEPS printf Enter number of time steps n Scant Ss taharis nsteps atoi tchar if nsteps lt 1 nsteps gt MAXSTEPS printf enter value between 1 and d n MAXSTEPS printf points d steps d n tpoints nsteps je All processes initialize points on line void init line void int ip Te E double x fac calculate initial values based on sine curve fac 2 0 PI k 0 for j 17 j lt tpoints j ktt 1 x double k double tpoints 1 values j sin fac x for i 1 i lt tpoints i oldval i values i M Calculate new values using wave eguation void do math int i 1 double dtime c dx ta
48. consisting of the entire array b will match The last example is a bit tricky in this case the out operation creates a tuple with the first element of a as its second field and any integer can be used in a formal to retrieve it 32 Using the Linda Operations Specifying Tuples and Basic Tuple Matching Rules Fortran 90 Array Sections Fortran Linda recognizes a subset of the array syntax used in the Fortran 90 standard within its operations This syntax provides an alternate way of referring to arrays and their subsections and may not be combined with the name ength notation we ve considered so far Array subscript references in Fortran Linda may take the following form ifirst ilast istride where jfirstis the smallest index of a slice of the array last is the largest index in the slice and stride is the stride if omitted the stride defaults to 1 A full array section is specified with this type of expression for each dimension in the array The shorthand form of a single colon alone refers to the entire range of values for that array dimension with stride 1 Here are some examples r sl a 100 100 15007 100 100 ont whole array sa 1 100 1 100 out second row b 2 2 ott divide an Z part 1 a t 2 250 outi i dryrde in 2 part 2 a t t Sle every other b 1 100 2 1 100 out The first out operation places the entire array a into the tuple space The second places only the s
49. cud este hm eee iy E Dep epi qe e RHET on 98 Viewing Process Information sa tiyini ia eiaei nia e en 99 Tuplescope Run Modes eee ee n rito ttes t tnt oa Enis ea hy 99 Using Tuplescope with a Native Debugger 100 The Tuplescope Debugging Language 4 4 4444 101 TDL Language Syntax bss cic caw ee ry va eee K y ee a E a 101 6 Linda Usage and Syntax Summary 105 Lida Operations eriep ikoe A V oe bie toe Woes Raa pob bosse ee ray RS 105 Formal GC Litda Syntax jake side sti ath eve nasal a dee p Sisk Wane 105 Timing Functions e o er oo blew See eile a eee RR Sees 106 Support Functions 5e ved le AW aa a eR rre ee 106 The cl and fle Commands creerse se ta Ree ond ose Be 107 Commarid Synta usb bon b osad une breed ke e aw edn EM Ter dees 107 Command Options Lesen e gece aaa ees NE hehe ee lado old ban ode 110 The ntsnet Command ie eh RR zde wie e Ae rt Re ia 111 DYNA corem o e eO EET EUN Eee ure diode re tede uen 111 Parameters tentar dde aeu sd vede Aa Da ol tated dE Orbe ats nt 111 Options Syntax Convention cilisis enn 111 Command Options aises ein e Re var pe d ei eee a ea m San prc a eta 111 ntsnet Configuration File Format lt ccc cece k 116 Resource Definition Syntax lice ee 116 R SOUtCeS 5543 scai o oec eu Dei d E oo oda 116 Map Translation File Format cece eee k 121 Environment Vatiables eser err A 6 oba EE ES 122 Tuplescope Referenc
50. example ntsnet can draw its configuration information from a variety of sources These sources are in order of precedence command line options ntsnet s application specific configuration file if any ntsnet s local configuration file ntsnet s global system wide configuration file ntsnet s built in default values 9 When they do not conflict the settings from all of these soutces ate metged together to create the ntsnet execution environment We ll cover each of these items separately in the context of actual execution tasks See the Linda Usage and Syntax Summary for a complete reference to all command line options and configuration file resources and formats ntsnet Configuration Files ntsnet uses several configuration files the global local and application specific configuration filee we ll use this term to refer to the specific files rather than in a generic sense from this point on which define program execution characteristics and the local and global map translation files which define directory equivalences on the various potential execution nodes in the network ntsnet first looks for an application specific configuration file named tsnet config appliration name where application name is the name of the application being executed with ntsnet application names will be discussed shortly ntsnet looks for an application specific configuration file in the following way first if the executable on
51. execution on each node This section looks at each method in turn both discussions assume that you have properly prepared executables for use with a debugger by compiling them with g ntsnet s Debug Mode The debug option to the ntsnet command initiates TCP Linda program execution in debug mode Including this option on the command line starts an xterm process running the debugger specified by the debugger resource on each participating node The value for the debugger resource defaults to dbx the other currently supported debugger is gdb the debugger from the Free Software Foundation Setting the value for debugger to none effectively disables debugging on a particular node as in this example Tsnet moliere debugger none Tsnet Node debugger gdb The second configuration file command sets the default value for the debugger resource to gdb while the first line prevents debugging on node 70 ere The debugger resource is node specific For example the following command will create three debugging windows executing the program test24 on nodes selected from the ntsnet node list in the usual way ntsnet debug n 2 test24 The node and application name will appear in the title bar of each window Once all of the debugger processes have started up you can set breakpoints run in single step mode examine variables and perform all other normal debugging functions for each process ntsnet facilitates program initiation by defining
52. is chosen arbitrarily When a match is found the actuals in the matching tuples are assigned to the formals in the corresponding fields of s Look for a tuple matching s in tuple space If a matching tuple is found actual to formal assignment occuts If no matching tuple exists the process blocks until one becomes available Predicate forms of rd and in respectively They do not block if no matching tuple exists but return 0 FALSE and exit If a match is found they return 1 TRUE and perform actual to formal assignment Each field of s containing a simple function call results in the creation of a new process to evaluate that field All other fields are evaluated synchronously prior to process creation When all field values have become available the tuple s is placed into tuple space Synchronously evaluates the fields of the tuple s and then places it into tuple space You can use the prefix __linda_ to construct an alternate name for any operation if the shorter name conflicts with other symbols The prefix begins with two underscore characters Formal C Linda Syntax linda call call type call body call type call body in linda in inp linda inp ra __ linda rd rdp linda rdp out Vinda out eval linda eval element element Linda User Guide 105 element formal actual formal lvalue length type name actual fvalue length length expression type name floa
53. iz 1 0 NXTILE NYPROB KOR k k kok k oko k oko ko k kok k RARA RARA k ok k kok RRA k kok RRA RRA RR K kok k kok ck kck ck kk ck kkk k Ste ic He E TT void step int it int ixmin int ixmax float ul NXTILE 2 NYPROB float u2 NXTILE 2 NYPROB void update Put boundary data into Tuplespace kg if ixmin 1 1 out west it ixmin sul 1 0 NYPROB if ixmax NXPROB 2 out east it ixmax amp ul NXTILE 0 NYPROB re Get boundary data from Tuplespace ui if ixmin f 1 1 int Bast up uxmim l Pearl O O eh if ixmax NXPROB 2 int west it ixmaxtl amp ul NXTILE 1 0 Update solution S update NXTILE 2 NYPROB ul u2 Linda User Guide 155 KOR KKK KK kok k ok ko k kok k RARA RARA RR KR kkk kkk k kkk kkk kk kkk k kkk k kkk k kkk k k kk k update KOKCKCKCKCKCKCKCkCKCkCk kCkCkCkCkCK Ck kCk Ck kCk Ck KCkCk kCkCk Ck kc k Ck kCkCk kok K kc k k kc k k kc kckckck ck k kk ckok kc kok c ke e xe kx x void update int nx int ny float ul float u2 int ix iy for ix 1 ix lt nx 2 ix 1 for iy 1 iy lt ny 2 iytt uz41x nyt1y ul ix nytiy parma cx Kiult ixtl j ny ftiy ult ix l nytiy 2 0 ul ix nytiy permscv oleistnytiytl uitix nytiy L 2 0 Klul ix ny y iy js KOR KKK KK kok
54. kok ko k KK IK I A I A K kok kkk kkk kk kkk kkk k inidat eK A kCkCk kCkCK Ck kCk Ck kCk k KCkCk kCkCk Ck kCK Ck K KCkCk kCk Ck Ck kc k k kc k k kc k ck kck ck kok ck I I I KH void inidat int nx int ny float ul int ix iy for ix 0 ix lt nx l ix 1 for iy 0 iy lt ny 1 iy kiul ix nytkiy El t ix nx xx Ij iy n iy 1 g KOK KK KK KK kok ko KK ok kok ok kok RRA pridat KOKCKCKCKCKCKCKCk KCkCk kCk Ck kCkCK Ck kCK Ck kCk Ck kCKCk kok k kCkCK Ck kCKCk kCkCk KCkCk k kc k k kc k k kc k kckck ck kck ck k kk ckck kc kok sk ke ke e kx x void prtdat int nx int ny float ul char fnam int ix iyi FILE fp fp fopen fnam w for iy ny 1 iy gt 0 iy for ix 0 ix lt nx 1 ix Foriner fo SO 2f lultix ny b ury s if ix nx 1 fprintt fp ys else fnrintfifp Xn fclose fp 156 Sample Programs 2D Heat Equation HEHEHE HEH HEH EEE EEE EEE EH EE HEE HEE HEE HEE HEE HEE HE HE HE HE HEE HR FILE make clinda_heat2D cl DESCRIPTION see clinda_heat2D cl USE make f make clinda_heat2D cl Note To compile using tuple scope us linda tuple_scope compiler option CFLAGS linda tuple scope FE HE TE FE FE k EH EH EH EH E FE FE FE HE FE FE HE TE FE HE EE EE EE HEH EH EH FE FE HH EE E TE FE E TE FE EE EE EE EE EE H
55. o mm Parallel Version KKK kok k oko k oko k kok k oko k RAR RR KARA k oko k kkk kok ok k oko kok RRA RRA RRA kkk kkk kk kkk kkk kk ck k kk LINDA Matrix Multiply C Version PILE elind mm cl OTHER FILES make clinda mm cl DESCRIPTION In this code the master task distributes a matrix multiply operation to n worker tasks NOTE C and Fortran versions of this code differ because of the way arrays are stored passed C arrays are row major order but Fortran arrays are column major order KOKCKCKCKCKCKCkCK Ck kCk Ck kCk k kok k Ck kCk Ck kc k Koko k kCk CK Ck kCK Ck kCKCk KCk k KCk Ck kok k kok Ck kk kck ck kck ck kck ck k kk kk include lt stdio h gt define NRA 62 define NCA 15 define NCB 7 define MAXWORKERS 20 real main int id LME numworkers nbr_rows offset i j k iT extra int worker worker function int work count number of work assignments made double a NRA NCA b NCA NCB c NRA NCB results NRA NCB ACkCk ck ck ck ck ck ck ck ck k ck kk k k k kk k k k i itializations KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK Get the number of tasks from the user KCKCKCKCKCKCkCkCK Ck kCk ck k kc k k kc K Ck kCk k kCK Ck KCkCk KCk Ck kCkCK Ck kCK Ck kCkCk KCk k KCk k kok Ck k kc k K kok kok kok ck ckck ck kok ck ke ke e Y numworkers 0 while numworkers gt MAXWORKERS numworkers lt 1 printf Enter number
56. of data types Tuples can have a maximum of 16 fields In C Linda tuple fields can be of any of the following types int long short and char optionally preceded by unsigned float and double struct union Arrays of the above types of arbitrary dimension including multidimensional 9 9 e arrays Pointers must always be dereferenced in tuples In Fortran Linda tuple fields can be of these types Integer 1 through 8 Real Double Precision Logical 1 through 8 Character Complex Complex 16 Synonyms for these standards types for example Real 8 Arrays of these types of arbitrary dimension including multidimensional arrays and or portions thereof Named common blocks 26 Using the Linda Operations Specifying Tuples and Basic Tuple Matching Rules Formal Tuple Matching Rules A tuple and a template match when They contain the same number of fields All corresponding fields are of the same type e The type of a field containing an expression is whatever type the expression resolves to The type of a field containing a formal 1s the type of the variable used in the formal In C fora structure or union field the type is extended to include the structure or union tag name The tag name and size of structures must match Tagless unions and structures are not allowed In Fortran common blocks match based upon their name alone Their internal structure is 70 considered
57. on the ntsnet command line it takes precedence over the resoutce value in the configuration files available Specifies whether a node is available for use as a worker This resource is node specific The default is true 116 Linda Usage and Syntax Summary bcast bcastcache cleanup debug debugger delay distribute fallbackload getload high ntsnet Configuration File Format Specifies whether or not the tuple broadcast optimization is enabled The default is false Specifies the size of the broadcast cache This size is a trade off between memory consumption and hit rate The default size is 1Mb This resource is only used when beast is true Specifies whether or not remote executables shall be removed from remote nodes after execution completes Executables are removed only if they were distributed by ntsnet in the current execution The local executable is protected from removal The default is true Specifies whether or not to run in debug mode see Debugging TCP Linda Programs on page 69 for more information The default is false If true it also changes the default value for kaon to false workerwait to 1000000 and maxwait to 1000000 and overrides the value of nice to be false Specifies the debugger to use when running in debug mode The default is dbx Specifies the delay period in seconds between invocations of rsh when ntsnet initiates execution on remote nodes The default value is 0 Speci
58. rd respectively that s what the p stands for They attempt to match the template specified as their argument to a tuple in tuple space in the same way as in and rd and they perform the same actual to formal assignment when a matching tuple is available As expected inp removes the matching tuple from tuple space while rdp does not When either of them successfully matches a tuple it returns a value of 1 in C and TRUE in Fortran If no matching tuple is available inp and rdp will not block Rather they will return a value of 0 C or FALSE Fortran and exit Using inp and rdp can complicate your program because they tend to introduce timing dependencies and non deterministic behavior that may not have been intended For example consider the following C Linda code fragments Master code real main for i 0 i lt tasks i O 0bL task 1 out tasks outed Worker code worker rd tasks outed while inp t sk 1 do task i Clearly if the rd of the tasks outed tuple were omitted the the worker code would be non deterministic It might get any number of tasks before the loop terminated which is not the intent What is perhaps less clear is that the program is still non deterministic even with the rd This is due to the fact that out is asynchronous There is no guarantee that all of the task tuples will be in tuplespace before the tasks outed tuple arrives It is
59. shall proceed otherwise execution shall terminate nodefile filename This option specifies the name of the file containing a list of nodes on which this application can run The default is tsnet nodes This resource is for backward compatibility with the old tsnet utility This file is only used if the nodelist resource is set to nodefile which is the default value See the description of the nodelist resource for more details nodelist node specifiers This option specifies a space separated list of nodes on which an application may run This list must be inclosed in quotes if more than one node specifier is used A node specifier can be any one or a combination of the types described below The keyword nodefile Linda User Guide 113 A node name A user defined resource See the description of the nodelist resource for more details Note If the nodelist option is not used and you have not specifically set the nodelist resource in the ntsnet configuration file s the application will run on the nodes contained in the tsnet nodes file in your current working directory opt resource value This option specifies a value to override any resource in the configuration file It provides a mechanism for overriding resources for which no specific command line option is provided p path This option specifies both the directory on a remote node where the Linda executable resides or will be distributed to and the directory th
60. some part of the output matrix C At the beginning of the program the master process creates tasks for each chunk of C that is to be computed separately Each worker removes a task from the shared data space It then reads the required rows of A and columns of B if necessary and forms the dot products When it is finished it places the resulting chunk of C into the shared data space and at the conclusion of the program the master gathers up all the chunks of C The granularity of the calculation can be adjusted by varying the amount of C that each task computes This approach to the problem is illustrated in the preceding diagram Notice once again the distinction between tasks and workers In this example the elements of C to be computed are divided into groups and each task consists of computing one of the groups The elements of C are not divided among the worker Linda User Guide 9 processes in any explicit way Worker processes do not know what tasks they will perform when they are created Workers acquire tasks as they are ready and perform whatever task they get This approach has many benefits For one thing it is generally easy to code since the worker processes don t need to worry about explicit interprocess communication that is taken care of by the parallel programming environment which manages the shared data space Processes read and write data via Linda s operations In addition this method also tends to be natura
61. specific instance nt snet Application names are usually the name of the corresponding executable program In order to make the separations between components possible however periods in application names must be translated to underscores Thus the application big job would appear in configuration file entries as big_job Application names can also be user defined if desired and the application name to use for a given run can be specified on the command line with ntsnet s appl option For example ntsnet appl big job medium job This option can be used either to apply one application s settings to a different application program or to specify the use of a user defined application name which need not correspond to the name of any executable program Note that periods in executable names are translated to underscores only when used as application names in the configuration files such translation should not take place at any other time such as when they are invoked in the ntsnet command line Node names can be full node names such as moliere frachem com ot node nicknames moliere In the node component of configuration file entries only periods again have to be translated to underscores so that ntsnet can figure out where the component boundaries are Anywhere else in the configuration file as part of resource values for example and on the command line no such translation is used Configuration file resources ate keywor
62. subsection of a multidimensional array you can do so by specifying a starting element and or length but a more powerful mechanism for doing so is provided by the Fortran 90 array syntax described in the next section The remainder of this section will discuss multidimensional arrays in C In C the basic principle to keep in mind is that multidimensional arrays match only when their types and shapes are the same the same holds true for sections of multidimensional arrays We will use these arrays in the examples int a 100 6 2 b 60 4 5 ar100 6 2 The following operations create and retrieve a tuple containing a multidimensional array Sut multi i in multa di The following in operation will not succeed because the shapes of arrays a and b ate different even though they have the same number of elements cut multi section ax in multi section b WILL NOT WORK Portions of multidimensional arrays may also be specified Here are some examples int a 3 5 2 b 51 2 cl2 i out section alol 01 35 int section 26i out 2d section a 0 2 in 2d section Tbi s out not an array a 0 0 D in not an array i just a scalar sss In the first pair of operations the construct a 0 0 points to the start of an array of length 2 which is why the formal involving array c matches it In the second pair of operations two 5x2 array sections with the second one
63. such statements still will retrieve whatever number of elements is present in the ten elements tuple Assuming that the first ten elements of array b have the same values as the corresponding elements of array a the following in operations would consume one of the ten element tuples without assigning any values in ten elements b l0 in ten elements b 10 Here array b 1s used as an actual and a matching tuple is simply removed from the tuple space assuming one exists Note that there are better ways to remove tuples from the tuple space without copying the data in them as we will discuss later Artay references can begin with any desired element as in these examples in ten elements amp b 4 1en in ten elements b 5 len These operations would retrieve a ten elements tuple place the ten array elements it held into the fifth through fourteenth elements of array b and set the value of 1en to 10 Although the examples in this section have used integer arrays exactly the same principles and syntax apply when accessing arrays with elements of other types Note that retrieving a tuple containing an array longer than the target array used in the template will result in writing past the end of the target array Linda User Guide 31 Multidimensional Arrays In Fortran array shape is ignored and so multidimensional arrays can be handled in a similar way to one dimensional arrays If you want to refer to a
64. tets tpoints lt 10 Y tpoints i 10 Print something out for validation printf first Sd pointe for validation Wo tpts for 1 0 i lt tpts i printt 4 2Ff values 1i printi anys Display results with draw wave routine draw wave amp values return 0 Linda User Guide 149 HEHE HH EH EH EH FE HE TE FE FE FE FE FE HE FE FE HE TE FE EH EE EE HEH EH HH TE FE FE HE FE HE EE EE EE EE EE HEH EH EE EH HE E E E HER H Linda Concurrent Wave Equation Program Makefile FILE make clinda wave cl DESCRIPTION see clinda wave cl USE make f make clinda wave cl Note To compile using tuple scope us linda tuple scope compiler option CFLAGS lm linda tuple scope k k FE E ERE FE HE FE FE HE TE FE E TE FE HE TE FE FE HE FE FE E TE FE HE TE FE FE HE FE FE EE EE FE HE TE FE EE EE EE EE EE HE TE FE EH EE E E E E CC ele CFLAGS lm wave clinda_wave cl S CC S CFLAGS clinda_wave cl o wave 2D Heat Equation in Serial Version e kk k kok k oko k ok ko RARA RRA kkk kkk k kkk RRA k kok k kok RRA RRA RRA k kok k kok k Koko k kk k k Ser FIL OTH DES two tem zer the dom FEF clu ial HEAT2D Example C Version E heat2D c ER FILES CRIPTIONS This example is based on a simplif
65. that remote TCP Linda processes ate initiated with rsh not rlogin Hence make sure that the PATH environment variable is set properly even if the login initialization file is not executed You can test this by running rsh manually and you can ensure this by placing the variable s definition in cshrc rather than login if you use the C shell In some network configurations it can be necessary to give remote hosts access to the local X server This is the purpose of the xhost command You will need to run xhost if you see error messages like this one Xlib Client is not authorized to connect to Server If it is required you can execute the xhost command manually For example the form xhost grants access to the local X server to all remote hosts You can also specify a list of nodes check the man page for details Alternatively you can modify the linda rsh shell script located in the bin subdirectory of the Linda tree adding the option access to the xon command which causes the latter script to run xhost automatically ntsnet s debug mode also changes the default values of some other resources The workerwait resource defaults to 1000000 seconds The maxwait resource defaults to 1000000 seconds The nice resource is overridden to be false ntsnet ensures a consistent dbx environment across all nodes by copying the dbxinit file from the local node to all participating remote nodes The dbx command ig
66. the master could easily generate tasks much faster than the workers could complete them and fill up tuple space in the process causing the program to run out of memory and terminate A technique known as watermarking can provide protection against this eventuality Watermarking involves making sure that there are no more than a fixed number of task tuples at any given time the high water mark so to speak Once this limit is reached the master process must do something else such as gathering results until the number reaches a lower bound the low water mark at which time it can go back to creating tasks When the number of tasks once again reaches the upper bound the process repeats Here is a version of the master process with watermarking 20 Call Get_Next DB Rec IF Rec EQ EOF Go TO 30 out task rec OK NTasks NTasks 1 IF NTasks LE UPPER BOUND Goto 20 DO While NTasks GT LOWER BOUND in result res Call Process res NTasks NTasks 1 EndDo Goto 20 30 Do 40 I 1 NTasks in result res Call Process res 40 Continue Database Searching Creating a new task increments the ntasks counter Once it reaches its maximum value the master switches over to gathering and processing results decrementing the ntasks counter which now holds the number of outstanding tasks since every time the master finds a result tuple it can be sure a task has been consumed When the number
67. the alias Irun within each debugging session to be the appropriate command line to begin application execution You should use this alias rather than the debugger s normal program initiation command e g run in dbx Once the program has finished executing the controlling master process will exit and the debugger prompt will appear in the corresponding window However the other worker processes will not return To terminate all program processes enter the quit command to the debugger for the master process and ntsnet will automatically terminate all of the other processes The debug resource application specific can be used instead of the command line option A value of true is equivalent to including the command line option the default value is false Linda User Guide 69 Hints on running TCP Linda programs in Debug Mode 70 Using TCP Linda Keep in mind that each process is handling a portion of tuple space in addition to running the application program Therefore when a process is paused for example at a breakpoint then no tuple space requests can be handled by it For this reason it s best to break only a single process at a time with all other processes either continuing or stepping through a blocked in or rd operation ntsnet relies on the command search path being set appropriately on all remote nodes Specifically the locations of xterm the debugger sh and rcp scp need to be in the search path Note
68. the rate limiting steps in the entire job For example if the comparison which took the longest was started last the other workers would all finish and sit idle waiting for it Sometimes such problems can be avoided by paying attention to the order in which records are obtained for example by making get_next_rec more sensitive to task size in our example This can complicate get next rec a great deal to the point where it too would benefit from becoming a parallel operation Of course the same kinds of considerations hold for the routine process as well At other times it is the comparison itself that needs to be parallelized It may not be sufficient to perform several comparisons at once when an individual comparison takes a very long time In either of these cases it will not be possible to take the generic approach to database searching that we have here Rather the specifics of the comparison or record retrieval or results processing algorithms will have to be examined explicitly and creating a parallel version of one or more of them will be necessary to achieve good performance In Chapter 7 of their book Carriero and Gelernter present an elegant combination database searching program that parcels out small comparisons as a whole and divides large ones into discrete pieces Linda User Guide 85 Molecular Dynamics 86 Case Studies Molecular dynamics calculations are performed to simulate the motion within molecules over time
69. the slider is at the far right end of the bar which represents normal full speed Moving the slider to the left causes program execution to proceed at a slower rate You can use the slider at any time during program execution to change the execution speed Slower execution rates represent a middle ground between single step execution mode in which execution breaks at every Linda operation and normal full speed execution 96 Using Tuplescope The Tuplescope Display A tow of buttons for the tuple class window at the bottom left of the control panel Click on a button to open its cotresponding tuple class window Buttons for currently open tuple class windows are grayed out When a process is initiated with an eval an icon for it briefly appears above the tuple class window icons This icon is a white outlined black box containing a white number within it in the diagram the sample icon contains the number 2 This number functions as a sort of internal process ID and is incremented every time a new process is created it does not correspond to any user assigned numbering scheme Once the process performs an operation on tuple space this icon disappears and the appropriate icon appears in one of the tuple class windows Tuple Class Windows A tuple class window has the following parts A sizing box black with two white veins located in the upper left corner of the window controls the size of the window Click the box to cho
70. usually abort the process and cause the Linda program to fail It is recommended that you quit from all native debugger processes before exiting from Tuplescope Pressing the Tuplescope Quit button while debugging windows are still open causes their processes to be terminated out from under them Tuplescope will make no attempt to stop the debugger processes so you will have to do it manually Some debuggers have difficulty shutting down in this state so you may have to use the UNIX kill command to stop those processes The Tuplescope Debugging Language Clicking on Tuplescope s Debug button brings up a menu that can be used to create compile and control debugging programs written in the Tuplescope Debugging Language TDL TDL is the means Tuplescope provides for users to specify that certain actions be taken on various program conditions The various items on this menu have the following meanings ltem Effect Edit program debug Edit the TDL program for this application Compile program debug Compile the TDL program for this application Clear Debugging Actions Cancel all debugging actions in effect Exit Debug Menu Close the Debug menu The Edit and Compile items edit and compile the file program debug where program is the name of the application running under Tuplescope Edit will open an editor in a separate X window using the contents of the EDITOR environment variable to determine which editor to run The Compile item
71. will block until the counter tuple s second field has its final value the number of worker processes A third approach involves retrieving the final data tuples created after the eval ed processes exit for example C Version Fortran Version in worker retval in worker iretval This allows the program to examine the return code from the function started by the eval operation While it isn t really necessary for a function as simple as hello it is a technique that is quite useful in mote complex programs Compiling and Running the Program To run this program you must first compile and link it The C Linda compiler has similar syntax to standard C compilers Its name is clc and its source files must have the extension cl Here is a clc command that would compile the program in the file hello world cl Linda User Guide 19 C Linda Compilation clc o hello world hello world cl The o option has the same meaning as with other compilers so this command would complle and link the source file hello world cl creating the executable program hello world The Fortran Linda version of the program is created using the fle command as shown below Fortran Linda Compilation flc o hello world hello world fl do args f Note that the extension on the Fortran Linda source file is 1 As illustrated additional source files Linda and non Linda alike may also be included on a Linda compilation command line when approptia
72. 0 Do 40 I 1 NTasks in result res Call Process res 40 Continue DO 50 I 1 NWORKERS out task dummy DII E 50 Continue Return End This program first starts the workers gets the target and places it into tuple space Then it loops retrieving one record from the database and creating a corresponding task tuple until there are no mote records Then it retrieves the results generated by the workers from tuple space and hands them to process Finally in its final loop the master process generates one additional task tuple for each worker These tasks serve as poison pills special tasks telling the workers to die The task tuple s third field holds either the value represented by OK meaning this is a real task or the one represented by DIE meaning terminate Linda User Guide 83 84 Case Studies Here is the corresponding worker Subroutine worker rd target target DO While TRUE in task rec flag If flag EQ DIE Goto 100 Compare rec target result out result result EndDo 100 Continue E The worker loops continuously reading tasks and compating records placing the results into tuple space for the master to gather later until it encounters the poison pill at which point it exits Straightforward as this version is it has some potential pitfalls The most serious occurs if the database has large records or large numbers of records or both In either case
73. 9 database searching 82 Linda User Guide i declarations in 2 dividing work among workers 74 Freewake 38 hello wotld 17 matrix multiplication 77 molecular dynamics 86 ntsnet configuration file entries 47 poison pill 83 ray tracing 73 Rayshade 73 simplification of 73 watermarking 84 executable locations 51 execution group 62 119 execution priority 51 extensions cl 19 C Linda soutce files 19 extensions of C Linda source files 19 F fallbackload resource 63 65 117 field types in tuples 26 file permissions 61 files gmon out 14 mon out 14 tsnet config application name 45 fixed aggregates 36 flexit 106 flhalt 106 flintoff 106 flinton 106 floffexit 107 flonexit 107 flprocs 107 formal 12 anonymous 36 Fortran 38 106 107 array numbering 41 ii Index array otder vs C 41 calling from C 41 Fortran 90 array syntax 33 Fortran common blocks 33 Fortran support functions 106 107 functions eval restrictions on 24 flexit 106 flhalt 106 flintoff 106 flinton 106 floffexit 107 flonexit 107 flprocs 107 lexit 106 Ihalt 106 lintoff 106 linton 106 loffexit 107 lonexit 107 Iprocs 107 print times 106 start timer 106 support 106 timet split 106 timing 106 G getload resource 63 65 117 gprof command 14 granularity 5 adjustability 6 9 81 definition 6 networks and 68 granularity knob 77 H hello_world program 17 high resource 51 117 119 HP UX 100 hpterm 100 I ignoring fields in tuples
74. Configuration File Format The maximum amount of time to wait for a valid execution group to be formed Note that maxwait specifies the total time to wait including the time specified in minwait it does not represent an amount of time to wait over and above the minwait interval The default is 30 seconds which is the same as the default for minwait unless debug is true when the default value is 1000000 seconds See the discussion of minwait below for more details about this resource Specifies the maximum number of workers started for a given application The default is the number of distinct nodes in nodelist minus one for the local node running the master ntsnet initially starts up the number of workers equal to the maximum of the minworkers and maxworkers resource values The master then waits the time period specified in the minwait and maxwait resources for the workers to join the execution group If at least minworkers join within that time execution shall proceed otherwise execution shall terminate See the discussion of minwait below for full details Specifies the minimum amount of time to wait to allow an execution group to form in seconds the default is 30 Execution will proceed according to the following criteria First if at any point before the minwait interval has elapsed maxworkers workers have joined the execution group execution will commence at once When the minwait interval expires if at least minworkers workers ha
75. Determining working directories and executable file locations on remote nodes using map translation and the associated configuration files if necessary Copying executable files to remote nodes if necessary Initiating remote processes Waiting for normal or abnormal termination conditions during program execution Shutting down all remote processes at program termination Removing executables from remote systems if appropriate The remainder of this chapter will look at these activities and the ways that the user can affect how ntsnet performs them in considerable detail Using the ntsnet Command Using the ntsnet Command The general syntax for the ntsnet command is ntsnet options executable arguments where options are ntsnet s options executable is the executable file to run on the local system and arguments are command line arguments for the specified network program ntsnet uses the command line options the location of the local executable and the settings in its configuration files to determine all of the remaining information it needs to execute the network parallel program Customizing Network Execution TCP Linda provides the ntsnet command to execute parallel programs across networks of machines ntsnet is designed for maximum flexibility Proper configuration makes running a TCP Linda program as simple as prepending its executable s pathname and any arguments with the ntsnet command as in the previous
76. E HE HE HE E E E E CC elg OBJ heat2D SRC clinda heat2D cl XLIBS CFLAGS B OBJ S SRE SICC S CFLAGS SRC S INCLUDE S LIBS S XLIBS o 0BJ Linda User Guide 157 158 Sample Programs Bibliography Nicholas Carriero and David Gelernter How to Write Parallel Programs A First Course The MIT Press Cambridge MA 1990 An introduction to developing parallel algorithms and programs The book uses Linda as the parallel programming environment David Gelernter and David Kaminsky Supercomputing out of Recycled Garbage Preliminary Experience with Piranha Proceedings of the ACM International Conference on Supercomputing July 19 23 1992 An overview of and preliminary performance results for Piranha Brian W Kernighan and Dennis M Ritchie The C Programming Language Second Edition Prentice Hall Englewood Cliffs NJ 1988 The canonical book on C Robert Bjornson Craig Kolb and Andrew Sherman Ray Tracing with Network Linda SIAM News 24 1 January 1991 Discusses the Rayshade program used as an example in Chapter 3 of this manual Harry Dolan Robert Bjornson and Leigh Cagan A Parallel CFD Study on UNIX Networks and Multiprocessors SCIENTIFIC Technical Report N1 Discusses the Freewake program used as an example in Chapter 2 of this manual Mark A Shifman Andreas Windemuth Klaus Schulten and Perry L Miller Molecular Dyna
77. EDITOR environment variable is opened in a new window Compile program debug Translate the TDL program to Tuplescope s internal form and put its statements into effect Clear Debugging Actions Cancel all debugging directives in effect via the current TDL program Exit Debug Menu Close the Debug menu TDL Language Syntax TDL statements have the following form if condition then action Conditions have one of the following formats select one item from each column This format tests the value in a tuple field and performs the action for matching tuples field N constant value I A Fs This format tests for the specified Linda operation and performs the action for matching processes 124 Linda Usage and Syntax Summary TDL Language Syntax linda op eval l out in ra block_in block rd This format tests for the specified process number and performs the action for matching processes process N A Vo Note that the brackets are part of the condition syntax and must be included Multiple conditions can be joined with and and or The entire condition is enclosed in parentheses when it is placed into the TDL statement Actions must be one of the following break Pause program execution hide Hide the triggering processes tuples color c Change the triggering process tuple to the color c one of red orange yellow green blue indigo and violet save Save the current contents of tu
78. In this chapter we will look at them in more detail and in the context of complete if simple parallel programs Note that this chapter contains examples of using both C based and Fortran based Linda and it refers to them as C Linda and Fortran Linda respectively Quick Start Hello world Fortran Fortran In this section we ll construct a parallel version of the canonical first example program hello world Here is the sequential version main printrf Hello worldXn Program Hello World Print Hello world End It would be absurd to try to perform the computation done by this program in parallel but we can create a parallel version where each worker process executes this program and says Hello world at the same time Here is a program that does so real main int argc char argv int nworker j hello nworker atoi argv 1 for j 0 j lt nworker j val worker hello j for j 0 j lt nworker j in done printf hello world is finished n return 0 Subroutine real_main integer I NProe Linda User Guide 17 Obtain number of workers amp store in NProc Do 10 I 1 NProc eval worker hello I 10 Continue Do 11 I 1 NProc in done 11 Continue Print hello world is finished Return End The first thing to notice about this program is its name and top level structure real main for the C Linda which requires that the top level routine
79. KK KK KK KK oko k KK KK worker task OKCKCKCKCkCkCk ck k I I ko ke X RARAS int worker int offset rows length int next offset int i jg kj double a NRA NCA b NCA NCB c NRA NCB First read in the a and b arrays Ed array x Pals rd array b bj This is the head of an infinite loop over tasks while 1 See if there is any work out there to do in offset offset if offset 1 out offset ljs return 0 There is still data to process Receive matrix data from master task in array data offset rows Update offset for the next guy if offset rows lt NRA next_offset offset rows 140 Sample Programs Matrix Multiply else next offset 1 out offset next offset Do matrix multiply for i offset i lt rows offset i for 3 0 j NCB j st t 20 for k 0 k lt NCA k el21 141 e 11191 ela ik biel ta Send results back to master task out results offset rows Eli end while loop It s all over return 0 EE AE aE aE aT aE AE k aE aE FE aE FE FE AE AE FE aE HE aE AE AE AE EH aE EH AE aa aH aa Ha aa aE aH aaa AE AE HEE aaa Linda Matrix Multiply Makefile FILE make clinda_mm cl DESCRIPTION see clinda_mm cl USE make f make clinda mm cl
80. KKKKKKK Get the number of tasks from the user Then define the number of worker tasks and the array partition size as chunksize KKEKK KKK KKK KKK KKK KKK A A X KKK KKK KKK KKK KKK Z A AX A KKK KK KKK KKK KKK KKK KKK KKK KKK k Edid numworkers 0 while numworkers gt MAXWORKERS numworkers lt 1 printf Enter number of workers between 1 and d n MAXWORKERS scanf d amp numworkers chunksize 100 KKK KK KR RIA RARA k kok k RAR kkk k master task OKCKCKCKCk kCk k k OR ck ko ke ke ke e x x f printf Starting LINDA ARRAY Example X X IN Initialize the array i 0 while i lt ARRAYSIZE data i 0 i put it out there out init arrav data Start the worker tasks i 1 while i lt numworkers printf Starting worker task d n i eval worker worker i i4 nbr chunks ARRAYSIZE chunksize i 1 index 0 while i lt nbr chunks out my section index chunksize index index chunksize dues end while Linda User Guide 129 print a few sample values i 1 while in result data i lt nbr chunks 7id index result put it into the data array data index index while data index lt index chunksize data data_index result data index data index print t n printf MASTER Sample results from worker tas
81. ail about how that happens 22 Using the Linda Operations Linda Operations Logically the fields of an eval are evaluated concurrently by separate processes evaling a five field tuple implicitly creates five new processes When every field has been evaluated then the resulting data tuple is placed into tuple space In current Linda implementations however only expressions consisting of a single function call are evaluated within the live tuple and actually result in a new process These functions can use only simple data types as their arguments and return values see below All other fields are evaluated sequentially before new processes ate created Here is a typical eval statement C Version Fortran Version evali coord x t x eval ocoord X Elx This eval will ultimately result in the same data tuple as the out operation we looked at previously Howevet in this case the eval operation will return immediately and a new process will evaluate x By contrast the out operation will not complete until the evaluation of x is complete and it has placed the data tuple into tuple space Compare these two C Linda loops Loop with out Loop with eval for 30 i lt 100 144 for 10 i lt 100 i out f values i 1 eval t values Ap fi The loop on the left will sequentially evaluate f i for the first 100 non negative integers placing a tuple into tuple space as each one completes The loop on
82. al version 7 0 Corresponds to TCP Linda version 7 0 September 2005 Printed in the U S A Acknowledgments Many Linda users and developers have contributed to the direction and quality of both the Linda products and this documentation In particular we d like to give special thanks to Harry Dolan Kevin Dowd and Joe Casper of United Technologies Research Center who produced the Linda version of the Freewake program described in Chapter 2 and T Alan Egolf the author of Freewake Craig Kolb the developer of the Rayshade program described in Chapter 3 Mark A Shifman Andreas Windemuth Klaus Schulten and Perry L Miller the developers of the Molecular Dynamics program described in Chapter 3 Paul Bercovitz who created the original version of the Tuplescope debugger described in Chapter 5 and Donald Berndt and Steven Ericsson Zenith authors of previous Linda manuals for SCIENTIFIC Contents Introduction esses j bd d oc AE 1 How to Use this Manual 0 005 006 eere e m9 rk nr eee eh ha Eh ire en 1 About the Example Programs sies e hr a er eR e Rees 2 Typographic Conventions ccc ce ee eee ee eens 3 1 Overview of Parallel Programming and the Linda Model 5 Approaches to Parallel Programming 4444444444440 7 Message Passitnous coe eee er eA He peque EP Pa e PEE a Distributed Data Structite s se shee re Gorin o Gewese REESE E EET 8 The Linda Model citrico ERU E Tr peor OG nee ak E
83. an Version 20 10 ell segmnt iseg calcdisp amp x iseg ifil iblade amp y iseg ifil iblade amp z iseg ifil iblade amp dy iseg amp dz iseg amp y amp z place results in tuple space amp dx iseg amp x out delta index dx dy dz Subroutine worker Double Precision x y z rd wake x y z nblades nfilmnt nsegmnt Do 10 I 1 VERY BIG NUM in index index out index index 1 GE nblades nfilmnt if index Return iblade index nfilmnt 1 ifil modulo index nfilmnt 1 Do 20 iseg 1 nsegmnt Gali caledisp x iblade ifil iseg z iiblade ifil iseg y iblade ifil iseg dx iblade ifil iseg dy iblade ifil iseg dz iblade ifil iseg Continue out delta index dx dy dz Continue Return End Each worker process first reads the position arrays and their index limits from tuple space The worker then loops continuously until all points are done At the beginning of each loop it removes the index tuple from tuple space increments the counter in its second field and then returns it to tuple space for possible use by other processes 40 Using the Linda Operations Example Freewake This counter stored in the variable index serves as a composite index combining the outer two loops of the original Fortran code Each task consists of executing the inner Fortr
84. an loop for a fixed pair of I and J values i e specific blade and filament indices A single counter tuple is easily retrieved incremented and returned to tuple space There are NFILMNT times NBLADES distinct pairs of I and J values so the worker first tests whether the counter s value is greater than or equal to this product equality is included in the test since index begins at 0 and runs to nfilmnt nblades 1 we ve given the equivalent variables in the Linda programs lowercase names The variable iblade is defined as index divided by nfilmnt and ifil is defined as index modulo nfilmnt Because these are integer operations and all fractions are truncated iblade will remain 0 until index reaches nfilmnt then become and remain 1 until index reaches nfilmnt 2 and so on At the same time ifil counts from 0 to nfilmnt 1 for each value of iblade over the same period The following table indicates the encoding of iblade and ifil within index for nfilmnt 3 and nblades 2 blade filament index iblade iseg 0 0 0 1 0 1 2 0 2 3 1 0 4 1 1 5 1 2 6 terminate Once iblade and ifil are computed a for loop calls a slightly modified version of the otiginal calcdisp routine once for each segment value using the computed iblade and ifil values each time In the C Linda version this loop differs from the Fortran inner loop it replaces in several ways First the arguments to calcdisp are explicitly the addresses of the relevant array el
85. an terminate normally when the last process usually real main finishes returns from its outermost call A program may terminate by having the final process call the C Linda support function lexit flexit is the Fortran Linda form but this is not required Note that if you want to call an exit function use lexit never use the standard system call exit The program can force termination by calling 1halt C or flhalt Fortran When such a call occurs in any process of the Linda program the entire computation will be terminated immediately and automatically Hany process terminates abnormally then the entire program will again be terminated automatically at once Linda User Guide 37 An individual process within the parallel program may terminate execution of user code by calling lexit or flexit This function terminates local user processing without affecting the other running processes if any See the Linda Usage and Syntax Summary for full details on available termination routines and their use Example Freewake We re now ready to look at another Linda program which illustrates all four operations and several tuple matching principles in action This program also illustrates the following techniques The master process becoming a worker after initialization is complete The use of a composite index to combine two existing indices The program we ll examine is named Freewake a computational fluid dynamics
86. ansform the serial program into a parallel program in the most straightforward way possible measure its performance and then search for ways to improve it If we focus on step 2 a natural question arises where does the parallelism come from The language of parallel programming can be quite ambiguous On the one hand there is talk of parallelizing programs a phrase which focuses on the programmers who convert sequential programs into parallel ones On the other hand much is said about exploiting parallelism implying that parallelism is already inherent in the program itself Which is correct Is it something you find or something you create The answer is of course it depends Sometimes a program can be trivially adapted to run in parallel because the work it does naturally divides into discrete chunks Sometimes a serial program must be restructured significantly in order to transform it into a parallel program reorganizing the computations the program does into units which can be run in parallel And sometimes an existing program will yield little in the way of independent computations In this case it is necessary to rethink the approach to the problem that the program addresses that is create new algorithms to formulate a solution which can be implemented as a parallel program t t of course even in these cases it may still be possible to take advantage of a parallel computer by running multiple concurrent sequential jobs
87. application It was developed to model the wake structure created by helicopter rotor blades and its influence on the blades themselves to aid in the design of new helicopters It was originally parallelized with C Linda but we will provide both C Linda and Fortran Linda versions here At its core the calculation is essentially a very complex N body problem the wake surface is divided into a large number of discrete elements The structure of this surface depends on the properties of the individual elements including their velocities For each time step the change in velocity of each element is a function of its interactions with all of the other elements and all of these changes in velocity determine the new structure of the wake surface Once it is obtained the program calculates the interaction of the wake and the blades themselves The vast majority of the computation is spent calculating the displacement of each point of the wake for each time step This is computed by these three nested loops See How and Where to Parallelize on page 14 for information about determining the most computationally intensive portions of a program Here is the original Fortran code DO I 1 NBLADES Typically 2 DO J 1 NFILMNT Typically 16 DO K 1 NSEGMNT Typically 512 CALL CALCBISP1 Silja Rie Vib Rly AIJ K DX I J K DY I J K DZ I J K END DO END DO END DO The subroutine CALCDISP ca
88. arrays in tuples simply by name i e without the colon notation For example if a is an array of declared length 20 then it can be referred to in a tuple by its name alone as in this example int a 20 b 20 out arrav a Similarly you can retrieve such an array by name 36 Using the Linda Operations Termination of Linda Programs rd aeray 25 in array Ta You can also treat multidimensional arrays and sections thereof this way int a 100 1 61 121 5 25 241 2 out milti section a 90 0 in multi section b 0 01 5 In these statements both arrays are essentially being used as pointers to the start of a fixed aggregate of length 2 Two array sections treated as fixed aggregates in this way must have the same length and shape to match Fixed aggregates never match arrays specified in the usual way Fixed aggregates match only other fixed aggregates Literal character strings specified without a terminal colon are treated as fixed aggregates Thus neither of the following in operations will find a match char a 20 8161 int len out wrong all cut also wrong hello in wrong a no match in also wrong stlen no match Termination of Linda Programs Since a Linda program can consists of many concurrent processes program termination requires a bit more care than for sequential programs Linda programs can terminate in one of three ways The program c
89. at shall be eded to prior to executing the remote Linda process Thus the p option simultaneously overrides both the rexecdir and rworkdir resources Since p specifies a local directory the value of p is subject to map translation The translation occurs before the p value overrides the rexecdir and rworkdir resources This option is intended to provided a mechanism very similar to the p option on the previous tsnet utility redirect This option turns on the tuple redirection optimization This is the default redirect This option turns off the tuple redirection optimization suffix This option causes a node specific suffix indicated by the suffixstring resource to be appended to the executable name This is the default Note that the default value of suffixstring is an empty string suffix This option indicates that TCP Linda is not to use node specific suffixes translate This option indicates that map translation shall be performed This is the default Note that it is not an error to have translation on and to have no map file In that case no translations will be performed translate This option indicates that map translation shall not be performed useglobalconfig This option causes ntsnet to use the resource definitions in the global configuration file The resource definitions in the global configuration file are used in addition to the command line options and the user s local configuration if one exists This i
90. at file and generating graph n prtdat NXPROB NYPROB u final dat KOR KK kok k oko ok ok ko k kok RRA kok k oko k oko kok RR RR RRA kkk RRA k kkk k kkk kkk k kkk kk ke kk Worker routine KOKCKCKCKCKCkCK Ck kCk Ck kCk Ck kok k kCkCKCk kCk Koko k Ck kCk Ck kok kok Ck kc kk k Ck kck ck kck ck k kk kk kk kk kk int worker int ixmin int ixmax float utile 2 NXTILE 2 NYPROB int ix iy iz ES void step ss Get parameters and initial data from Tuplespace af rd struct parms paris imi initral data asin Penatile 0 1 0 eps Ze Set edges of utile to grid boundary values if appropriate ui if ixmin 1 1 int Ieft T utile O0 O 0 lt for iy 0 iy lt NYPROB 1 iy utile 1 0 iy utile 0 0 iy if ixmax NXPROB 2 ind right 2 utile 0 NXTILE 1 0 2 for iy 0 iy lt NYPROB 1 iy utile 1 NXTILE 1 iy utile 0 NXTILE 1 iy El 154 Sample Programs 2D Heat Equation for ix 0 ix lt NXTILE41 ix 4 utile 1 ix 0 utile 0 ix 0 utile 1 ix NYPROB 1 utile 0 ix NYPROB 1 FE Iterate over all timesteps ui iz 0 for it 1 it lt parms nts ittt step it 1 ixmin ixmax amp utile iz 0 0 sutile 1 iz 0 0 az 1 Tus Put results into Tuplespace k out result id ixmin ixmax out result ixmin amp utile
91. be given this name rather than the usual main Similarly the top level Fortran Linda routine is named real main note also that it is defined as a Subroutine notas a Program The C Linda program requires one command line atgument the number of wotker processes to cteate and to save space we ve eliminated the code that checks whether or not it got a valid argument There are a number of ways that the Fortran Linda version could obtain the number of worker processes we won t dwell upon those possibilities at this point Next the program s first loop initiates nworker worker processes using the eval operation each executing the function hello Here is hello C hello int 1 printf Hello world from number d n i out done return 0 Fortran Subroutine Hello ID Print Hello world from number ID out done Return End hello is only a minor variation of our original sequential program It prints the Hello world message along with the number it was passed from real main this integer serves as a sort of internal process number Each message will look something like this Hello world from number 3 The routine hello places a tuple containing the string done into tuple space just before exiting These tuples are then gathered up by the master process real main in its second loop This technique has the effect of forcing real_main to wait until all the worker processes have terminated before exiti
92. causes the file to be translated into an internal form used by Tuplescope Successful compilation results in the message Compilation done Otherwise an error message is printed Once compiled the statements in the debugging program go into effect until cancelled by the Clear Debugging Actions item TDL Language Syntax TDL programs consist of one or more lines of the following form if condition then action where condition is a test condition and action is some action to be taken when the test is true 1 Note that the parentheses are part of the condition Conditions are formed from the following components item operator test_value Note that the brackets are a required part of the syntax There are three distinct kinds of conditions Linda User Guide 101 102 Using Tuplescope Tuple field comparison tests where e7 is field N where N is an integer operator is one of the C operators gt and lt and est_value is a constant This sort of test selects tuples on the basis of one or more of their fields For example This test chooses tuples whose second field contains the value 2 Character strings used as constants must be enclosed in double guotation marks Single characters must be enclosed in single quotation marks Note that tuples from distinct tuple classes can be selected by the same tuple field comparison Fields containing aggregates may not be used in such conditions Tuple space operation tests w
93. ce eee eee 63 Special Purpose Resources y siria seins tre e adi Wee av n vie ets 67 Tuple Redirection Optimization lt e 67 Disabling Global Configuration Files eee eee eee 67 Generating Additional Status Messages 0 6 cee eee eee eee 67 Process Initiation Delays 1 nee ence ee 67 Appropriate Granularity for Network Applications lt 68 Forcing an eval to a Specific Node or System Type lt lt 68 Debugging TCP Linda Progtams 5 eere rte eek Rn ee 69 ntsnet s Deb g Mode s esses eer ex Cy Er e e a 69 Hints on running TCP Linda programs in Debug Mode 70 Running TCP Linda Programs Without ntsnet 444444444 71 4 Case Studies ers er ere Osea AO keer see NR AR 73 Ray Tracing e zo boa ieee x ns GAY BU va Veta Gad Wed e es Au VR Bink We e 73 Matrix Multiplication icc eder EIE ry eg d NR Helo de nae eB 77 Database Searching desees Bes fy eR ORE REA each Re eee rede udis 82 Molecular Dynamics 2406 o mestre ene Ro UR e scene t UR D bts 86 5 Using Tuplescope x Mia A dd 95 Program Preparation 4 cess rk e Rm ere eR Aura darn d UR E Ds 95 Invoking TFuplescope eee e y e Re hme nm ehh heit kt alo da m eon 95 Th Tuplescope Display soi dio ere e eren eed ed eet dde a 96 The Control Paneler esee ee rm eer Cr Ee p CE PUN Caepio ds 96 Tuple Class WII dO WS senil oh vp wat C epa ee i e V P RES RS 97 ii Contents Niewino ADOTEDALES v
94. ched or come closest to matching so far When all relevant records have been tested output will print the final results This version could search any kind of database given appropriate versions of get_next_record compare and process get_next_record could be implemented to return every record in the database in sequential order or according to some sorting criteria or it could return only selected records those most likely to match the target for example think of searching a database containing fingerprints Database Searching compare might return a yes or no answer depending on whether target matched the current record or not or it might return some value indicating how close a match the two were In the first case process would only need to keep track of positive results matches while in the second it would probably want to report on some number of best matches at the conclusion of the program Transforming this program into a parallel version is fairly straightforward Each task will be one comparison This time we ll use one tuple for each task holding the actual database record rather than a single counter tuple Here is the parallel master routine Subroutine real_main Do 10 I 1 NWORKERS eval worker worker 10 Continue Call Get_Target target out target target NTasks 0 20 Call Get_Next DB Rec IF Rec EQ EOF Go TO 30 out tesk rec OR NTasks NTasks 1 Goto 20 3
95. cons for processes that have accessed tuples in this class The form of the icon varies depending on the operation that the process performed and its status These are the possible icons all icons display the process number in their center Icon Appearance Meaning Solid black arrow pointing up Successful in operation Solid black atrow pointing down Successful out operation White arrow in black box pointing up Successful rd operation Solid black diamond Blocked in operation White diamond Blocked rd operation There are examples of each type of icon in the Tuplescope display diagram Viewing Aggregates 98 Using Tuplescope The Display Aggregates item on the Modes menu controls the display of aggregates Click the Modes button to display its menu and click the Display Aggregates item to togele its current state If the item is selected a check mark appears to the left of its name You must select the Exit Modes Menu item to close the modes menu Tuplescope Run Modes When Display Aggregates is on the Aggregates menu button is active and you can use its menu to select the data format for subsequent aggregates displays It contains the choices Long Short Float Double Character and Hexadecimal Only one format choice is active at a given time a check mark appears to the left of its name All other formats are grayed out and unavailable The default format is Long To select a different format first deselect the current format by choo
96. consumption and hit rate The default size is 1 MB This resource is only used when beast is true This option indicates that remote executables should be removed when execution completes This is the default Note that the local executable is protected from removal This option indicates that remote executables should not be removed when execution completes Linda User Guide 111 d d debug distribute distribute Synonymous with distribute distribute Run application in debug mode see Debugging TCP Linda Programs on page 69 for more information This option also changes or overtides the values of several ntsnet resources see the discussion of the debug resoutce later in this chapter for details This option causes executables to be copied to remote nodes prior to execution Executables shall only be copied to nodes which are actually going to take part in the execution After execution completes ntsnet automatically removes the remote executables that it just distributed The local executable is protected from removal See the cleanup command line option or resource for information on preventing the automatic removal of remote executables This option indicates that executables are not copied This is the default fallbackload load getload getload h h help high high This option specifies the load average the scheduler shall use for a node if the RPC call to get system load average fails
97. cont command in dbx You may now use the native debugger to examine the process The figure below illustrates a sample combination debugging session ex Medes frenos Run Break Continue Debus Save it EBM globals task result worker Tuplescope windows exec 2 worker ng dbx version 3 1 for AIX ane mn Joe Type help for help sl alobals INT INT reading symbolic information m dbx 9 Native debugger window By default the debugger started is dbx except under HP UX where it runs xdb in an hpterm window A different debugger can be specified by setting the DEBUGGER environment variable to it If the executable is not in the current path give the entire pathname as its value otherwise its name alone is sufficient Currently supported debuggers are dbx gdb and xdb under HP UX It takes a little practice to understand all the nuances of Tuplescope native debugger interactions The trickiest part is usually figuring out where the next continuation command needs to be executed If the Tuplescope Continue button does not resume execution try issuing a next command to the native debugger process es 100 Using Tuplescope The Tuplescope Debugging Language To exit from a native debugging session without affecting Tuplescope detach from it before quitting in dbx give the detach command followed by quit Quitting without detaching first will
98. cutables 51 specifying resources without option equivalents 114 syntax 45 111 wait option 62 115 workerload option 65 115 workerwait option 115 o on command 118 operations 11 105 alternate names for 26 blocked 13 predicate forms 25 out 11 22 105 field evaluation order 22 ovethead 6 P parallel programming issues 7 steps 5 parallel programs interprocess communication options 8 parallel resource value 52 parallelism 5 Linda User Guide iii and hardware environment 6 fine vs coarse grained 6 parallelization level to focus on 81 passive tuple 11 permissions 61 poison pill 8 portability 1 POSTCPP_CC environment variable 123 predicate operation forms inp 25 rdp 25 print_times 106 process scheduling 61 processor definition 7 prof command 14 profiling 14 program examples See examples program termination 37 Q quick start 17 Network Linda 43 R ray tracing 73 rd 11 12 21 105 compared to in 21 rdp 25 105 real_main 18 37 creating from existing main 74 redirect resource 67 120 resource value 50 resources 46 ntsnet command line options and 48 50 parallel value 52 58 120 process scheduling 61 specificity types 46 49 specifying those without iv Index command line options 114 syntax 116 rexecdir parallel value 58 rexecdir resource 52 59 61 120 parallel value 52 rsh command 118 running C Linda executables 20 running C Linda programs on a network 21 rworkdir res
99. dd significantly to the time needed to process the tuple Many non shared memory Linda implementations provide a way for the user to request that the array not be copied The user does this by negating the length In this case a supporting Linda implementation simply records the address of the data This address is then used to locate the data that must be retrieved To satisfy a rd or an in Any changes made to the data after the out but prior to retrieval will be reflected in the data sent Since the user generally cannot know when the system internally retrieves the data the tuple s content in the presence of such modifications has to be viewed as nondeterministic While it 1s possible to design a parallel algorithm that exploits this non determinism using a negated length field is tantamount in most cases to a promise to oneself that the array will not be modified until by some means it is known that all rds and ins against the tuple have completed Refer to the release notes to see if a particular implementation supports this optimization For more on arrays Fixed Aggregates in C on page 36 Linda User Guide 29 Pointers and Assumed Size Arrays C pointers to arrays and Fortran assumed size arrays must a mays specify an explicit length when they are used as an actual within a tuple as in these out operations C char b 2Z0 py Al30 int len p b out array2 pr20 in arrayZ d len Fortran Dimension b d 30
100. de resource value where the various components have the following meanings Program is either the class name Tsnet or a specific instance of this class 1 e ntsnet In the future there may be alternate versions of Tsnet type programs such as xtsnet but currently there is only nt snet Appl is either the class name Appl or a specific application name such as ping The application instance names cannot contain a period you must convert periods to underscores Node is the class Node or a specific node name such as mysys The node instance names can be either the node s official name or a nickname The node instance names are node names found in either the etc hosts file the NIS hosts database or the Internet domain name database An example of an official name is fugi mycompany com A typical nickname for this node is fugi If a node name contains a period you must convert the period to an underscore The other option would ec be to use a nickname not containing the character Resource is a vatiable name recognized by ntsnet that can be assigned values Value is the value assigned to the resource If both the appl and node components are required for a given resource definition the appl component must precede node If an incorrect format is used the resource definition will be ignored by ntsnet Note All resources are application specific unless otherwise specified Also if the corresponding option is used
101. de specific The default is 1 suffix Specifies whether or not to append a node specific suffix indicated by the suffixstring resource to the executable name The default is true suffixstring Specifies a suffix to be appended to a particular executable name when run on a particular node the default is the null string meaning no suffix This resource is node and application specific threshold Specifies the maximum load allowed on a specific node The ntsnet scheduler is prevented from starting another worker on this specific node when this threshold is reached This resource is node specific The default is 20 translate Specifies whether map file translation is used The default is true useglobalconfig Specifies whether the global configuration file is used The default is true useglobalmap Specifies whether the global map translation file is used The default is true 120 Linda Usage and Syntax Summary Map Translation File Format user Specifies a username to use on remote nodes instead of your local username If this resource is unspecified the remote username is the same as your local username This resource is node specific verbose Specifies whether or not ntsnet works verbosely The default is false If the verbose resource is true ntsnet displays remote commands being issued and information about each node specified in the nodelist resource veryverbose Specifies whether the maximal amount of status messages s
102. ds whose values specify some kind of execution behavior option or application or node characteristic Consider these examples Tsnet Appl Node rworkdir tmp Tsnet Appl maxprocspernode 4 Tsnet Node speedfactor 1 These three entries all refer to application and node classes only thereby serving as default values for instances not specifically defined in other entries The first line sets the default working directory for remote nodes to tmp on each node The second line sets the maximum number of processes per node to 4 and the final line sets the default node speed factor to 1 These entries are completely general applying to every application program and or node Contrast them to the earlier example which applied only to the explicitly named applications and nodes Resource Types One can also freely intermix classes and specific instances as in these examples Tsnet hello world Node rworkdir tmp hello Tsnet Appl moliere rworkdir xtmp The first example sets the default working directory to tmp hello for any remote node when the program hello world is run The second example sets the default working directory on node moliere to xt mp whenever a network application without its own entry for this resource runs remotely on it Many resources have corresponding ntsnet command line options These options are designed to be used to override configuration file settings for a specific program run although they can be used inst
103. du Eie eas 10 How and Where to Parallelize c iced ee e asi em Ree e cete es 14 2 Using the Linda Operations 2 22cs ea hr Rn 17 Quick Start Hello world 0 eect nen 17 Compiling and Running the Program eee 19 C Linda Compilation escoria whe a dada hie des 20 Fortran Linda Compilation 0 0 ce eee eee ee 20 Linda Operation edge ale eee ep eg nee rho E aa 21 Dr toi A gue E o ade cdi alan salad oan R bad bb Sis 21 O A REO ee RET VS Als GAM Gad Reg bs 21 o PC CP 22 IU etd esce eM eie up eu P PU QR M UE eee paie edd meas ease 22 eval s Inherited Environment 4444444444400 23 eval Function Restrictions ass ceder ene nee bnew E nem e e oV VA A E SUE 24 Predicate Operation Forms inp and fdp eee 25 C Linda Alternate Operation Names 4 26 Specifying Tuples and Basic Tuple Matching Rules ooooooomommmmmo o 26 Formal Tuple Matching Rules 4 eee eee ees 27 No 0 tt u Pa A A tek ak HASTE nas R ede na 28 ATTAYS 45 bees BN eS Gee GaSe ed ban Eder eho eRe RAO Ta Meee 28 Avoiding Array COPIES isade reame erre rhe ds 29 Pointers and Assumed Size Arrays 0 eect etnies 30 Multidimensional Arrays 4444444000000 ence eens 32 Fortran 90 Array Sections 0 60 5 c ees cece ee ee ee eee e e ese 33 Fortran Named Common Blocks 0 0 0 cece eee eee 33 fcu PD 34 Varying Length Structures iiis sss ies re ee et e a eee 35 C Character
104. e workerdone tuples and merges their results into the arrays used by the remainder of the unaltered sequential program The calculation then continues as in the sequential version with the final calculated energies corrected for temperature at the end of each iteration Once all time steps have been competed real main kills the workers and cleans up tuple space finishing the master functions it started when the program began As we have seen master responsibilities are distributed among three separate routines in this program each controlling a different phase of the calculation 5 Using Tuplescope Tuplescope is an X based visualization and debugging tool for Linda parallel programs In addition to the usual debugger features such as single step mode and breakpoints Tuplescope can display tuple classes data in specific tuples and visual indications of process interaction throughout the course of a program run Linda programs are usually debugged using the Code Development System and Tuplescope This combination offers the most convenient and least intrusive method of debugging Tuplescope is part of the Linda Code Development System which simulates parallel program execution in a uniprocessor environment It requires a workstation running X Windows This chapter assumes knowledge of standard debuggers and debugging activities Refer to the manual pages for your system or to the relevant books in the Bibliography for information on
105. e Linda product family TCP Linda the Linda Code Development System CDS and Shared Memory Linda SM Linda The Linda product family are parallel programming languages based on the C and Fortran languages that enable users to create parallel programs that perform well in a wide range of computing environments This manual includes examples of using both the C based and Fortran based Linda products and refers to them as C Linda and Fortran Linda C Linda combines the coordination language Linda with the programming language C similarly Fortran Linda combines the coordination language Linda with the Fortran programming language Parallel programs are created with Linda by combining a number of independent computations processes into a single parallel program The separate computations are written in C or Fortran and Linda provides the glue that binds them together Linda has several characteristics which set it apart from other parallel programming environments Linda augments the serial programming language C or Fortran It does not replace it nor make it obsolete In this way Linda builds on investments in existing programs Linda parallel programs are portable and they run on a large variety of parallel computer systems including shared memory computers distributed memory computers clusters and networks of computers With few exceptions Linda programs written for one environment run without change on another
106. e rows to each successive workers until the numbet of elements it has exceeds the target number or until all the rows are gone This Linda User Guide 91 92 Case Studies simple heuristic works quite well so long as the number of workers is much much smaller than the number of atoms a condition invariably satisfied by real world molecular dynamics problems Once the starting position in the variable st art and the length out size have been computed the nbwconfig tuple is sent to tuple space its second field is workerid the worker s ID number assigned at startup The tuple s final three arguments are the total number of atom pairs and two constants The final for loop in the routine gathers the results of the special charges calculations performed in parallel by the workers The partial results from the qsum tuple are summed up by nb_energy_setup and used in the remaining sequential initialization activities after which the routine returns Before continuing with the main thread of the computation let s look briefly at the beginning of the worker function nb energy worker in the configuration for each worker in nbwconfig workerid start outsize fub n T xzpfac cont c if workerid num workers 1 PU A PAL B TB pack atc Farlen out done reading else for rdct 1 rdct lt num workers rdct in done reading in A At B B pack atc arlen T
107. e same behavior as under previous releases of Network Linda which used the tsnet command providing backward compatibility if you do nothing How ntsnet Finds Executables The nodelist and nodefile command line options correspond to the nodelist and nodefile resources respectively The value for nodelist must be enclosed in double guotation marks if multiple nodes are listed ntsnet nodelist moliere leopardi sappho hello world 4 Specifying Execution Priority Two resources control the execution priority of processes started by ntsnet The high resource application specific indicates whether processes are nice d or not it defaults to true If high is set to false then processes are run at lowered priority The command line options high high abbreviable to h and h correspond to this resource The nice resource node and application specific allows you to specify whether processes should be nice d or not on a node by node basis for each application For example the following lines state that processes running hello world on z0 ere should be nice d but those on chaucer shouldn t be although generally processes that run on chaucer are nice d Tsnet hello_world moliere nice tru Tsnet hello_world chaucer nice false Tsnet Appl chaucer nice tru The default value for the nice resource is true If the value of the high resoutce is set to true then it overrides the setting of the nice resource
108. e steps int npts first i k int nmin nleft int worker tpoints 0 nsteps 0 while tpoints lt NWORKERS tpoints gt printf Enter number IAXPOINTS of points along vibrating string n scant sd amp tpoints if tpoints lt NWORKERS 1 end while Cpoints gt printf Enter value between d and d n while nsteps lt 1 printf scanf nsteps gt MAXS Sd amp nsteps EPS Enter number of times steps n if nsteps lt 1 nsteps gt MAXS1 1 end while EPS printf Enter a value between 1 and din Put out total points and time steps out globals tpoints nsteps NWORKERS Assign chunks of work nmin tpoints NWORKERS nleft tpoints NWORKERS k 0 for i 1 i lt NWORKERS i npts i lt nleft nmin 1 nmin first k 1 eval worker worker i first npts k k npts Get results get_results tpoints NWORKERS int worker int id int first int npoints 4 int left right int tpoints nsteps nwrkers double ret val 146 Sample Programs AAXPOINTS NWORKE RS MAXPOINTS zj MAXSTE get global values printrf d first d npolnts d n id first rd globals tpoints nsteps nwrkers Initialize line init line tpoints npoints first Determi
109. e template does not require a string of a specific length This requirement holds for such templates even when the length of the literal string in the tuple and the declared length of the array used in the template are the same even if hi had been hello the colon would still be needed in the out operation Anonymous Formals A formal need not include a variable name but instead can use a data type name in C and in Fortran can use a typeof construct with a data type name as its argument Here are some examples C struct STYPE s in data Pint in struct struct STYP in int array Pant i 33 Fortran Common sample array num junk in data typeof integer in array typeof real 8 in common typeof sample Such formals are called anonymous formals Anonymous formals within in operations still remove a tuple from the tuple space but the data in that field is not copied to any local variable It is effectively thrown away This construct can be useful when you want to remove a tuple containing for example a large array from a tuple space Anonymous formals allow it to be removed without the time consuming copying that would result from an in operation A rd operation with a template containing only actuals and anonymous formals has no effect if there is a matching tuple but still blocks when none is available Fixed Aggregates in C In C you can reference fixed length aggregates such as
110. e this is a mapto entry it will never be used to translate a generic directory to a working directory on gogo an appropriate restriction given that automounted directories should not be explicitly referred to by their temporary mount points The last item we need to consider is translation rules for node degas This node is a special case in that users do not have permanent home directories there and are accordingly placed in the root directory when they log in However the system administrator feels that it is undesirable for remote processes to run from the root directory and wants them to run from tmp instead So we need to equivalence the working directory we re using to tmp Unfortunately given that there are no subdirectories under tmp corresponding to the various users who might run remote processes on the system we cannot write one simple rule to cover everyone Instead user chavez would need to include a rule like this one for het working directory within her local map translation configuration mapfrom home chavez work 1 degas tmp Map Translation The local directory u home chavez work is translated to the generic directory home chavez work which in turn is trans lated to the specified remote directory for each listed remote node a null translation occurs for nodes auroraandblake as well as for any other node not mentioned in the map file leaving the remote directory as home chavez work H
111. ead of the configuration files if desired Here is an example command which runs hello world overriding its usual maxprocspernode resource setting ntsnet maxprocspernode 3 hello world Command line options override any configuration file settings Remember also that local configuration file settings take precedence over those in the global system wide configuration file Resource Types In addition to their scope node specific application specific or node and application specific resources can also be classified by the kind of value they expect Some resources like those we ve looked at so far requite a value an integer or a directory for example Many others are Booleans and expect a true or false setting for their value For such resources the following values are all interpreted as true true yes on 1 These values are all interpreted as false false no off 0 For example the following entry indicates that the node marlowe is not currently available Tsnet marlowe available no ntsnet command line options corresponding to resources taking Boolean values use the following syntax convention If the option name is preceded by a minus sign then the resource is set to true and if it is preceded by a plus sign the resource is set to false For example the command line option useglobalconfig sets the resource useglobalconfig to true stating that the global ntsnet configuration file should be consulted The option
112. eate many processes each responsible for computing one row of the output matrix C If the matrices are large enough it might not be possible for each process to hold all of B at once In this case each process might then hold the row of A cotresponding to its output row and at any given time one column of B The process computes the dot product of its tow and the column it currently holds producing one element of its output row When it finishes with one column it sends it on to another process and receives a new column from some process Once all processes have received all the columns of B and have finished their final dot product the matrix multiplication is complete although the completed rows would still need to be explicitly moved from the component processots to the desired output location The message passing approach to the problem is illustrated in the diagram below It has a number of implications for the programmer First the program needs to keep track of which process has what data at all times Second explicit send data and receive data operations must be executed whenever data needs to move from one process to another Unless they are coded extremely carefully such bookkeeping and communication activities can cause bottlenecks in program execution The term processor is used in a generic sense here to designate a distinct computational resource whether it is one CPU in a multiprocessor computer or a separate computer o
113. eceives the join message is the worker added to the execution group and counted toward the minimum and maximum number of workers The following table summarizes the preceding information At a given time 7 having received p join requests the master process will act as follows States amp Outcomes p lt minworkers minworkers lt p lt maxworkers p gt maxworkers lt minwait wait wait SUCCESS minwait lt lt maxwait wait SUCCESS SUCCESS gt maxwait failure SUCCESS SUCCESS Selecting Nodes for Workers Table 1 lists the resources used to determine whether a node is used for a worker process When it wants to start a new worker process ntsnet determines which node to start it on in the following manner First it calculates a new adjusted load for each node assuming that the process were scheduled to it using the following formula initial load Nmaster masterload Nworker workerload speedfactor This quantity represents a load average value corrected for a variety of factors The various components have the following meanings initial_load If the getload resource is true its default value this is the load average obtained from the node over the period specified in the loadperiod resource If the remote procedure call to get the load average fails the value in the fallbackload resource is substituted If getload is false then zwitial_load is set to 0 One potential use of the fallbackload resource
114. econd row of array b The third and fourth operations divide array a in half and place the two halves into the tuple space defaulting the beginning and ending elements in the third dimension to the first and last array element respectively The final operation illustrates the use of the stride value Fortran Named Common Blocks Entire named common blocks can be transferred to a tuple space as a single unit Named common blocks ate referenced by their common block name enclosed in slashes For example the following operations place and retrieve the designated common block Common example a b c d n out common example in common example The out operation places the entire common block into the tuple space and the following in operation matches and retrieves this same tuple For matching purposes common blocks with identical names match and the internal structure of the common block is ignored Blank common may not be placed into tuples using this method The best solution in such cases is usually to convert it to a named common block Linda User Guide 33 C Structures From the point of view of syntax structures work very much like scalars For example these two out operations create identical tuples struct STYPE S t pr p s out structure s out structure p Hither of these in operations will retrieve one of the tuples created by the previous out operations in structure t in
115. ed for the various remote systems e g executables with different compilation options all required executable files must be in the same directory as the local executable and they will be copied to the target remote execution directories as determined by the specific values set in the rexecdir resource for each node or by map translation The distribute resource is false by default The cleanup resource application specific determines whether remote executables are removed after the execution completes Its default value is true Note that the local executable is never removed regardless of the setting of cleanup The cleanup resource setting is relevant only when distribute is true The distribute resource may also be set with the distribute distribute command line options both abbreviable to d The cleanup resource can be set with the cleanup cleanup command line options Architecture Specific Suffixes By default the names of executable files on all nodes ate the same as the executable name given on the ntsnet command line When executables are distributed prior to execution this executable is the one that is copied by default Howevet it s possible that different flavors of the executable should be run on different nodes ntsnet provides a mechanism for specifying which executable to use based on an architecture specific suffix For example ntsnet can be told to use executables having the extension 1686 on some nodes and t
116. either constants or expressions which resolve to constants and some of which hold placeholders for the data in the corresponding field of the matched tuple in tuple space These placeholders begin with a question mark and are known as formals When a matching tuple is found variables used as formals in the template will be assigned the values in corresponding fields of the matched tuple Here is an example C Form Fortran Form simple i simple TEJ In this template the first field is an actual and the second field is a formal If this template is used as the argument to a rd operation and a matching tuple is found then the variable i will be assigned the value in the second field of the matched tuple A template matches a tuple when They both have the same number of fields The types values and length of all actuals literal values in the template are the same as the those of the corresponding fields in the tuple The types and lengths of all formals in the template match the types and lengths of the corresponding fields in the tuple We ll consider these conditions in more detail in Chapter 2 for now let s look at some examples If the tuple H Also known as an anti tuple 12 Overview of Parallel Programming and the Linda Model The Linda Model C Form Fortran Form cube 8 512 cube 8 512 is in tuple space then the statement C Form Fortran Form rtd cube 8 1 rdi cube 8 21
117. either normal or single step mode depending on the setting of the Single Step item on the Modes menu When single step mode is in effect Tuplescope executes until the next Linda operation takes place Tuplescope assumes that required source files are in the current directory It is not possible to execute a program more than once within a single Tuplescope session To rerun a program exit Tuplescope and restart it Clicking on the Break button will cause program execution to pause at the next Linda operation Execution will resume when you click on the Continue button which may also be used to resume execution in single step mode To exit Tuplescope click on the Quit button Linda User Guide 99 Using Tuplescope with a Native Debugger The following method is recommended for combining Tuplescope with a native debugger like dbx Compile the program for Tuplescope and the native debugger by including any necessary compiler options on the clc command i e g for dbx linda tuple scope for Tuplescope Execute in single step mode in Tuplescope until the desired process is created Click on the new process icon with the middle mouse button This will create a new window running the native debugger attached to that process Set desired breakpoints in the native debugger Then turn off single step mode in Tuplescope Give the continue execution command to the native debugger to resume execution of that process e g use the
118. el Programming and the Linda Model How and Where to Parallelize This sort of display summarizes the number of times a given routine was called Often it is helpful to also know where a routine was called from A call graph table will indicate this information Here is an example called procedure fcalls calls from line calling proc file exp T557120 36 78 48 gauss gaus3 t 3022848 14 71 63 gaus3 gaus3 f 3022848 14 71 79 gaus3_ gaus3 3022848 14 71 95 qgaus3 gaus3 f 503808 2 45 143 Gauss gaus3 t 503808 2 45 127 gaus3 gaus3 f 503808 2 45 111 gaus3 gaus3 f 503808 2 45 159 gaus3_ gaus3 f 503808 2 45 L73 Gauss gassi 503808 2 45 LIL gaus3 gaus3 t Bart 1007616 15 03 111 tunes funez tt 1007616 15 03 110 furc3 funca E 1007616 15 03 108 tunes fune3 tf 1007616 15 03 109 func Lunes 503808 Td 44 tunes Eines E 503808 3 51 147 func3 tfunc3 f 503808 3 148 func3 func3 f 503808 Tol 149 fune3 funes E Here we can easily see that the exponential function is called literally millions of times all from within one routine We would want to try to do some of those calls in parallel if they are independent Linda User Guide 15 16 Overview of Parallel Programming and the Linda Model 2 Using the Linda Operations In the last chapter we discussed the process of creating parallel programs from a general point of view We also introduced tuples and the main Linda operations
119. ements since Fortran subroutine arguments are always passed by reference Second the first and third array indices are reversed due to the differing Fortran and C multidimensional array ordering conventions The Fortran arrays have not changed in any way so the location denoted by the Fortran X i1 i2 i3 for example must be accessed as x i 3 12 11 from C Third C arrays begin at 0 while the Fortran arrays begin at 1 However this is easily accounted for by making iseg run from 0 to nsegmnt 1 rather than from 1 to nsegmnt as it did in the Fortran loop Fourth we ve also added the addresses of the arrays x y and z to the subroutine s arguments in Fortran CALCDISP accesses them via a COMMON block not shown here Finally we ve translated the subroutine name to lowercase and appended an underscore a common requirement when calling a Fortran subroutine from C After the inner loop completes the worker sends the displacement values it has created to tuple space and the outer loop begins again When the counter in the index tuple exceeds its maximum value each worker process will terminate the next time it examines it Linda User Guide 41 As we ve seen it was very easy to transform Freewake into a parallel program with Linda because all of its work is isolated so well within a single routine In the next chapter we ll look at several more complicated case histories 42 Using the Linda Operations Quick Start 3
120. en local directoty trees and directory trees on remote nodes If your network presents a consistent view of a common file system via NFS or AFS for example then you will not need to worry about map translation On the other hand if your networked file systems ate not completely consistent if a file system is mounted as home on one system and as net home on another system for example then map files can be a great help in automating TCP Linda execution Basically map files provide a way of saying When I run a program from directory X on the local node always use directory Y on remote node R Map translation occurs for both the execution directories locations of executable files and working directories on each node How ntsnet Finds Executables Map translation means that ntsnet translates local directories to their properly defined eguivalents on a remote node before specifying an executable on or copy an executable to that remote node executable file distribution is discussed later in this chapter If map translation is enabled as it is by default but no translation is explicitly specified for a given directory and or node whether because its rules haven t been specified or no map translation file exists the rule is simple look for the same directory on the remote node If enabled map translation occurs whether the local directories ate specified explicitly as when the full executable file pathname is given on t
121. ergy setup str coor count Perform sequential setup work global data all workers get once Oub A ESTELARES B sStr B 2 pack ate packatomte istr nti taskav total work workers target pairs per worker taskav nb num nb num 1 2 num workers create worker tasks start beginning of worker s domain outsize size of worker s section for index 1 w id 0 w id lt num workers w id start index increments of index follow compute start amp length for this worker tasksum holds the number of pairs given to this worker so far increment it with each successive value of index size of current matrix row until task gt target or we re out of pairs for tasksum 0 tasksum lt taskav amp amp index lt nb num l index tasksum index if w id num workers 1 last worker outsize nb num 1 start else outsize index start start workers initialization phase out nbwconfig wid start outsize nb num expfac confac Fill worker specific data arrays and send to tuple space end worker init loop get partial g sums and nb list size calculated by workers qasum gbsum 0 0 for i 0 gt i lt num workers itt in qsum uasum ubsum ucount qasum uasum qbsum ubsum count ucount Molecular Dynamics Finish initialization return After performin
122. es data IRR M eene el e Saltem uus Va 123 Menu Buttons oou cen oie ure A RA ra ca ee a etae 123 The Modes Menu oic re A a a a REPRE 123 The Debug MER 1 0 s e Rr vie xD ERE TOES od Y Ee 124 TID Loap Syntax pi sem eed yer EPOR Ree LET DESC ES Ee eges 124 7 Sample Programs eel n 127 Array ASSI E SA AAA EPIIT eo O AS E Gye 127 Serial Version oi aje e pem ees EIE IRAE ARS re ET OA P ees 127 Parallel Version z aeneo AA noone erin Sipe one Sao TUS 128 pi Calculation zoe ate esten pu V Bay as dae eel ae SESS etam a A 132 Serial Version s imos in ed ke ees eh e ere ene ER Sees 132 Parallel Verlo cce eee ete hcm eme P gie tel e t eere els 134 Matri Multiply usce dere RR ERA SERA HRS waned RC ER o t ee As 137 RN uei sO skens re aces voee tS ep Se ee ee Sere ese RE ERGY CERIS BBA 137 Parallel Versiones ose seven ee ERE REI exequi e pre cuente I 138 Concurrent Wave Equation cette eens 142 Serial Version aue tg wie ees A lacks eee oeste an sha daa no ena o Sealers eel utet 142 Parallel Verso eee e HR ERR Rep eere Ee p reg 145 Linda User Guide iii 2D Heat EQU ts sage as a v pl hey ea eS en Oe EE SP See ae 150 DELAL VERSION cte li ee RD ed A CA aM ST de odd a Dah nde uod 150 Parallel VersIOD us cece metet teet E a wo ara dele a Sak Base ee A los 152 Bibliograplij cobos hess wee oboe ot ad e Vet ek i lndex eb a aa ee a a VE Va fg NU ees i iv Contents Introduction This manual describes th
123. est The first line sets the default location for remote executables to usr local bin on the remote system meaning that ntsnet should use this directory on each node as the default location for the executable file to startup Subsequent lines set different default values for the application hello world and for the node o ere The final line sets a specific value for the application hello world when running remotely on molere the executable for hello world on ofer resides in the directory usr linda bin test The rexecdir resource can also be given the special value parallel which is its default value This indicates that map translation is to be performed on the directory where the local executable resides for evety node which has no specific remote execution directory set Map translation involves taking the name of the local directory containing the executable to be run and determining what the equivalent directory is for each remote node participating in the job These equivalent directories are determined according to the rules set up by the user in the local and or global map translation files The format of these files is discussed in the next section The p command line option specifies the values for both rexecdir and rworkdir overriding all other methods of specifying them p is discussed later in this chapter About Map Translation 52 Using TCP Linda As we ve stated map translation is a way of defining equivalences betwe
124. f tuple space in addition to running the application program Therefore when a process is paused for example at a breakpoint then no tuple space requests can be handled by it For this reason it s best to break only a single process at a time with all other processes either continuing or stepping over an in operation To do so effectively you need to have compiled the application with g Linda User Guide 71 72 Using TCP Linda Ray Tracing 4 Case Studies This chapter provides detailed discussions of several real applications parallelized with Linda In some cases the descriptions will follow the programming process recommended in Chapter 1 by first transforming a sequential program into a parallel program in a straightforward manner and then if necessary optimizing it to produce a more efficient parallel program in others we ll focus on topics and issues of special importance Space limitations will not allow for a full treatment of any of these programs but these case studies illustrate many important techniques useful to any programmer working with Linda Image rendering is a very computationally intensive application The Rayshade program written in C renders color images using ray tracing It is capable of including texturing effects to simulate different materials in the image multiple light sources and types point spotlight diffuse and atmospheric effects like fog or mist Rayshade computes the effects
125. fies whether or not the executable s shall be distributed to the remote nodes Executables are distributed only to those remote nodes that are actually going to take part in the execution After the execution completes ntsnet automatically removes the remote executables that it just distributed The local executable is protected from removal The default is false See the cleanup resource for information on preventing the automatic removal of remote executables Specifies the load average the scheduler shall use for a node if the RPC call to get system load average fails The default is 0 99 The value specified can be any real number gt 0 If failure of the RPC call indicates that the node is down this option can be used to set fallbackload to a very large value effectively making the node unavailable to ntsnet Specifies whether ot not to use load averages when scheduling workers on the nodes in the network The default is true This can be used to make worker scheduling consistent between different runs of ntsnet It also makes sense if the rstatd daemon is not available on the network Specifies whether all workers shall run at normal ptiority and Linda internodal communication should run at full speed The default is true If the high resource is false the workers run nice d unless specifically overtidden on a per node pet application basis using the nice resource note that high being true overrides the setting for nice Also when
126. g the same initial setup steps as the sequential version the parallel version of nb energy setup places some global data in tuple space The majority of the code shown in this routine is concerned with dividing the work up among the worker processes The tasks created here define the range of the data each worker is responsible for This represents a somewhat different technique from the usual mastet worker scenario In the latter the work is divided into tasks which are independent of any particular worker and each worker grabs a task when it needs one Here the total work is divided among the workers at the beginning of the calculation the wotk each worker does is essentially part of its structure in this case a function of its wotker ID number This technique of dividing the problem space into discrete chunks and assigning each to one worker is known as domain decomposition The parameters calculated here and communicated to each worker via tuple space will be used in both the summing of special charges done in the setup phase with results collected at the end of this routine and in the actual nonbonded interactions computation later in the program The scheme for dividing the work here is somewhat complex but it is designed to ensure good load balancing among the worker processes this is always a concern when a domain decomposition approach is taken For a molecular dynamics calculation the total number of nonbonded interactions fo
127. h of the nodes Most of the work can be done by this map entry which uses home as its generic directory map home chaucer u erasmus mf flaubert u home gogol net aurora home This entry translates the listed local directories to the generic directory home The translation applies to the entire tree starting at the specified locations Thus in our example ntsnet will translate the local current working directory on flaubert u home chavez work to the generic directory home chavez work we don t have to explicitly construct a rule for the current directory When it needs to detetmine the wotking directory for a remote process participating in the execution of the application test24g it will use this genetic directory So when it starts a remote worker process on chaucer it will use the directory u chavez work as the current directory translating home to u as specified in the rule for node chaucer When ntsnet starts a remote worker process on blake it still attempts to translate the generic directory but no rule exists for blake In this case a null translation occuts and the directory remains home chavez work which is what is appropriate for this system The rule we ve written will allow us to run our ntsnet from the work subdirectory of cbavex s home directory on nodes aurora and blake and on any of the listed nodes except gogol in each case the current working directory will translate to home
128. h shell script called by ntsnet to start up a worker process on a remote node This resource provides a hook enabling users to control the behavior of the shell script which can itself be modified by the user In the default implementation of linda rsh located in the Linda bin subdirectory only the string on is meaningful as a value to this resource If on is passed to linda rsh then the on command will be used instead of rsh to initiate the remote process This is a node specific resource Specifies the number of minutes over which the machine load is averaged This is the load average then used by the worker Typical values for loadperiod ate 1 5 and 10 The default is 5 Specifies the load that the master real_main process is considered to put on the node The value specified can be any real number gt 0 The default is 1 Typically 1 or some smaller fraction is used If the master process uses much less CPU time than the workers then masterload should be set smaller than workerload Specifies the maximum number of nodes on which to execute The default value is the number of nodes in the node list maxprocspernode 118 Linda Usage and Syntax Summary Specifies the maximum number of Linda processes started on any given node the application is running on On the local node maxprocspernode includes the master The default value is 1 maxwait maxworkers minwait minworkers nice nodefile nodelist ntsnet
129. he colon when omitting the length while in Fortran the colon may also be omitted although including it is correct too For example the first out operation in each language places the entire array a into the tuple space If you only want to place part of an array into the tuple space then you can include an explicit length in the out operation In this way the second out operation in each language places only the first ten elements of array a into the array2 tuple The array format is basically the same for arrays used as formals in templates The one difference is that an integer variable is used for the length parameter and its value is set to the length of the array in the matching tuple Thus the final pair of in operations in the preceding example will both result in the value of len being set to 10 The semantics of Linda ensure that a user may alter the contents of an array immediately upon return from the out the tuple generated by the out will still reflect the contents of the array just prior to the out To enforce these semantics Linda makes a copy of the array before returning from the out Shared memory implementations of Linda allocate space for this copy in the shared memory region Non shared memory implementations of Linda allocate memory in the address space of the process executing the out Avoiding Array Copies This copying can be problematic for very large arrays It can neatly double memory requirements and a
130. he command line or determined implicitly using the TSNET_PATH for example Thus map translation will occur for both of the following commands ntsnet tmp test24 ntsnet test24 In the first case ntsnet will translate the directory you ve specified tmp for each node where the application test 24 will run if no explicit translation has been defined then ntsnet will perform a null translation and look for the program test 24 in tmp on each remote node as well For the second command ntsnet will first determine the location of the test 24 executable on the local node using TSNET PATH or its default value and then translate that location for each remote node involved in the job If the rworkdir resource is set to parallel the default value then the current working directory is also subject to map translation The Map Translation File ntsnet uses the local and global map translation files tsnet map and common lib tsnet map relative to the Linda tree respectively to determine the correspondences between directories on different nodes The first matching entry is used to perform each translation The map translation mechanism is extremely powerful and can be used to define equivalences among systems whether or not their file systems are linked with NFS Map translation is a two step process Rather than having to specify the exact translation for every pair of hosts within a network map translation allows you
131. he worker next obtains the current coordinates of the atoms in wakeup wakeflag workerid if wakeflag 1 return 0 eat poison and die if workerid num workers 1 rdi coor Teilen out read _ coords 7 else Linda User Guide 93 94 Case Studies for rdct 1 rdet lt num workers rdct in read coords in eoor c len Most workers rd this tuple however the last worker waits until all the other workers have rd it using the same technique of gathering semaphore tuples we saw earlier before removing it from tuple space with the in operation In addition to freeing the memory this is necessary so that the new coordinates can be unambiguously transmitted to tuple space on the next iteration The worker then calculates the nonbonded interactions for its pairs of atoms It uses code differing only in its loop limits from that in the sequential version of nb energy for i190 gi start i lt outsize i git f do many computations The initial value of the vatiable gi and the limit against which the variable i is tested were both obtained from the nbwconfig tuple Once this calculation is complete the worker sends its results to tuple space out workerdone fu start outsize elec evdw esup count It then waits for the next wakeup tuple commencing the next iteration of its while loop eventually it retrieves a poison pill and exits nb energy ultimately gathers th
132. held at any given time 2 c1ump M total elements would fit within the available amount Or if the program were running on a network containing processors of greatly differing speeds then it might sometimes be preferable to make the Matrix Multiplication task size smaller so that there are enough total tasks to keep every processor busy for essentially the entire execution time with faster processors completing more tasks in that time Building such adjustability into a program is one way to ensure easy adaptability to different parallel computing environments Even this version makes some optimistic assumptions however For example it assumes that the master process has sufficient memory to read in both matrices before sending them to tuple space If this assumption is false then the disk reads and out operations would need to be interleaved to minimize the amount of data the master has to hold in local memory We ll close this consideration of parallel matrix multiplication by looking at a case where one might not want to use it Here is a portion of a subroutine from a computational chemistry program written in Fortran Subroutine gaus3 x n Loops containing many independent exponentials call matrix multiply chi coo psi call matrix_multiply chix coo gx chiy coo gy call matrix multipl ehiz coo gz y call matrix multiply y call matrix multipl GhidZ coo d2 Return End
133. here fe is linda op operation is either or and est_value is one of the following eval out in rd block_in and block_rd This kind of test detects the occurrence of a particular kind of Linda operation Here is an example linda_op eval This condition detects the occurrence of an eval operation Process comparison tests where ez is process operation is one of lt and and est_valueis an integer representing a Tuplescope process number This sort of test detects when any tuple space operation is performed by any process whose process number fulfills the condition For example this condition detects when any process with process number less than 5 accesses tuple space process lt 5 Multiple conditions can be joined with and or or as in this example process lt 5 and linda_op out The entire composite condition is enclosed in parentheses in the complete TDL statement as we ll see below Action can be one of the following Action break Effect Pause program execution resume execution with the Continue button Ifa tuple is associated with the tuple space operation that triggers a break Tuplescope turns it solid black The process which performed that tuple space operation is marked by a notch in its icon Ordinary display of all tuples and process icons is restored when you click the Continue button The Tuplescope Debugging Language hide Suppress display of matching tuples or
134. hese statements retrieve the nbwconfig task tuple and the global values tuple from tuple space After it starts up the worker will block until the nbwconfig tuple is available The appearance of this tuple is a trigger that initiates active computation It is in this sense that we refer to it as a wakeup for the wotker causing it to resume active execution after a significant pause sleep The if statement checks whether this is the worker with the highest worker ID If not it rds the globals tuple and then creates a done reading tuple If it is the last worker then it ins all the done reading tuples from the other workers and then ins the globals tuple removing it from tuple space This technique is useful when you want to remove large unneeded sets of data from tuple space or when some data must be removed because new or updated versions will replace it We ll see an example of the latter later in the program The wotker next computes its portion of the special charges sum and dispatches the results with an out operation It then enters its main infinite while loop performing a few data initialization operations fot the coming nonbonded interaction calculation and then waiting for its next wakeup tuple again tied to its worker ID and appropriately labelled wakeup Molecular Dynamics qasum qbsum 0 Calculate new values out qsum qasum qbsum count while 1 vdw elec esup 0 0 count
135. hould be displayed The default is false The veryverbose and verbose resources are independent workerload Specifies the load that a worker will put on a node The value specified can be any real number gt 0 The default is 1 Typically 1 or some smaller fraction is used A larger value could be used to increase the chances of having one Linda process running on each node workerwait Specifies the time in seconds that a worker waits for a response to its join message from the master The default is 90 unless debug is true in which case it is 1000000 If a worker does not get a response within the specified time telling the worker that it has joined the worker will exit and therefore not participate in the execution of the application Map Translation File Format This section documents the format of the map translation files used by TCP Linda These files ate t snet map in the user s home directory the local map file and lib tsnet map the global map file located relative to the Linda tree Map file entries have one of the following formats map generic directory nodel specific directory node2 specific directory mapto generic directory nodel specific directory node2 specific directory mapfrom generic directory nodel specific directory node2 specific directory Linda User Guide 121 Note that generic directory need not be a real directory location at all but can be any
136. ied dimensional heat equation domain decomposition The initial perature is computed to be high in the middle of the domain and o at the boundaries The boundaries are held at zero throughout simulation During the time stepping an array containing two ains is used these domains alternate between old data and new data OKCKCKCKCKCkCkCk Ck kCk ck kCkc k kCkCKCkCk CK Ck kk Ck kCk Ck kCk CK Ck kCk Ck kk Ck kCkCk KCk Ck kCk Ck k kc k Ck kck ck kok ck k kk k kk kok ck ke de lt stdio h gt define NXPROB 11 define NYPROB 11 str t Farms E f Joa Loa E CE t cy int nts parms 10 1 0 1 BUT mai n f int at u 2 NXPROB NYPROB voi kkk ini iX iy iz it d inidat prtdat update CKkCkckCckck AKA AKA AKA A KA ck ZA Z X A X KA X X ck X ck ck X AZ A X AZ ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck k kk k k kk kk k Initialize grid OKCKCKCKCKCkCkCk Ck kCk ck kCk k kCk CK kCk CK Ck kCkCk kCkCk RR RR k k kCkCk IA k kc k kckck ck kok ck kok ck k kk kok e ke dat NXPROB NYPROB u 150 Sample Programs 2D Heat Equation prtdat NXPROB NYPROB u initial dat for ix 0 ix lt NXPROB 1 ix wpr pzs 107 nl llizilV u 1 ix NYPROB 1 u 0 ix NYPROB 1 for iy 0 iy lt NYPROB 1 iy urr ro liv 34101 107 liy u 1 NXPROB 1 iy u 0 NXPROB 1 iy KKK KK KK oko k oko k k k
137. ing the same routine used by the sequential program However instead of an immediate call to out line the computation is followed by an out operation placing the results into tuple space Sometime later do_t race will retrieve the line and then make the call to out line to write it to the output file Meanwhile the worker has started its next task It will continue to perform tasks until all of them are done worker bears a striking resemblance to the original version of do trace it adds a call to rayshade_main unnecessary in the sequential version and it lacks only the call to outline from its main loop which remains behind in the master process version of do_trace This separation results from the need to pass data between master and workers via tuple space It is also much easier to code than sending explicit messages between all of the various processes Matrix Multiplication Matrix multiplication is a common operation central to many computationally intensive engineering and scientific applications We ve already looked at matrix multiplication in a general way in Chapter 1 Here we will look at a couple of approaches to parallel matrix multiplication and conclude by considering when one should and should not use them This example again uses C Linda This case study illustrates the following techniques Adjustable granularity and granularity knobs Deciding where in a program to parallelize The basic matrix mu
138. interactions NSteps out worker wakeup out current coordinates in wakeup Exit rd or in coordinates v Calculate nonbonded interactions out results in results amp fill matrix Calculate new atomic Exit coordinates For each loop iteration the master calculates the bonded interactions and other energy terms and then divides the work of the nonbonded interactions among the workers placing the data the worker will need into tuple space Eventually the master gathers the workers results using them along with the quantities it computed itself to calculate the new position and velocity for each atom in the molecule Here is real_main which is a renamed version of the original main real_main argc argv T 300 0 process args argc argv Molecular Dynamics startup workers for i 0 i lt num workers i eval worker nb energy worker i num workers Read data Initialize parameters amp data structures verl init str coor vel s Te temperature str kinetic_energy str vel verl scale coor sqrt T Te for i 0 i lt NSteps i verl_step str coor vel s Te temperature str kinetic_energy str vel verl scale coor sqrt T Te kill workers amp clean tuple space for i 0 i lt num workers i out wak up l 2 in worker Pant best Ah use C Linda exit function
139. ion file syntax is modeled after the Xlib resources of the X Window System However ntsnet and TCP Linda are neither part of X nor do they require it Some users may find a general introduction to X resources helpful see the Bibliography for the appropriate references A resource is basically just an execution parameter characteristic The configuration file defines values for the various available resources and specifies the contexts application programs and or execution nodes for which the value applies The ntsnet configuration files consist of lines of the following form program application node resource value where program is the system program name to which the resource being defined is applied application is the relevant user application program node is the relevant node name resource is the resource characteristic name and value is the value to be set for this resource in this context We ll look at each of these parts in turn To begin with the system program for the ntsnet configuration file entries will always refer to TCP Linda We ll look at the exact syntax of this component in a moment At this point in time this component is simply a carryover from the X syntax A resource can be either application specific node specific ot apply to both applications and nodes For example the resource rworkdir which specifies the working directory on a remote node is application and node specific meaning that a diffe
140. is designed to prevent attempts to start new workers on nodes that have gone down If the RPC to get the load average for a node fails and fallbackload is set to a large value then it will be quite unlikely that the node will be chosen by the scheduler Setting fallbackload to a value greater than threshold speedfactor will ensure that it is never chosen N n 1 if the master process is running on the node 0 otherwise T ntsnet makes use of rstatd to gather this information If this service is not provided on a node default values are assumed Linda User Guide 63 masterload The second term in the numerator of the formula for the adjusted load means that the value of the masterload resource is added to the raw load average if the master process is running on the node This enables an estimate of how the CPU resources the master will eventually use to be included in the adjusted load average even though it is not currently consuming them Set masterload to a smaller value than its default of 1 if it does not consume significant resources during execution for example if it does not itself become a worket N optant The number of worker processes already started on the node 64 Using TCP Linda Table 1 Process Scheduling Resources ntsnet Worker Process Scheduling Resource Meaning processes that can be run on any node Includes the master process on the local node Default Value Equivalent ntsnet Option s
141. k trace jit line y out buf Out result v out but Xres return The worker begins by reading the command line arguments from tuple space into a local array with the rd operation It then calls rayshade_main to perform the necessary initialization and setup activities It is often just as fast to have workers each repeat the sequential program s initialization process rather than doing it only once and then attempting to transmit the results of that process to the workers through tuple space However when you choose to have the worker repeat the program initialization it is important not to repeat steps that must occur only once Here the workers call rayshade main with a third argument of 1 ensuring that picture initialization is not repeated The wotker function next retrieves the global parameters from tuple space and sets the sampling parameters using the same code as originally appeared in do trace Matrix Multiplication Each iteration of the while loop computes one scan line of the final image and places it into tuple space for later collection by the master process executing do trace It removes the scaninfo counter tuple from tuple space decrements it and puts it back so that other processes can use it If the counter has fallen below zero all scan lines are finished the loop ends and the worker returns terminating its process If the counter is non negative the worker generates that scan line by call
142. k d n id prank data d d Wn index datalindex printf data d din index 10 data index 10 print 0 data d d n index 20 datalindex 20 Printe M Wy bb now kill off the workers to do 1 e chunksize 0 i 1 while i lt numworkers out my section 0 0 iri printf MASTER All Done n by telling them there is no more work KOR KKK KKK KK KK kok k koe ke ke e e k worker task TK A kck ck kck ck RK int worker int id local variables int index int data ARRAYSIZI int chunksize 1 int next index Di 1 Get the initialized array from tuple space rd init array data loop until no more work 1 while Index into my portion of the array that was passed in in my section index chunksize 130 Sample Programs KE Get the chunksize py ret THE DESC USE Note K a CFLAGS array C Array Assignment check to see if there is more work if chunksize y d there is no more work so let other tasks know return 0 Do a simple value assignment to each of my array elements i index while i lt index chunksize data i i 1 Itt Send my results back to the master ut result data id index data end of loop utn 0 FEFE HE TE FE FE FE FE FE HE FE EEE EE HH HER HER ER HE
143. kCkCK Ck kCK Ck kCKCk KCkCk k kc k k kc k k kc k k kc k Ck kck ck k I kk A x include lt stdio h gt defineARRAYSIZE600000 main 1 int i loop variable floatdata ARRAYSIZE the intial array KOR KKK KKK KK KK kk k koe ke ke e e k initializations OKCKCKCKCkCKCk ck k kk kk k kk kk X kk e e kx x printf Tyn tkrk Starting Serial Example EXARHEREFRA QU fElush stdout Initialize the array for i 0 i lt ARRAYSIZE i data i 0 0 Do a simple value assignment to each of the array elements Linda User Guide 127 for i l i lt ARRAYSIZE i data i i 1 printf Sample results n pesmttrit data 1 f n data 1 printf data 100 f n data 100 print data 1000 f n n data 1000 rtriuwshisbdout printf All Done Xn k k k AE FE ERE Serial Array Makefile FILE make ser array c DESCRIPTION See ser array c 4 USE make f make ser array c k k k FE FE FE FE AE ERE CC array Ser array c s4CG sef array c o array Parallel Version KR k ko kok k kok k oko ko kok k oko k kok ARK RR RRA k ok k RA k kok RRA RRA kkk kkk kk kck ck kok ck kk ck k kk B LINDA Example Array
144. ker L M N rd columns columns len while 1 im task Pay if L gt Ly out tssk i break else out task itl vales y BBE for j 0 j lt N j result j dot product row columns j M M out result i result N return The worker next reads in the row of A that it needs Its final loop computes each element of the corresponding row of C by forming the dot product When all of its entries have been computed the worker sends the completed row of C to tuple space and begins the next task Matrix Multiplication Unfortunately this version has a disadvantage that not only inhibits it from always performing well but sometimes even prevents it from running at all It assumes that each worker has sufficient local memory to store the entire B matrix For large matrices this can be quite problematic For such cases we could modify the master to place each column of B into a separate tuple and have the workers read them in for each row of A a column at a time but this would add significant communications overhead and probably yield poor performance Instead a more flexible program is needed where the amount of work done in each task can be varied to fit memory limitations and other restrictions imposed by the parallel computing environment Here is a version which fulfills this goal master L M N Allocate memory for matrices Read in matrices put matrices int
145. kok I IK k kkk k kkk k kkk k kkk k kkk ck kokck k kkk I k kk k Iterate over all timesteps KKK I RRA RR k I kCkCk RARA RR RR RARA RA k k kCk Ck kc k Ck kck ck kok ck ckck ck I RRA ARAS iz 0 for it 1 it lt parms nts it update NXPROB NYPROB amp u iz 0 0 amp u 1 iz 0 01 iz 1 iz prtdat NXPROB NYPROB amp u iz 0 O final dat KR KK kok k RARA kok k RR KAR KR k kok k oko k A k oko kok k kok k kkk K KC k kok k kok kok ck I e k k subroutine update XKOKCKCKCKCKCkCKCk kCk Ck kCkCk kCkCKCkCkCK Ck kCk Ck kCKCk I I kCK I K KCkCk A I kk I e e kx void update int nx int ny float ul float u2 int Xx 1y for ix 1 ix lt nx 2 ix for iy 1 iy lt ny 2 1y Ue ti ole nyt y parms cx ul ix l j nytiy fui 1x 1 nytiy 2 0 Ul ixtnytiy parms cy Kiul ix n tiy rl ul ix nytiy L 2 0 lyltix nriy js KOR KKK KKK kok kok k oko k A A A A k kok k kok k A k kok ck kok ck I HK kk subroutine inidat kok kok Ck KCk Ck kCkCk kCkCk Ck kCK Ck kCk Ck kCKCk KCkCk kok k Ko k Ck kCKCk kCkCk kCkCk A kCk Ck kc kckckck ck kok ck k kk ckck X kk kk void inidat int nx int ny float ul 4 int qx iy for ix 0 ix lt nx 1 ix for iy 0 iy lt ny 1 iy iplris aytiy o Elost ix nx ix 1 iy ny iy I e Linda User Guide 151 KR k kok ok k oko k ok ko ok k k kok ok
146. kok k Ck kCKCk KCkCk kCk Ck k kc k Ck kc k k kc k ck kck ck k kk ckck kc kk ke e kx kx n 1 n lt darts n 1 generate random numbers for x and y coordinates r double random cconst x coord 240 B XD r double random cconst y coord 2 0 r 1 0 if dart lands in circle increment score if sqrix coord sgr y coord lt 140 score calculate pi pi 4 0 double score double darts return pi EEEH f Serial pi calc Makefile FE AE TE AE AE AE FE FE FE AE AE AE FE FE FE AE AE AE FE AAA AE E FE AE E AE AE E FE AE AE AE AE AE FE AE E AE AE AE E FE AE AE AEE AE FE FILE make ser_pi_calc c DESCRIPTION See ser_pi_calc c USE EEEH QE make f make ser_pi_calc c FE AE TE AE AE AE FE FE FE AE AE AE FE FE FE AE AE AE FE FE FE AE AE AE E FE FE AE AE AE AE FE AE AE AE AE AE FE AE AE E AE AE E FE AE E AE AE AE FE AE FE AE AE AE FE AE E AE AE AE AE FE FE AE AEE FE FE cc pi calc ser pi calc c BICOI Ser pi calc c o pi calc Linda User Guide 133 Parallel Version A ACA F F F F LINDA pi Calculation Example C Version FILE clind pi _cale el OTHER FILES dboard c make clinda pi calc c DESCRIPTION Linda pi calculation example program C Version
147. lculates the displacement of a single point in three dimensional space Each call to CALCDISP is independent It takes the x y and Z coordinates of a point in space elements of the arrays X Y and Z respectively and produces the displacement in each direction of the specified point in space as its only output elements of the arrays DX DY and DZ Thus performing some of those calls at the same time would have a large effect on overall program performance 38 Using the Linda Operations C Version Fortran Version Example Freewake Here is the key part of the code which serves as the master process and coordinates the calls to CALCDISP put data into tuple space out wake x y z nblades nfilmnt nsegmnt out indesx 0 create task counter for i 0 i lt NWORKERS i start workers eval worker worker worker then become a worker for index 0 index lt nblades nfilmnt index gather data from tuple space in delta index TEmp x tmp y Ptmp x Put data into the displacement arrays DX DY and DZ e Put data into tuple space out wake x y z nblades nfilmnt nsegmnt C Create task counter start workers then become Ge a worker yourself out index 0 Do 10 I 1 NWORKERS eval worker worker 10 Continue Call worker Do 20 index 0 nblades nfilmnt 1 in delta index tmp x tmp y tmp z Put data into the displace
148. lel Programming Shared Data Space est Tasks B Worker reads required data Worker sends results Worker Processes Distributed Data Structures Matrix Multiplication In a distributed data structures parallel program workers retrieve tasks from the shared data space complete them and then repeat the process until all tasks are done Here each task is to compute a portion of the result matrix C Each worker reads the data it needs for its current task here the relevant portions of A and B from the global data space and places its results there when finished as well The master process which generated the tasks and placed them into the shared data space also eventually gathers the results placed there by the worker processes Note that the tasks and workers are also independent of one another The total work is not split among the workers rather the total work is split into a number of chunks and each worker performs task after task until the entire job is done In this case it is the task size and not merely the number of workers that primarily determines the granularity of the program and this granularity can be adjusted by varying the task size If we look again at our matrix multiplication example each task might consist of computing
149. ling for this rendering operation do trace s main loop executes For each scan line in the final image it calls trace jit line and outline The former handles the ray tracing computations through many subordinate routines and outline writes the completed line to a file This sequential program already divides its work into natural independent units each scan line of the final image The parallel version will compute many separate scan lines in parallel We ll look at how Rayshade was restructured to run in parallel in some detail To begin with a new real main routine was created This is not always necessary sometimes it is best just to rename the existing main to real main In this case main was renamed rayshade main and real main calls it This was done because the original main routine needs to be called by each worker process as we ll see Here is real main real main argc argv for i Or 1 arge cb strepy args i afravlil cut comm args args iargc return rayshade_main argc argv 0 real_main s tasks are vety simple place a copy of the command line arguments into tuple space accomplished by copying them to a local array which is then used by the out operation and then call rayshade_main with its original arguments and one additional new one Here is rayshade_main the additions made for the C Linda version are in boldface rayshade_main argc argv worker Setup parse options a
150. llel programming techniques while others discuss the example programs in greater detail or from different perspectives About the Example Programs 2 Introduction The chapters in this guide that describe using Linda include many code samples and fragments as examples All of this code is derived from real programs but in most cases has been shortened and simplified usually to make it fit into the allowed space Typically declarations and preprocessor directives are omitted except when they are vital to understanding the program Also sections of code that are not relevant to the point being made are replaced often by a one line summary of their function set in italics Blank lines without initial comment indicator have been inserted into Fortran programs for readability Therefore although the code examples are derived from real programs they do not generally constitute working code Many examples are provided in both C Typographic Conventions and Fortran versions differences between the two languages are highlighted in the text where appropriate For examples of working code see the chapter Sample Programs on page 127 Typographic Conventions Fixed width type is used for all code examples whether set off in their own paragraphs or included within the main text For example variable names and filenames refetred to within the text are set in fixed width type Boldface fixed width type is used in examples to indicate te
151. llo world from number running on cervantes Hello world from number running on sappho Hello world from number running on blake Hello world from number running on virgil Hello world from number running on leopardi oor JJ BS B Hello world from number running on goethe hello world is finished The ntsnet command initiates a Linda program on a network ntsnet and its configuration and options are described in detail in Using TCP Linda Linda Operations rd This section describes the Linda operations we looked at in the previous chapter in more detail including some simple examples Additional examples are given in the next section Tuples and Tuple Matching The in operation attempts to remove a tuple from tuple space by searching for a data tuple which matches the template specified as its argument If no matching tuple is found then the operation blocks and the process executing the in suspends until one becomes available If there are one or more matching tuples then one of them is chosen arbitrarily The matching tuple is removed from tuple space and each formal in the template is set to the value of its corresponding field in the tuple For example this in operation removes a tuple having the string coord as its first field and two other fields of the same type as the variables x and y from tuple space it also assigns the values in the tuple s second and third fields to x and y res
152. lly load balancing Workers continually execute tasks as long as any of them remain If one worker runs on a faster processor than some others it will finish each task more quickly and do proportionately more of them Of course there are times when reality isn t quite this simple so we ll look at some techniques to ensure good load balancing in Chapter 3 The Linda Model Linda or more precisely the Linda model is a general model of parallel computing based on distributed data structures although as we ve noted before it can be used to implement message passing as well Linda calls the shared data space tuple space C Linda is an implementation of the Linda model using the C programming language and Fortran Linda is an implementation of the Linda model using the Fortran programming language Processes access tuple space via a small number of operations that Linda provides For example parallel programs using C Linda are written in C and incorporate these operations as necessary to access tuple space In this way C Linda functions as a coordination language providing the tools and environment necessary to combine distinct processes into a complete parallel program The parallel operations in C Linda are orthogonal to C providing complementary capabilities necessary to parallel programs C Linda programs make full use of standard C for computation and other non parallel tasks C Linda enables these sequential operations to be divided amo
153. ltiplication procedure is well known multiplying an L by M matrix A by an M by N matrix B yields an L by N matrix C where C is the dot product of the 7 row of A and the column of B We won t bother translating this procedure into a sequential C program Instead let s look at a simple straightforward parallel matrix multiplication Here is the master routine real main read mats rows columns L M N for i 0 i lt NWORKERS i eval worker worker L M N tor i0 i lt Le i Linda User Guide 77 78 Case Studies out row i rows il M out columns columns M N out task z for i0 i lt bz i in result i result i len The function begins by reading in the matrices we re ignoring the details of how this happens Next it starts the worker processes places each row of A and the entire B matrix into tuple space and then creates the task tuple Finally it collects the rows of C as they are completed by the workers The worker process begins by rding the B matrix from tuple space After doing so it entets an infinite loop from which it will exit only when all tasks are completed Each task consists of computing an entire row of C as specified by the task tuple The worker process retrieves this tuple increments its value and places it back into tuple space checking to make sure it 1s not already greater than the maximum number of rows to be computed wor
154. me and user defined Linda User Guide 119 resources The default is nodef ile plus the local node name The key word nodefile refers to the nodefile resource value which is a file containing a list of node names User defined resources provides a way to specify a list of node names symbolically The user defined resource must be preceded with the indirection symbol The maximum number of indirections is 16 redirect Specifies whether or not tuple redirection optimization is used The default is true rexecdir Specifies the directory on a remote node where the TCP Linda executable resides Or if distributing it also specifies the directory on the remote node where the Linda executable shall be distributed to prior to execution This resource is node and application specific The default is the key word Parallel The Parallel keyword indicates that ntsnet should use the map file to translate the name of the local executable directory for that remote node rworkdir Specifies the remote node s working directory This resource is node and application specific The default is the key word Parallel The Parallel keyword indicates that ntsnet should use the map file to translate the name of the local working directory for that remote node speedfactor Specifies the relative aggregate CPU capability of a particular node The larger the relative speedfactor the more capable that particular node is of running multiple workers This resource is no
155. ment arrays DX DY and DZ 20 Continue The first out operation places the position arrays x y and z into tuple space in the wake tuple later the workers will each rd it Then the master process creates the index tuple from which the workers will generate tasks In the first for loop the master process next starts NWORKERS worker processes At this point the master process has completed all necessary setup work so it becomes a worker itself by calling the same worker function used in the eval operations This is a common technique when few startup activities are required and worker processes run for a long time If the master did not become a worker then its process would remain idle until the workers finished After the workers finish the master process executes the final for loop which gathers the results produced by all the workers removing them from tuple space and placing them in the locations expected by the original Fortran program Here is a simplified version of the worker routine Linda User Guide 39 C Version worker rd wake x y z nblades nfilmnt nsegmnt while 1 loop until work is done in index index get current task index out index index 1 increment and put back if index gt nblades nfilmnt test if done break iblade index nfilmnt blade ifil index nfilmnt filament for iseg 0 iseg lt n segment Fortr
156. mics Simulation on a Network of Workstations Using a Machine Independent Parallel Programming Language SCIENTIFIC Technical Report INZ Discusses the Molecular Dynamics code used as an example in Chapter 3 of this manual Mike Loukides UNIX for FORTRAN Programmers Sebastopol CA O Reilly amp Associates 1990 Chapter 5 discusses the use of UNIX debugging utilities including dbx Chapter 8 provides a good overview of standard UNIX profiling utilities Tim O Reilly et al X Window System User s Manual Volume 3 of The X Window System series Sebastopol CA O Reilly amp Associates 1988 1992 Chapter 10 of the standard edition and Chapter 9 of the Motif edition discuss the theory philosophy of X resources which has been used in the design of the Network Linda configuration files Linda User Guide B i B ii Bibliography Symbols notation 29 notation 12 A actual 12 adjustability 81 aggregates viewing via Tuplescope 98 alternate operation names 26 linda eval 26 linda in 26 linda inp 26 linda out 26 linda rd 26 linda rdp 26 anonymous formal 36 anti tuple 12 application specifying to ntsnet 48 application specific configuration file 45 array copying 29 fixed length 36 length 28 arrays 28 36 colon notation 29 35 Fortran 90 syntax 33 Fortran vs C 41 multidimensional 32 of structures C 34 available resource 65 116 B bcast resource 117 bcastcache resource 117 blank common 33 blocki
157. mum and maximum times to wait for nodes to join the execution group If maxwait is omitted it is set to the same value as minwait Both default to 30 seconds Execution will commence once the execution group is set based on the values of the minwait maxwait minworkers and maxworkers resources see the discussion of these resources in the next section for details This option specifies the load that a worker will put on a node The value specified can be any real number gt 0 The default is 1 Typically 1 or some smaller fraction is used A larger value could be used to increase the chances of having one Linda process running on each node workerwait seconds This option specifies the time in seconds that a worker waits for a response to its join message from the master The default is 90 If a worker does not get a response within the specified time telling the Linda User Guide 115 worker that it s joined the worker will exit and therefore not participate in the application execution ntsnet Configuration File Format This section serves as a reference for the format of both the user local ntsnet configuration file tsnet config and the global ntsnet configuration file 1ib tsnet config relative to the Linda tree When setting Boolean resources in the configuration files values can be specified as true ot false yes or no on or off or as 1 or 0 Resource Definition Syntax Resources program appl no
158. n a network Linda User Guide 7 Process i 1 send Bj 1 Message Passing Matrix Multiplication receive Process i receive Bj send Process i 1 send B j 1 receive In message passing parallel programs there is no global data All data is always held by some process and must be explicitly sent to other processes that need it Here each process holds one row of A and computes one element of C while it has the corresponding column of B The columns of B are passed from process to process to complete the computation Distributed Data Structures Distributed data structure programs are a second approach to parallel programming This method decouples the data required for the calculation and the distinct simultaneously executing processes which each perform a part of it making them autonomous Distributed data structure programs use a shared data space with the individual processes reading data from it and placing results into it The data structure is distributed in the sense that different parts of it can reside in different processors but it looks like one single global memory space to the component processes In this sense a distributed data structure might be termed virtual shared memory VSM Although Linda can be used to implement both types of parallel programs the most natural one for u
159. ndom void double dboard int darts double x coord x coordinate between 1 and 1 y coord y coordinate between 1 and 1 pi pi pr random number between 0 and 1 int score number of darts that hit circle Linda User Guide 135 n unsigned long cconst used to convert integer random number between 0 and 2 31 to double random number between 0 and 1 cconst 2 lt lt 31 1 score 0 KOR KK kok k oko k oko ko k kok k kok kok ok RARA k oko k kok RR KARA RRA RAR kok K kok I I I kkk k tt 5 F DI Throw darts at board Done by generating random numbers between 0 and 1 and converting them to values for x and y coordinates and then testing to see if they land in the circle If so score is incremented After throwing the specified number of darts pi is calculated The computed value of pi is returned as the value of this function dboard Note the seed value for rand is set in pi_calc kok kok k A kCkCk kCkCK Ck kCk Ck kCKCk kCkCk kCkCk Ck kCK Ck kCkCk KCkCk kCkCk kCk Ck Ck kc k Ck kc k ck kck ck kck I I I ke e kx f for n 1 n lt darts n generate random numbers for x and y coordinates r double random cconst x coord 2 0 om 1 03 r double random cconst y coord 2 0 Y 1 05 if dart lands in circle increment score if sqr x coord ar y coord lt 1 0 score calculate pi pi
160. ne left and right neighbors if id nwrkers right ds else right id 1 if id 1 left nwrkers else left id 1 Update the line update left right tpoints nsteps npoints first id Output the results Concurrent Wave Equation npoints the first value 0 was only used for passing data to neigbors so reset array to send back only values we are interested in ret val values l out results id first ret val npoints fe Initialize points on line des int n t Line int tpoints int points int first local variables int X Je Ki double x fac Calculate initial values based on sine curve fac 2 0 PI k first 1 for j l J lt npoints j ktt 4 x double k double tpoints 1 values j sin fac x for i l i lt npoints i oldval i values i Linda User Guide 147 Update all values along line a specified number of times d int update int left int right int tpoints int nsteps int npoints ine first int Ta d local variables double dtime c dx tau sqtau int iteration 1 j Update values along the wave for nstep time steps dtime 0 3 o 1 0 dx 1 0 tau le dtime Y dx Sqtau tau tau Update values for each point along string for i l i lt nsteps i iteration i Exchange data wi
161. ned the execution group When the minwait interval has elapsed if at least minworkers have joined the execution group execution will start Otherwise ntsnet will continue trying to create workers until the maxwait interval has elapsed which includes the time already spent in minwait As soon as minworkers workers have started execution will immediately commence Once maxwait seconds have passed and the execution group is still smaller than minworkers the startup process will fail and execution will not proceed The default value for both maxwait and minwait is 30 seconds The values for maxwait and minwait may also be specified using the wait command line option Its syntax is wait minwait maxwait For example w 30 60 would set minwait to 30 seconds and maxwait to a total of 60 seconds If wait is given only a single value as its parameter then that value is used for both resources Similarly the values for maxworkers and minworkers may also be specified using the n command line option Its syntax is n minworkers maxworkers ntsnet Worker Process Scheduling For example n 2 4 would set minworkers to 2 and maxworkers to 4 If n is given only a single value as its parameter then that value is used for both maxworkers and minworkers Once ntsnet has attempted to start a process on a node it waits for a message from the worker indicating that it has joined the execution group Only after the master r
162. ng 13 25 C cds 95 character strings C 35 cl file extension 19 cle command 19 c option 110 D option 110 g option 110 help option 110 I option 110 L option 110 loption 110 linda compile args option 110 linda info option 110 linda link args option 110 linda t scope option 110 linda tuple scope option 95 110 o option 20 110 passing switches to native compiler 110 specifying executable 110 syntax 107 v option 110 w option 110 cleanup resource 59 117 C Linda operations 11 Code Development System 95 colon notation 29 35 common blocks 33 compiling 19 110 composite index 41 computational fluid dynamics 38 configuration files disabling global 67 formats 116 121 map translation 53 ntsnet 45 Index D data tuple 11 database searching 82 dbx Tuplescope and 100 debug resource 69 112 117 119 121 debugger use with Tuplescope 100 DEBUGGER environment variable 122 debugger resource 69 112 117 delay resource 67 117 directory Linda 46 discarding tuples 36 distribute resource 59 60 117 distributed data structures 8 benefits 10 load balancing in 10 distributed master 89 E environment variables 51 95 122 eval 11 22 105 compared to out 22 completion 12 creating processes 23 creating workers 11 data types in 24 field evaluation 23 forcing to specific node 68 function restrictions 24 retrieving resulting data tuples 19 eval server 62 106 examples comments in 73 counter tuple 1
163. ng itself a recommended Linda 18 Using the Linda Operations Quick Start Hello world programming practice Each in operation removes one done tuple from tuple space and it will block if none is currently present and wait until some worker finishes and sends its tuple there This same effect could have been achieved by means of a counter tuple which each worker process increments as it finishes In this case real main would create and initialize the counter C Version Fortran Version out counter 0 out counter 0 and each worker would update it as its last action C Version Fortran Version in oco unter 215 in counter J out counter j l out counter J l These statements remove the counter from tuple space assign the current value of its second field the number of processes that have finished so far to the variable j and then increment it and place the tuple back into tuple space Note that only one process can access the counter tuple at a time and so some processes may have to wait for others to finish before they can terminate In this case the waiting time is minuscule so for this program the concern is a non issue However in general it is best to avoid building unnecessary dependencies among processes into a program With a counter tuple the final loop in real_main would be replaced by the statement C Version Fortran Version in counter nworker in counter NProc real_main
164. ng the available processors Since C Linda is implemented as a precompiler C Linda programs are essentially independent of the particular native C compiler used for final compilation and linking Fortran Linda operates in an analogous manner Linda programmers don t need to worry about how tuple space is set up where it is physically located or how data moves between it and running processes all of this is managed by the Linda software system Because of this Linda is logically independent of system architecture and Linda programs are portable across different architectures whether they re shared memory computers distributed memory computers or networks of workstations Data moves to and from tuple space as tuples Tuples are the data structures of tuple space A tuple is a sequence of up to 16 typed fields it is represented by a comma separated list of items enclosed in parentheses Here is a simple example C Form Fortran Form Simple 1 simple 1 Tt Pronounced two pull with the emphasis on the first syllable 10 Overview of Parallel Programming and the Linda Model The Linda Model This tuple has two fields the first is a character string and the second is an integer and in this case both of them contain literal values Variables may also be used in tuples C Form Fortran Form easy i easy i This tuple also has a string as its first field and a second field of whatever type the vatiable i is
165. nore IO is also often useful when debugging TCP Linda programs This is not necessary for gdb Debugging TCP Linda Programs Running TCP Linda Programs Without ntsnet Master session Worker sessions TCP Linda programs can also be executed manually by invoking ntsnet with the launchByHand option These are the steps for doing so hd Establish a session on each desired remote node via rlogin for example It will be most convenient to start each discrete session in a separate window In general start as many processes as you would when executing the program with ntsnet If you plan to use a debugger start it on the desired remote nodes Run ntsnet launchByHand in a separate session In the session where you will start the master process make sure that the LINDA_PATH environment variable is set correctly It should point to the top level Linda installation directory and the directory specification must include a final slash Begin program execution using the commands output by ntsnet cut and paste will reduce errors one per remote session These have the general form application appl arguments LARGS linda args node name application LARGS linda args where application is the program command app arguments are the application s arguments if needed linda args are any TCP Linda run time kernel options and node name is the name of a remote node Keep in mind that each process is handling a portion o
166. ntsnet Command The ntsnet Command Suppress warning messages Suppress warning messages about text beyond column 72 text is still ignored flc only ntsnet options executable arguments Syntax Parameters options executable arguments One or more command line options listed below Command line options override configuration file settings Executable file to execute Arguments to the executable program Options Syntax Convention When setting boolean resources on the command line ntsnet uses the convention that an option name preceded by a minus sign sets the corresponding resource to true and one preceded by a plus sign sets the corresponding resource to false Command Options appl name bcast bcast bcastcache s cleanup cleanup This option causes ntsnet to use name as the application name for the purposes of querying the configuration file database Normally ntsnet uses the executable name as typed on the ntsnet command line as the application name in the configuration file database This can be useful if several different executables use the same configuration parameters Note that appl has no corresponding resource parameter in the configuration file This option enables the tuple broadcast optimization This option disables the tuple broadcast optimization This is the default This option specifies the size of the broadcast cache in bytes This size is a trade off between memory
167. o tuple space for i 0 i lt L i clump rows M clump columns M clump out 1 rows rows M clump cut i oolumns columns M clump start workers and make first task for i 0 i lt workers i eval worker worker L M N clump out task U for i 0 i lt L i clump result M clump in result matrix i result The master begins by allocating memory for and reading in the matrices Next it places the matrices into tuple space in chunks each of which is clump rows or columns of its matrix clump functions as a granularity knob in this program a parameter whose value determines the task size at least in part Changing clump s value directly affects how large each chunk is The master then starts the workers and creates the task tuple as before Its final loop retrieves the result matrix C from tuple space again in chunks of clump rows Code to reassemble C from these chunks is omitted Both versions of this program assume that A is stored by row in the rows array and B is stored by column in the columns array Such a strategy makes accessing the proper elements of each one much easier Linda User Guide 79 80 Case Studies Here is the corresponding worker function worker L M N clump Allocate memory for matrix chunks while 1 in task rows index if rows_index lt L out task rows index clump else cGut task
168. o use ones with the extension i586 on others Using this mechanism it s even possible to support accidental homogeneity in which different machine architectures support the same data layout but obviously great care must be taken here Linda User Guide 59 60 Using TCP Linda These suffixes are used when the suffix resource application specific is true the default value Which suffix to use for a given node is specified by the suffixstring resource application and node specific The default suffixstring value is the null string so even though suffixes are used by default this fact has no effect until some specific suffixes are defined using suffixstring Here is a section of a ntsnet configuration file illustrating a common use of these features Tsnet Appl Node rexecdir parallel Tsnet test24 nodelist chaucer moliere sappho Tsnet test24 suffix true Tsnet Appl chaucer suffixstring athalon Tsnet Appl moliere suffixstring pIII Tsnet Appl sappho suffixstring xeon Tsnet test24 sappho suffixstring xOptBlasLib Tsnet Appl aurora suffixstring duror These entries would result in the following executables being used for these nodes when running the application test24 located in whatever directory resulted from map translation chaucer test24 athalon moliere test24 pIII sappho test24 xOptBlasLib aurora local test24 duron If the distribute resource for the application test 24 is t
169. of outstanding tasks reaches its minimum allowed value the do loop ends the master begins to make new tasks and the process begins again After all needed tasks have been created the master still needs to gather any remaining results from tuple space which is the purpose of the second while loop UPPER_BOUND and LOWER_BOUND allow this program to adapt to its environment Their values can be adjusted based on the size of tuple space on the sizes of individual database records of the database as a whole on the relative speeds of the get_next_record compare and process functions and so on It can be as important to make sure that there are enough tasks in tuple space as to make sure there aren t too many When there aren t enough tasks to keep all the workers busy then work starvation sets in and performance diminishes Thus if LOWER_BOUND were too low there might be periods where workers would actually have to wait for their next task a condition which is virtually never desirable Load balancing is another consideration that often comes into play This program will perform fine if all of the comparisons take about the same amount of time as would be the case when comparing fingerprints However there are many cases where different comparison operations take vastly different amounts of time comparing DNA sequences for example In such cases the program must ensure that the more time consuming comparisons do not become
170. of all of these factors on the image also taking into account reflections and shadows This case study illustrates the following techniques Dividing the main task among the workers Adapting the sequential program structure to the master worker paradigm Rayshade s main routine is shown below As mentioned in the Introduction we ve removed some sections of code and replaced them with descriptive comments and we ve ignored others like declarations altogether main argc argv Setup parse_options argc argv read_input_file Initialization startpic start new picture More setup raytrace After some initial setup Rayshade processes the command line options validating them for correctness and determining which options have been selected It then reads in the image and then performs some additional initialization steps Linda User Guide 73 74 Case Studies Next Rayshade calls the function startpic which logically creates a new picture folowed by some additional setup activities Finally it calls rayt race which initiates the actual rendering The bulk of the work is handled by the routine do trace called by raytrace do trace called by raytrace Set sampling parameters Trace each scanline writing results to output file for y StartLine y gt 0 y trace jit line y out buf outline out buf After testing and setting parameters controlling the samp
171. of this code differ because of the way define NRA 62 number of rows in matrix A define NCA 15 number of columns in matrix A define NCB 7 number of columns in matrix B main 1 int iz jp kit mise 7 double a NRA NCA matrix A to be multiplied b NCA NCB matrix B to be multiplied C NRA NCB result matrix C Initialize A B and C matrices for i150 i lt NRA i for j 0 j NCA j eLxpp3le itj for i0 i NCA i for j 0 j lt NCB j bli lil 3 3 for i 0 i lt NRA i for j 0 j lt NCB j celta 0 0 Perform matrix multiply for i 0 i lt NRA i for j 0 j lt NCB j for k 0 k lt NCA k cltlldits alil k SIRI TI Okay it s a trivial program printf Here is the result matrix n for i 0 i lt NRA i printe ya for j 0 j lt NCB j printf 56 25 Te od jd qs Linda User Guide 137 printi ass HEE HEHE EEE EEE EEE EEE EEE EEE EEE EE HEHEHE EE HEE HEE HEE HE HE HE EE E E HEE HE E E E HEE E E E HH HE E Serial Matrix Multiply Makefile FILE make ser_mm c DESCRIPTION See ser mm c USE make f make ser mm c k k k k k AE aE aE A aE AE HE aE aE aE aE Ha a AE AE FE aE aaa aE a a aE E aE HE aaa AE aa aaa Ea aE E E ec mm ser mm c CC ser mm c
172. of this wildcard character is supported Here is an example entry using the ampersand wildcatd mapto net amp sappho blake t 73 These entries map the root directory to net hostname for the nodes sappho and blake Thus the local directory tests new on sappho would be mapped to the generic directory net sappho tests new by this entry The ampersand character can also be used within the actual translation specifications map working home test work 8 How ntsnet Finds Executables This example equivalences the directories home test work hostname for all of the nodes within the network Here is a more complicated example map net amp O fp This entry maps the local root directory on all nodes to the generic directory net hostname where hostname is replaced by the name of the local node It also translates directories of that form back to their local equivalents when performing translation on remote directories preventing unnecessaty automounting When wildcarded entries produce multiple matches for a given translation the longest matching string is used When there are two matching strings of equal length then the more specific match 1 e containing fewer wildcards is chosen Distributing Executables ntsnet can distribute executable files prior to program execution This is controlled by setting the distribute resource to true application specific When different executables are requir
173. of workers between 1 and d n MAXWORKERS scanf d amp numworkers 138 Sample Programs Matrix Multiply BRR KKK kok k kok k kok k Koko ke ke ke ke KARA master Ltask KKK KK KR KK KK ke ke ke e RARAS ye Initialize A and B for i 0 i lt NRA i for 3 0 j lt NCA j alili si ji for i 0 i lt NCA i for 3 0 j lt NCB j Bl ji si jz put out the arrays for the tasks to read out arrav b bi out array a al Put out work for them to do ten columns at a time work count 0 nbr rows 10 offset 0 extra NRA nbr rows if extra gt 0 1 out offset offset out array data offset extra offset offset extra work count s else out offset offset while offset lt NRA out array_data offset nbr_rows offset offset nbr_rows work count Start up worker tasks id 1 while id lt numworkers printf Starting task td Xn id eval worker worker id t Receive results from worker tasks Linda User Guide 139 for k 1 k lt work_count k in results offset nbr_rows c put results into results array for i offset i lt nbr_rowstoffset i for j 0 j NCB j 1 results i j clill j Print results for i 0 i lt NRA i for j 0 3 lt NCB j 1 printf 2f results r 3l3 printr An KR KKK K
174. ok kok ok kok k ok k oko k oko oko k oko k kok k kok Koko k kok k kok ck kokck kck ck Koko kkk subroutine prtdat XKOKCKCKCKCKCkCK Ck kCkCk kCkCk kCkCK Ck kCK Ck kc k k kCKCk kCkCk kCkCk kok k Ck kCKCk kok k kc k k kc k k kc k Ck kc k ck kok ck kok ck ckck ck ckok sk ke kk void prtdat int nx int ny float ul char fnam int ix iy FILE fp fp fopen fnam w for iy ny 1 iy gt 0 iy 4 for ix r ix lt nx l ixit 4 fprintf fp 8 3t ultix nytiy if ix nx 1 fprinbrt tg T ya else fprintfi fg Aevi fclose fp HEHEHE EEE EEE EEE EEE EEE EEE EEE EEE TEEGATE HEE HEE HEE HE HE HE EE ERE E ERE E HE Serial heat2D Makefile FI DESCRIP USI mak T ES oF TION p make ser heat2D c See ser heat2D c f make s P r heat2D c EAE AE AE AE AE TEE aE AE AE AE FE EE ER HAA AAT AE k aT aE FE AE AE AE AE aE aE HEE aE AE E aE aE E AE AE aE a aE AE AE aE aE a aE aE a aE Ea aE u ec heat2D ser_heat2D c S CC ser_heat2D c Parallel Version o heat2D P e ok k kok k oko k oko k kok kk kkk k kkk k k k k k k k k k k kk k k k k kk k k kCkCk Ck kc k Ck kCk Ck kCk ck k k kkk k kkk k k k k k be high in the midd boundaries are held time stepping LINDA FILE clinda heat2d OTHER FILES
175. ompiler to use for compiling files defaults to f77 in most cases and to xIf under AIX LINDA FORTRAN LINK LINDA PATH 122 Linda Usage and Syntax Summary Used by the linda fortran link shell script specifies the command to use for linking the executable same defaults as for LINDA FORTRAN Specifies the path to the Linda installation directory The directory specification must contain a terminal slash LINDA PROCS POSTOPP QC Tuplescope Reference Used by the Linda Code Development System as the return value for the Iprocs and flprocs support functions under CDS Iprocs is not truly meaningful and is provided only for compatibility with TCP Linda Used by the postcpp cc shell script specifies the C compiler to use for compiling c1 files defaults to cc POSTFL FORTRAN Used by the postfl fortran shell script specifies the Fortran TSNET PATH Tuplescope Reference Menu Buttons Modes Aggregates Run Break Continue Debug Save Quit The Modes Menu Single Step compiler to use for compiling f files generated by Fortran Linda from 1 source files same defaults as for LINDA FORTRAN Used by ntsnet specifies its search path for local executables Its value is a colon separated list of directories Set debugging modes on or off Specify format for aggregates displays active when Display Aggregates is on Available formats are Long Short Float Double Character and Hexadecimal
176. ons the following steps are necessary to create and run a TCP Linda program Make sure that the bin subdirectory of the Linda distribution tree is in the search path Define the set of hosts to be used for program execution by creating the file tsnet config in your home directory containing a single line like this one Tsnet Appl nodelist moliere sappho blake shelley Replace the sample node names with the appropriate ones for your network T TCP Linda is also available for certain distributed memory parallel computers However the discussion here centers on networks although identical considerations apply in both cases This discussion applies only to TCP Linda version 2 4 7 or higher Linda User Guide 43 What ntsnet Does 44 Using TCP Linda Compile the program using a command like the following Cle 6 hello world hello world sel See the Quick Start section of Chapter 2 for a more detailed discussion of this and the following step Finally run the program preceding the normal invocation command and arguments with ntsnet ntsnet hello world 4 ntsnet will automatically run the program on the defined set of nodes ntsnet is responsible for the following tasks 9 Parsing the command line options and configuration file entries Querying remote systems for load averages Locating local executable files Determining what set of nodes to run on based on its scheduling algorithm
177. ose minimum medium or maximum window size Mouse Button Effect Left Make the window minimum size Middle Make the window medium size Right Make the window maximum size A textual representation of the tuple class positioned to the right of the sizing box This shows the tuple class s structure Here is an example gl bals INT INT Clicking with the left mouse button on the tuple representation string will close the tuple class window Linda User Guide 97 Spherical icons for each tuple that exists in this class Left clicking on a tuple opens a small window that displays its contents For example globals 1000 225 If a tuple contains a large amount of data scroll bars will appear and you can scroll through it viewing one part of it at a time You can also scroll with the keyboard arrow keys Key Scrolling Effect Up arrow Scroll to previous line Down atrow Scroll to next line Ctrl up arrow Scroll to previous page Ctrl down arrow Scroll to next page Left arrow Move to the beginning of the tuple Right arrow Move to the end of the tuple If the tuple contains an aggregate by default the aggregates s contents is not shown instead the word BLOCK appears The Modes and Aggregates menus control the display of aggregates see Viewing Aggregates below Clicking on an individual tuple window with the right mouse button will refresh its contents clicking on it with the left mouse button will close it I
178. other provisions have been made using the current directory as the default directory in each case The p option is no longer necessary in such cases Permissions and Security Issues ntsnet assumes that all remote systems and directories are accessible By default it uses the local username for running remote processes The user resource node specific can be used to specify an alternate username on a remote system For example the following configuration file entry tells ntsnet to use the username guest when running process on node sappbo Tsnet sappho user guest There is no provision for specifying passwords for use on remote systems The standard TCP IP mechanisms the etc hosts equiv file and individual user rhosts files should be used to assure access ntsnet Worker Process Scheduling The TCP Linda System provides considerable control over how processes are scheduled on temote nodes ntsnet uses the node list resources we ve already looked at to determine the list of nodes where the application may potentially run Whether a node is actually used depends on a number of other factors which we ll examine in turn Forming The Execution Group The list of potential nodes on which an application may run is determined by the nodelist resource the set of nodes on which an application actually runs is called the execution group ntsnet begins with the node set and successively starts remote processes using the scheduling rule
179. ou are using TCP Linda Chapter 3 Using TCP Linda describes the special features and considerations of the Linda implementation for networks of UNIX workstations The first section provides an introduction and quick start for new users of TCP Linda The remainder of the chapter describes the general features of TCP Linda To see examples of using Linda with the C and Fortran programming languages Chapter 4 Case Studies presents several extended program examples illustrating the process of transforming a sequential program into a parallel program with Linda Chapter 5 Using Tuplescope describes the Tuplescope visualization and debugging tool and the Linda Code Development System CDS and how to use them for high level debugging of Linda programs running on a single node Chapter 6 Linda Usage and Syntax Summary provides a programmer s reference to the features of the Linda programming environment including both C Linda and Fortran Linda language constructs and the elements of the Linda Toolkit Chapter 7 Sample Programs contains five sets of example programs that demonstrate how to use Linda to parallelize serial programs Each example includes the serial version of a program written in C and a parallel version created with Linda The examples are presented in order of increasing complexity The Bibliography lists books and articles that may be of interest to Linda users Some items provide more advanced treatment of para
180. ource 52 120 parallel value 53 58 S sample programs 127 152 2D heat eguation 150 array assignment 127 concurrent wave equation 142 matrix multiply 137 pi calculation 132 scalars 28 secutity 61 shape of arrays 32 shell scripts location 118 speedfactor resource 65 120 start timer 106 status messages 67 stride 33 strings 35 structures arrays of 34 C 33 varying length 35 suffix resource 60 120 suffixstring resource 60 120 support functions 106 Syntax formal C Linda 105 syntax requirements 18 T task adjusting size of 80 compared to worker 9 TCP Linda 43 68 appropriate granularity for 68 file permission requirements 61 process scheduling 61 running programs 21 searching for executables 51 secutity 61 shell scripts 118 specifying the working directory 61 tuple space size 22 TDL 101 template 12 matching rules 26 37 termination 37 threshold resource 65 66 120 throwing away data 36 timer split 106 timing functions 106 translate resource 57 120 tsnet command 50 61 tsnet config file 46 tsnet config application_name file 45 tsnet map file 46 53 TSNET_PATH environment variable 46 51 123 tuple allowed field types 26 arrays in 28 36 arrays of structures in 34 character strings in 35 classes 96 colon notation 29 common blocks in 33 counter 19 data 11 data types in evals 24 definition 10 discarding 36 formal to actual assignment 12 Fortran 90 array syntax 33 ignoring fields in 36 length of
181. out varying struct s5s bytes sizeof STYPE sizeof int buf len l current structure length out varying struct sibytes The first out operation creates a tuple with a varying structure as its second field The second out operation creates a tuple whose second field is also a varying structure for this instance the current length is set to the size in bytes of STYPE including the first element of array buf plus the size in bytes of the remainder of array buf the product of the size of one element and its number of elements minus 1 C Character Strings In keeping with C usage character strings are simply arrays of characters and the normal array considerations and matching rules apply The only exception occurs when comparing string constants with character arrays The length of a string constant is the number of characters in it plus one for the null terminator Thus the string hello has a length of 6 and a five element character array will not match it it requites a six element array char s 6 int len out string hello length 6 in string 38 Linda User Guide 35 The array colon notation may also be used with strings for fields where the length of the string is variable out sterimg Litas in string ra len 7 Note that the literal string in the out operation needed to include the colon in order for the template in the in operation to match since th
182. oving it from existing node sets true node specific workerload The third term of the numerator adjusts the load average for the number of workers the node already has running even though they are not yet consuming substantial CPU resources You can alter this value from its default of 1 to reflect your estimate of how much load a single worker process adds Linda User Guide 65 66 Using TCP Linda speedfactor The speedfactor resource s value is used to normalize the corrected load averages for differing CPU speeds and hence different total capacities on different nodes Generally a value of 1 is given to the slowest system type within the network Faster systems will then be assigned correspondingly higher speedfactor values Multiprocessor systems ate also given higher speedfactor values generally set to the value appropriate for a single CPU multiplied by the number of processors Once all the adjusted loads are calculated ntsnet finds the node having the lowest value which presumably will be the most lightly loaded when execution begins It then checks its adjusted load against the value of its threshold resource If the adjusted load doesn t exceed it ntsnet next checks the number of processes already started If this number is less than the value of the maxprocspernode resource another process is started otherwise ntsnet continues on to the next least heavily loaded node This scheduling proce
183. ow ntsnet Finds Executables local generic remote home chavez work mabig a mapfrom flaubert u home chavez work aurora home chavez work gt blake home chavez work m gt chaucer u chavez work m degas tmp erasmus mf chavez work I flaubert u home chavez work gt gogol net aurora home chavez work This particular rule facilitates generic to remote directory translation enabling remote processes to be started on degas from any other node using tmp as the working directory Note that the generic directory we ve specified is the one to which her home directory will be translated from any other node This rule will work only for user chavez other users would need to construct their own versions This rule will not allow chavez to run ntsnet on degas however since it uses mapfrom If it were necessaty to be able to use degas as the local node then the rule should be written as a map The preceding diagram illustrates the map translation process using all three rules for our sample ntsnet command executed on flauber We ve noted that generic directories need not exist as real directory locations Here is a map file entry which takes advantage of this fact map special chaucer u guest linda blake home guest1 app1 linda sappho usr guests linda bin aurora public bin zola fuses local bin special is not a real directory on any node this entry defines the s
184. pecified directories on the nodes listed as equivalent Map translation can be disabled by setting the translate resource application specific to false it is true by default or by using the corresponding command line options translate translate Remember also that map translation is used only for determining executable file locations on nodes for which the setting of rexecdir is parallel Linda User Guide 57 Similarly the remote working directory is determined by map translation of the local current working directory only for nodes where the rworkdir resource is set to parallel also its default value In particular map translation does nothing to filenames internal to the application Map Translation Entry Wildcards 58 Using TCP Linda There are two wildcard characters allowed in map translation file entries An asterisk can be used as a wildcard in local and remote node names An ampersand amp can be used to substitute the current node name within a translation For example the following mapto entry handles the most common version of automounted directories mapto net gt tmp mnt net It specifies the same remote directory for every remote node for the generic directory net The asterisk wildcard character can be used as above to represent the full node name It can also be used as the initial component of a node name to specify wildcarded subdomains medusa com No other placement
185. pectively C Version Fortran Version in coord 2x y inm eoord 2x Py If no matching tuple exists in tuple space then the operation will block The following tuple searches for the same sort of tuple but specifies that the value in its second field must match the current value of x C Version Fortran Version ini ocoor xe Ty in coord x Ty The rd operation functions identically to in except that it does not remove the matching tuple from tuple space For example the following rd operation will attempt to match the same kind of tuple as in the examples with in except that this time the value in the second field must be 3 Linda User Guide 21 C Version Fortran Version rd ooord 3 Tyl rdi eoord 3 Ry When the rd operation successfully matches a tuple the value in its third field will be assigned to the variable y The tuple will remain in tuple space available for use by other processes out The out operation adds a tuple to tuple space Prior to adding it out evaluates all of its fields resolving them to actual values out returns after the tuple has been added to tuple space For example the following out operation places a coord tuple into tuple space C Version Fortran Version out eoord 3 10 ceuta ecard 3 105 If any of the fields in an out operation contain expressions they are evaluated before the tuple is placed into tuple space For example this out operation will com
186. ple space to a file named program N dump where program is the application name and N is a integer incremented each successive save operation Linda User Guide 125 126 Linda Usage and Syntax Summary 7 Sample Programs This chapter contains five sets of example programs that demonstrate how to use Linda to parallelize serial programs Each example includes the serial version of a program written in C and a parallel version created with C Linda The examples are presented in order of increasing complexity Array Assignment Serial Version KR k k kok k oko k oko k oko k oko k kok I kok k k ok k oko k Koko RRA RARA k kok RRA RRA kkk kkk kk kok ck kkk ck ck ck k kk Serial Example Array Assignment C Version PILhE array c OTHER FILES make array c DESCRIPTION In this simple example an array is initialized and values assigned In the parallel version the master task initiates numtasks 1 number of worker tasks It then distributes an equal portion of an array to each worker task Each worker task receives its portion of the array and performs a simple value assignment to each of its elements The value assigned to each element is simply that element s index in the array l Each worker task then sends its portion of the array back to the master task As the master receives back each portion of the array selected elements are displayed XKOKCKCKCKCKCKCK Ck kCk Ck kCk Ck kCkCK Ck kCKCk kCk Ck kCKCk kCkCk
187. processes This can be used to filter out unwanted tuples or processes color color Change matching items to the indicated color Co or must be on one of red orange yellow green blue indigo and violet On monochrome displays the colors are mapped to distinct black and white pie shaped icons save Dumps the contents of tuple space to a disk file This is equivalent to clicking on the Save button during program execution Save operations are not legal when the Dynamic Tuple Fetch option is in effect Here are some examples of complete TDL statements if linda_op out then color red if process lt 5 or process gt 7 then hide if linda op out and field 2 0 then break The first statement turns the process icons for processes performing out operations the color red The second statement hides all process icons except those for process numbers 5 and 6 The final statement causes execution to pause whenever a tuple is placed in tuple space whose second field is nonzero Note the syntax of the TDL statements including both the parentheses and square brackets which must surround conditions Linda User Guide 103 104 Using Tuplescope 6 Linda Usage and Syntax Summary Linda Operations in s rd s rdp s amp inp s eval s out s Withdraw a tuple matching s from tuple space If no matching tuple is available execution suspends until one is If more than one matching tuple exists one
188. pute the value of the function with the argument x and then place a tuple with that value in its third field into tuple space the second field will of course contain the current value of x C Version Fortran Version out Tepora x se 4 Cle OOS e RL The evaluation order of the fields is not defined and cannot be relied upon For example in the following C Linda out operation x may or may not be incremented before it is used as the argument to the function f C Version Fortran Version out coerd FE El outst coord oz EXI Similarly in the Fortran Linda version there is no way to determine which routine will be called first should both and g modify x These types of constructions should be avoided No specified action is taken 1f tuple space is full when an out operation attempts to place a tuple there Current Linda implementations will abort execution and print a diagnostic message Typically such events are treated by programmers along the lines of a stack or heap overflow in conventional programming languages the system is rebuilt with a larger tuple space Future Linda versions may raise exception flags Under TCP Linda the size of tuple space is limited only by the total available virtual memory of all participating nodes eval As we saw in the previous chapters an eval operation creates a process tuple consisting of the fields specified as its argument and then returns Here we ll go into more det
189. r an N atom molecule is approximately N N 1 2 and so we want each worker to do about N N 1 2 num workers of them If there are N atoms there are N possible interactions but we throw out pairs where an atom is interacting with itself The factor of 1 2 comes from the fact that the interactions are symmetric A s effect on B is equivalent to B s effect on A and thus only need to be computed once We should also remove bonded pairs but if this number is small compared to the total number of interactions then our formula is a reasonable approximation to the ideal number of interactions per worker In some cases the total work can just be divided as evenly as possible among the wotkers Here we might like to just give each worker the number of interactions we computed above cotrecting the last worker s amount for any remainder and other integer rounding we had to do However the interaction pairs are actually stored as rows in a lower triangular matrix a matrix with all of its non zero elements on or below the diagonal and for maximal efficiency we want to give only complete rows of the matrix to each worker Given this constraint it is necessary to figure out how many rows to give each wotker so that the number of elements each one gets comes out about the same and close to the target number To do so the program uses the fact that there are I non zero elements in row I of a lower triangular matrix and assigns consecutiv
190. r vel s Te temperature str kinetic energy str vel verl_scale coor sqrt T Te for i20 i lt NSteps i verl step str coor vel s Te temperature str kinetic_energy str vel verl_scale coor sqrt T Te exit 0 The key routines here are verl init which performs the setup work and verl step which oversees the calculation at each time step The actual computations are performed by routines called from these functions As we ll see it is in the latter that the major changes for the C Linda version appear Linda User Guide 87 88 Case Studies The following diagram illustrates the structure of the parallel version of the molecular dynamics program The master still does much of the initial setup work but one step summing the special charges is divided among the workers Once the master has gathered and processed the workers results from the setup phase the main calculation loop begins Some parts of it remain with the master in fact the nonbonded interactions so dominate the execution time that it is not worth parallelizing any of the other steps Structure of the Parallel Molecular Dynamics Program eval workers Read data worker from disk setup in task amp globals Y compute charges Y out charges Setup seq setup out globals outtasks in 8 sum changes i Tuple Repeat Calculate Space NSteps bonded Repeat times
191. rent value can be set for it for every application program and node combination For example you can specify Customizing Network Execution a different working directory when running program big job on node zoZere and on node chaucer and you can specify different working directories for programs big job and medjob on node molere In contrast the resource speedfactor is node specific meaning that you can specify a different value for each potential execution node but not for node application program combinations the value for a node applies to all TCP Linda applications that run on it Such resources usually specify intrinsic characteristics of a node which don t depend on the application program being run For example speedfactor specifies how fast a node is relative to other nodes in the network something which is relatively constant across different applications at least in theory Finally the resource maxprocspernode is an application specific resource meaning that its value can be specified separately for different application programs The value set for an individual application is used for whatever nodes it may execute on This resource specifies the maximum number of processes per node that can be run for a given application program the default is 1 Here are some example entries some sample tsnet config file entries ntsnet hello_world moliere rworkdir tmp ntsnet hello_world maxprocspernode 1 n
192. rgc argv Ray Tracing read input file Initialization if worker return startpici Start new picture More setup raytrace The third argument is a flag indicating whether the caller is a worker process or not a value of 1 means it is a worker This flag is used to ensure that the worker exits at the proper time and the remaining initialization steps are executed only at the beginning of the job t A few lines were also added to parse_options to handle additional options specifying the number of workers to use for the parallel version rayshade main still ends by calling rayt race No changes were made in that routine but a lot was done to the support routine do_trace that raytrace calls do_trace becomes the master process in the parallel version of Rayshade It s important to note that the master does not always need to be the top level routine real_main or even the original main function any routine can become the master process Here is do_trace do trace out JitSamples JitSamples out TopRay TopRay for i 0 i lt Workers i eval worker worker out scaninfo StartLine for y StartLine y gt 0 y ini result w Y out bufi outline out buf The new version begins by placing two parameters needed by lower level routines into tuple space The worker processes will read them and pass them along to the routines that need them The function then
193. rue then files of these names will be copied to the remote nodes prior to execution Otherwise they must already exist there in the proper directory The command line options suffix suffix may also be used to specify the value of the suffix resource for the current run ntsnet Worker Process Scheduling Specifying the Working Directory for Each Node The working directory used for program execution on a remote node is also mapped from the current directory on the local node if the rworkdir resource application and node specific is set to parallel for that node Alternatively an explicit working directory can be set for an application node or application node combination by giving an explicit directory as rworkdir s value The p option can be used to specify a remote directory for both rexecdir and rworkdir in a single option It overrides any other method of setting either resource including other command line options Note that the old tsnet command requirement of including p with cwd as its option is no longer necessary If the current directory is NFS mounted on all desired nodes using the same path everywhere then a command like this ntsnet hello world will work properly even assuming no settings have been made via the ntsnet configuration file It will run the executable hello world located in the current directory on all relevant nodes by default the nodes in the file tsnet nodes in the current directory if no
194. s closed tuple window open windows grayed out click to open X tuple space partition Dog X le space partition Que X tuple space partition Que sl globals INT INT ZI task INT gj result INT window sizing box Ex blocked process after m pocese locker A g completing an out Li o 9 9999909 process after procen blocked A 22 s 9 successful in process after successful rd tuples click to view Tuple class windows The Tuplescope display consists of two parts the control panel window and separate windows for each tuple class tuple classes consist of the compiler generated distinct partitionings of tuple space based upon tuple field types and any unique constant strings Note that the display above is deliberately artificial and was constructed to present all of Tuplescope s features in a compact form rather than to represent any real or even possible situation The Control Panel The control panel is a separate Tuplescope window that contains these items The name of the executable appearing on both the window title bar and in a box in the upper left of the display A row of menu buttons along the top of the window labeled Modes through Quit Click a menu button to display its associated menu if any or perform its designated action Aslider control bar in the upper right corner of the window that has a sliding bar that moves along a tick marked area The default position for
195. s described below until a sufficient number of them are running Linda User Guide 61 62 Using TCP Linda Technically the processes started by ntsnet are known as eva servers An eval servet is a process started by ntsnet that waits to execute a Linda process The process on the local node is known as the master in this context and it too eventually becomes an eval server The eval servers on remote nodes and any additional eval setvers on the local node are known as workers Note that this terminology is independent of the master worker concepts within the application program When used in this chapter the terms master and worker refer exclusively to the initiating process on the local node and all other eval servers started by ntsnet respectively How many worker processes are desired is controlled by the maxworkers and minworkers resources application specific The default values are the number of distinct nodes in the nodelist resource not counting the local node for maxworkers and 1 for minworkers The master attempts to start maxworkers workers when execution begins if at least minworkers eventually join the execution group the job proceeds Otherwise execution terminates More specifically execution will begin according to the following rules ntsnet will attempt to start up to maxworkers worker processes Until the time specified by the minwait resource execution will begin immediately whenever maxworkers workers have joi
196. s from each worker for wrker 0 wrker lt nworkers wrker in pr results round nbr pi recv keep running total of pi pisum pisum pi recv end for ea worker loop Calculate the average value of pi for this iteration pi pisum homepi nworkers 1 1 includes master s calc Master calculates the average value of pi over all iterations avepi avepi round nbr pi round nbr 1 prince After 3d throws average value of pi 10 8f n DARTS round_nbr 1 avepi end for each round loop return 0 KOR k k kok k oko k ok ko k kok k kok kok ok RARA RR RR A I A A k kok k I OI I eK kk Start of Worker Task Fe AK RARA RR RR A k oko k kok k Koko Koko k kok k kok kok k K kk kk kk k kk ke ke ke ke e e ke x int worker int id local variabl es double my_pi int round Set seed for srandom id random number generator equal to task ID Calculate pi for round 0 using dartboard algorithm round lt ROUNDS round my_pi dboard DARTS out pi_results round my pi return 0 KKK KK RARA RAR RAR ok ok oko kok ok k oko RR RRA I A k kkk kok ck kok ck kck ck kok ck Koko k kk dboard KOKCKCKCKCkCkCKCk kCkCk kCkCk A kCk Ck kCkCk KCkCk kCk Ck kCK Ck k Ck kCkCk kCk Ck kCk k Ck kCk Ck kc k Ck kck ck kok ck kck ck ckck I KK fdefine sqr x x x long ra
197. s the default 114 Linda Usage and Syntax Summary The ntsnet Command useglobalconfig useglobalmap useglobalmap v v vv vv verbose verbose veryverbose veryverbose This option causes ntsnet not to use the resource definitions in the global configuration file ntsnet will only use the command line options and the user s local configuration file if one exists This is useful if a user does not wish to use the configuration file installed by the system administrator This option causes ntsnet to use the global map translation file The translations in the global map translation file are used in addition to the uset s local map translation file if one exists This is the default This option causes ntsnet not to use the global map translation file ntsnet will only use the user s local map translation file if one exists This is useful if a user does not wish to use the map file installed by the system administrator Synonymous with verbose verbose Synonymous with veryverbose veryverbose This mode displays remote commands being issued and other information useful for debugging configuration and map files This option turns off verbose mode This is the default Turns on very verbose mode which produces the maximum amount of informational status messages This option turns off very verbose mode It is the default wait minwait maxwait workerload oad This option specifies the mini
198. s the value of the LINDA PROCS environment variable or the value 6 if it is not defined The clc and flc Commands Command Syntax cle options source files fle options source files clc expects source files to be of one of the following types c cl o 10 flc expects source files to be of one of the following types 1 0 10 By default these commands produce the executable a out override this name with o The following figure illustrates the compilation and linking process Linda User Guide 107 Command line clc o sundae choc cl straw lo vanilla c shell script calls cpp clc parser executable Compilation executable aka the Linda engine ac _Itxxx o vanilla c l i E Itcp es support lo Itcp es support d0 Linda Runtime Support Itcp linda a i etc a 7 shell script 4 eo linda_cc_link calls cc amp Id i C Linda Compiling and Linking Process Linking 108 Linda Usage and Syntax Summary The clc and flc Commands Gumurj 0Je e sqnjs e epui do uoddns eumuny epur UeJ110J epul yxxxu ueJ140J epuil Fejiuea uonejiduo3 J eTTTUPBA OT Me13s TI 00UD eepuns o OTF eui PUBWIWIOD ss9901g Buryur pue Buijiduo3 epurT1 ue140 4 yu ue0j epul z jeue i 9p02 UeILIO4 ejiduioo o ui vu o xxxe 48 op woddns se day o uoddns se doy ueJjuoJ jpsod a 109 Linda User Guide
199. se with Linda is the distributed data structures using virtual shared memory All interprocess communication is accomplished via this global data space Processes never explicitly send messages to one another but rather place data into the shared data space When another process needs that data it obtains it from the shared data space Linda handles all data transfer operations so the program need not worry about the exact mechanism by which it occurs Linda operations require no additional overhead over any message passing scheme and in fact sometimes are more efficient Programs using the distributed data structure method often use a master worker computation strategy Under this approach the total work to be done by the program is broken into a number of discrete tasks which are stored in the global data space One process known as the master is responsible for generating the tasks and gathering and processing the results Actual program execution involves a number of component processes known as workers Each worker removes a task completes it and then grabs another continuing until some condition is met For example it may encounter a special type of task known as a poison pi telling it to die or it may be terminated by some other mechanism Depending on the situation the master process may also perform task computations in addition to its other duties 8 Overview of Parallel Programming and the Linda Model Approaches to Paral
200. sing its menu item The check mark will then disappear and the other formats will become active You may then select the desired format Exit from this menu by choosing the Exit Aggregates Menu item What format a tuple containing an aggregate uses depends on the Dynamic Tuple Fetch setting on the Modes menu If Dynamic Tuple Fetch is active then an aggregate is displayed in the current Aggregates menu display format when you click on its tuple icon If Dynamic Tuple Fetch is not in effect then the format that was in effect when the tuple entered tuple space will be used regardless of the current setting Viewing Process Information Clicking on a process icon displays a scrollable file window containing the text of the source file corresponding to it Tuplescope requires source files to be in the current working directory The line containing the tuple space operation corresponding to that icon is indicated by a caret Scroll bars or the keyboard scrolling keys can be used to examine the source file To close the window click the left mouse button anywhere within it Tuplescope Run Modes Tuplescope has a number of different modes for program execution First execution speed can be controlled with the slider control on the Tuplescope control panel discussed previously This is independent of the other run controls we ll look at in this section Clicking on the Run button will commence program execution The program will execute in
201. ss continues until maxworkers processes are started or until no qualifying node can be found The goal of the scheduling process is to minimize the load on the maximally loaded node Here are some sample ntsnet configuration file entries showing the uses of these resources Tsnet Appl getload true Tsnet Appl loadperiod 10 use the default speedfactor of 1 for these systems Tsnet Node slowmach priestley pinter fastguy is 5 25X slowmach Tsnet Appl fastguys sand stahl goethe has 2 heads and gogol has 4 each is 1 5X slowmach Tsnet Node multiprocs goethe gogol Tsnet Appl nodelist slowmach fastguys multiprocs scheduler parameters Tsnet Appl masterload 5 Tsnet Appl workerload 1 maxprocspernode is set high so gogol can get a lot Tsnet Appl maxprocspernode 8 Special Purpose Resources Tsnet stahl speedfactor 5 25 Tsnet sand speedfactor 5 25 Tsnet goethe speedfactor 3 Tsnet gogol speedfactor 6 This file attempts to set speedfactor values for the various nodes reflecting their relative CPU capacities The maxprocspernode resource is set to a high value so that the multiprocessor system with 4 CPU s can run up to two workers per processor Special Purpose Resources ntsnet provides several resources for use with various optional features which will be described in this section Tuple Redirection Optimization By default the TCP Linda system uses up e redirection
202. starts Workers worker processes each running the routine worker t While this program structure in which the workers each call the main routine is unusual the technique of having workers repeat the setup code is not because it is often faster Linda User Guide 75 76 Case Studies Once the workers have all started do trace creates the task tuple scaninfo As in the Freewake example Rayshade uses a counter tuple to indicate which task in this case which scan line should be done next Here the counter is initially set to the maximum scan line number to be computed each worker will decrement the counter as it grabs a task and when the counter falls below 0 all tasks will be complete The second for loop gathers completed scan lines from tuple space again counting down from the maximum scan line number and sends each one to the output file The lines need to be gathered in order so that they are placed into the file in the proper location they may not artive in tuple space in that order however so do_trace may spend some time blocked The last piece needed for the parallel version of Rayshade is the worker function worker rd comm args 7 args arge for i 0 i lt afgo tTf i argv i char argsti rayshade main argc argv 1 rd lopRay 7 TopRay rd JitSamples JitSamples Set sampling parameters while 1 in scaninte 7 Y out scaninfo yal if y lt 0 brea
203. string In this case the entry has the effect of setting up equivalences among the listed set of remote directories Wildcards are allowed in map translation file entries The asterisk character can be used for any node name or as the first component of a node name e g com The ampersand character amp substitutes the current node name at the time and in the context in which the translation is taking place within a directory pathname It can be used in either generic or specific directory specifications See the discussion in Chapter 4 for full details on map translation file entries Environment Variables The following environment variables are used within the Linda system DEBUGGER LINDA CC LINDA CC LINK LINDA CLC LINDA FLC LINDA FORTRAN Specifies the debugger to use when combining a native debugger with Tuplescope The default is dbx xdb under HP UX Used by the 1inda cc shell script specifies the C compiler to use for compiling c files defaults to cc Used by the linda cc link shell script specifies the command to use for linking the executable defaults to cc Used by the C Linda compiler specifies which type of executable to build linda tcp or cds Code Development System Used by the Fortran Linda compiler specifies which type of executable to build linda tcp or cds Code Development System Used by the 1inda fortran shell script specifies the Fortran c
204. structure p Structure fields are one way to create tuples containing one record of data For example the following loop retrieves NREC records from a database and places them into the tuple space Each record is identified by the integer record number in the tuple s second field ine dos struct REC Ss for 150 i lt NBEC i 4 get next rec amp s Guti record iy 8 Structures may also be used as actuals in templates in structure t In this case the structures in the tuple and template must have the same structure tag the same size and they must be identical on a byte by byte basis including values in any padding bytes When using such constructs be careful to take structure padding into account An atray of structures is treated just like any other array int len struct STYPI se 20 20 dub struct stray si matches sets l n 10 int struct array t len 34 Using the Linda Operations Specifying Tuples and Basic Tuple Matching Rules Varying Length Structures You can use the colon notation with structures to specify them as varying In this case the length is taken to be the size of the structure in bytes This construct was designed for structures having varying arrays as their last elements Here are two examples struct STYPE double guy Dy es int buf_len int bwt 1 V int bytes STYPE s declared struct length
205. t double struct union unsigned int long short char Timing Functions The C Linda function names ate listed below start timer Initializes and starts the stopwatch in the current process A separate call to start timer is required in each process where timing is desired timer split string print times Support Functions Takes a stopwatch reading when called and labels the time split with the specified string length 32 bytes The maximum number of timer splits is 32 Prints a table listing all time splits executed so far for this process Each row includes the time split and its associated string The C Linda function names are listed below The Fortran Linda versions have an f prepended to the C Linda function name fllexit status f lhalt status f lintoff f linton 106 Linda Usage and Syntax Summary Replacement for the C exit function The lexit routine allows an eval process to abort the execution of the routine invoked through the eval operation but still continue as an eval server The status value int passed to lexit is placed into the corresponding field of the live tuple subject to typecasting restrictions Terminates execution of the entire Linda application not just the local process after calling any termination handlers specified by lonexit see below Provides the exit value returned by ntsnet Blocks the interrupts associated with tuple space handling
206. ta tuple is placed into tuple space While it is not literally true that an eval creates the processes that evaluate its arguments it can be helpful to think of it this way For example this eval statement will result in a process being created to evaluate its third argument f i C Form Fortran Form eval test i f i eval test i f 1 evals ate often used to initiate worker processes as in the following loop Linda User Guide 11 C Form Fortran Form for i20 i lt NWORKERS i Do 5 I 1 NWORKERS eval worker worker eval worker worker 5 Continue This loop starts NWORKERS worker processes In this case the primary function of the eval is simply to start the process rather than to perform a computation and place the result into tuple space Formally howevet when each worker finishes a tuple of the form C Form Fortran Form worker 0 worker 0 is placed into tuple space assuming that the workers terminate normally and adhere to the usual UNIX retutn value convention of zero for success The other two operations allow a process to access the data in tuple space A rd operation reads a tuple from tuple space and an in operation removes a tuple from tuple space Both rd and in take a template as their argument A template specifies what sort of tuple to retrieve Like tuples templates consist of a sequence of typed fields some of which hold values such fields are known as actuals
207. te Here is a sample run Program hello world 8 Execution o world from number lo world from number lo world from number lo world from number lo world from number lo world from number lo world from number gt Aw a amp O EG lo world from number I 0000000 0 g oO lo_world is finished It is to be expected that the messages from the various processes will display in non numerical order since we ve done nothing to force them to display sequentially Linda programs are essentially asynchronous and there is no guarantee that a particular process will execute before any other Indeed we would not want to do so since we re trying to achieve a simultaneously executing parallel version of hello world To run on a network the program is compiled and linked in essentially the same way but running it requires using a slightly different command For this version we might want to add the node name to the output line from each worker C gethostname name 20 printf Hello world from number bd running on s Mn i name Shown below are the commands to create and run the modified program C Linda version is shown 20 Using the Linda Operations Program Execution with TCP Linda Linda Operations clc o hello world hello world cl ntsnet hello world 8 Hello world from number running on moliere Hello world from number running on ibsen He
208. th left hand neighbor if first 1 r out RtoL left iteration values 1 in LtoR id iteration values 0 end if Exchange data with right hand neighbor if firsttnpoints 1 l tpoints 1 out LtoR right iteration values npoints in RtoL id iteration values npoints 1 end if Update points along line for j 1 j lt npoints j Global endpoints if first j 1 1 first j 1 tpoints newval j 0 0 else Use wave equation to update points newval j 2 0 values j oldval j sqteu values j 1 2 0 values 31 values 3 1 end if end npoints for for j 1 j lt npoints j oldval j values j values j newval jl 148 Sample Programs Concurrent Wave Equation end nsteps for loop return 0 Receive results from workers and print mf int get_results int tpoints int nworkers local variables int ly jy K tpts first Length double results 1000 Store worker s results in results array for i 1 i lt nworkers i 1 Receive number of points and first point in results 1 first resultsilength put it into the results array first first 1 adjust first to array starting at U j 0 for k first k lt length first k j values k results jl
209. that divides the work into many tiny tasks is said to be fine grained while one that divides the work into a small number of relatively large ones is called coarse grained There is no absolute correct level of granularity Neither coarse grained nor fine grained parallelism is inherently better or worse than the other However when overhead overwhelms computation a program is too fine grained for its environment whatever its absolute granularity level may be The optimum level depends on both the algorithms a program implements and the hardware environment it runs in A level of granularity that runs efficiently in one environment for example a parallel computer with very fast interprocessor communication channels may not perform as well in another such as a network of workstations with much slower communication between processors There are two ways to address this issue and we ll look at examples of both of them in the course of this manual First many problems offer a choice of granularity level For example if a program must execute eight independent matrix multiply operations a parallel program could perform all eight of them at the same time or execute eight parallel matrix multiplication operations one after another Which approach is correct depends on the structure of the overall problem and each is better than the other in some circumstances The other solution is to build adjustable granularity into programs so that they can
210. the high resource is false Linda internodal communication is throttled so Linda User Guide 117 kainterval kaon lindarcparg lindarsharg loadperiod masterload maxnodes that it does not flood the network and thereby degrade the performance of the network Linda application and other network users For small networks 2 4 nodes specifying high as true will probably not make a difference On large networks specifying high as true and thus asking the Linda kernel not to throttle internodal communication may cause the network to flood Specifies how often in seconds each Linda process sends out a keep alive message The default is 100 seconds The range of legal values is 100 to 31536000 one year The range is silently enforced This resource is only useful when the keep alive mechanism is used that is when kaon is true Specifies whether of not the keep alive mechanism is used The default is true unless debug is true in which case it is false Specifies a string to be passed to the linda rcp shell script called by ntsnet to distribute executables to remote nodes This resource provides a hook enabling the user to change the behavior of the shell script which can itself be modified by the user The default implementation of linda_rep located in the Linda bin subdirectory takes no arguments and so ignores the value of this resource This is a node specific resource Specifies a string to be passed to the linda rs
211. the particular data records are is less important than the general issues all such searches raise many of which are applicable to other types of problems as well Readers wanting a more rigorous and specific treatment of this topic should consult Chapters 6 and 7 of How to Write Parallel Programs by Carriero and Gelernter This case study discusses the following techniques and issues Distinct task tuples Using watermarking to avoid overwhelming tuple space Task ordering to aid in load balancing e Dealing with unequal task sizes Here is a simple sequential program in Fortran to search a database Program DB S arch Call Get target target 10 Call Get next DB rec If rec EQ EOF go to 100 Call Compare target rec result Call Process result Goto 10 100 Continue This program compates a target record or key against records in a database The following discussion assumes that the operation of comparing two data records takes a substantial amount of time Many such database applications exist including ones designed for DNA sequence searching credit application retrieval and many other sorts of complex string matching The program hides all the details of the operation in its component subroutines get_next_record retrieves another record from the database compare compares the record to the target and process takes compare s result and keeps track of which records have mat
212. the right will create 100 concurrent processes to evaluate the function f for each i value As each process finishes the resulting data tuple will go into tuple space eval s Inherited Environment In C Linda the environment available to evaled functions consists solely of the bindings for explicitly named parameters to the function Static and global initialized variables in the function are not currently reinitialized for the second and following evals and thus will have unpredictable values The created process does not inherit the entire environment of its parent process Thus in the preceding eval example the environment passed to the function f will include only the variable x Under Fortran Linda created processes inherit only the environment present when the Linda program was initiated with no modifications due to execution of user code In many implementations this is achieved by saving a clean copy of the program image from which to clone new processes Some distributed memory Linda systems satisfy these semantics exactly only when the number of evals does not exceed the number of processors Linda User Guide 23 Consider the following program structure Block Data Integer val Common params val Data val 5 End Subroutine real main Integer val Common params val val 1 Call 3 eval worker 3 Return End Subroutine F I Integer val Common params val Return End
213. these topics Program Preparation Linda programs must be compiled with the linda tuple_scope option to use Tuplescope The environment variable LINDA_CLC or LINDA_FLC should also be set to cds for Code Development System For example the following commands will prepare the C Linda program test24 for use with Tuplescope setenv LINDA CLC cds clc o test24 linda tuple scope test24 cl o oe Invoking Tuplescope Once an executable has been prepared for use with Tuplescope simply invoking it will start the debugger For example the following command would initiate a Tuplescope session with the test 24 application we prepared above 9 test24 arguments Of course this command would need to be run from a shell window within an X Windows session If desired you may also include X Toolkit options following the program arguments to customize the resulting Tuplescope windows T f you use the Bourne shell the equivalent commands are LINDA CLC cds export LINDA CLC Linda User Guide 95 The Tuplescope Display Below is a canonical diagram of the Tuplescope display note that exact window properties and look depend on the window manager in use Tuplescope control panel executable name Jak Hodes arme Run Break Contre Debug Seve Gus EBM L J T process starting up menu buttons run speed slider globals l task result worker tuple window icon
214. tiation Delays ntsnet initiates worker processes by running the rsh command in the background on the local node creating each successive process as soon as the previous command returns Under some unusual network circumstances such a procedure can overwhelm a network server process and result in errors The delay resource is provided to handle such situations It specifies the amount of time to pause between successive rsh command Linda User Guide 67 initiations in milliseconds the default value is 0 If you experience such problems try setting its value to 1000 1 second The delay command line option may also be used to specify the value for this resource Appropriate Granularity for Network Applications A network environment often has a significantly different granularity profile from other parallel computing environments such as shared or distributed memory multiprocessors because its communications speed is so much slower This is the result of the differing communications bandwidth achievable via Ethernet and typical hardware interconnects in parallel computers In general networks impose relatively high communications overhead and parallel programs need to be relatively coarse grained to achieve good performance On a typical 100mbs Ethernet network an out in combination takes approximately 500 microseconds and maximum throughput is about 10MB per second Thus to out and then in a tuple of size S takes about S 10000000
215. to specify two rules for each host to and from a generic value known as a generic directory A generic directory is a string often a directory pathname used essentially as a key into the various entries in the map translation file The generic directory serves as the target for specific local and remote directories as ntsnet attempts to translate them for use on various nodes Map translation file entries use the following commands mapto Map a specific local directory to a generic directory mapfrom Map a generic directory to a specific directory usually on a remote node Linda User Guide 53 54 Using TCP Linda map Eguivalent to a mapto and a mapfrom mapping specific directories to and from a generic directory Here is the general syntax for a map translation file entry mapverb generic dir node name specific dir node name specific dir The generic directory is always translated to a specific directory before being used as a location for executables or as the working directory on any node Thus there is no requirement that it even exist as a real directory path In fact it is possible to use any arbitrary symbolic name as a generic directory aurora home NFS mounted directory ae as t s vi a a e chaucer 2 blake lu a home a e 4 e e a a a a E d degas ES at erasmus Amp z ox mf a A ot a a a gt oa a a s
216. tsnet moliere speedfactor 2 Lines beginning with an exclamation point are comments The second line sets the working directory to tmp on the node moliere when the application hello_world is run there The third line sets the maximum number of processes that can run on any one node when the application hello world is executing to 1 and the final line sets the speed factor for the node molere to 2 where the default value is 1 indicating that it s about twice as fast as the norm In configuration file entries the program application and node components if used can be either the class name or a specific instance of a class Class names recognizable by their initial capital letter act as wildcards stating that the entry applies to all instances of the specified class In this way they can setve as default values for specific instances specific applications and or nodes for which no explicit entries have been created The following table lists the possible values for each of these three components of a configuration file entry Linda User Guide 47 48 Using TCP Linda Item Associated Class Name Example Instance program Tsnet ntsnet application Appl user application program name node Node node name user defined node resource Currently nt snet is the only valid specific instance of the class Tsnet so the two are effectively equivalent Thus for every entry the program component will be either the class Tsnet ot its only
217. u sqtau dtime 0 3 1 0 dx 1 0 tau e dtime Y dx Sqtau tau tau newval i 2 0 values i oldval i isqtau values ri 1 2 0 values il v lues 1 1 Linda User Guide 143 EA Update the points a specified number of times void update int i jy tpts update values for each point along string update points along line for j 1 j lt tpoints j global endpoints Xr Xi 1 1j tpolnts newval j 0 0 else do math j for j 1 j lt tpoints j oldval j values j jl newval jl values j J print it out for validation tpts Epoints 10 Epoints 10 print i tfirst 3d points for validation n tots for i 07 1 tpEs i prantr s4 2f values i prints ya s Main program main int left right get program parameters and initialize wave values init param init line update values along the line for nstep time steps update return 0 144 Sample Programs k sy Concurrent Wave Equation HEE k k EEE EEE EEE EEE EEE EEE E E E FE HE EE HEE HEE HEE HE HE HE EE E ERE E E E E HE SERIAL Wave Equation Makefile FILE make ser_wave c DESCRIPTION See ser_wave c USE mak f make ser_wave c HEE EEE EEE EEE EEE EEE HEHEHE EEE EE HEE HEE HEE HEE HEE HEE HE HE HE EE ERE E E E E HE
218. uple space by matching a tuple to a template not by say specifying a memory address or a position in a list This characteristic defines tuple space as an associative memory model The examples so far have all used a string as the first element of the tuple This is not required but is a good practice to adopt in most cases because it makes Linda programs more readable and easier to debug It also helps the Linda compiler to easily separate tuples into discrete classes enabling more efficient matching The next chapter will look at more complicated tuples and tuple matching scenarios and will present some simple examples of parallel programs using Linda Linda User Guide 13 How and Where to Parallelize This section contains a short discussion of how to find the computational kernels in a program It discusses the UNIX profiling utilities prof and gprof The steps described here are independent of Linda and are usually done before parallelizing the program They are designed to help you determine where to focus your efforts within the original application This section by nature is introductory and brief consult the relevant manual pages and related works in the Bibliography for more detailed discussions of these commands and profiling in general To use the UNIX profiling utilities the p for prof or pg for gprof options must be included on the link command Note that they ate not needed for compilations only for linking For example
219. useglobalconfig sets the resource to false and the global configuration file will be ignored The polarities of the plus and minus signs may seem counterintuitive at times just remember that minus means on which is the usual convention used by most X applications Note that all options which require parameters for example the command line option maxprocspernode option we looked at earlier are preceded by a hyphen a prepended plus sign has no meaning for them and will generate an error Their values follow them separated by a space Linda User Guide 49 Not all resources have named command line options The opt option is provided so that any resource s value can be specified from the command line It takes a valid configuration file entry as its parameter enclosed in double quotation marks ntsnet opt Tsnet moliere available no hello world A third type of resource enables users to create named node lists Here are some examples Tsnet Appl fran moliere gaughin voltaire pascal Tsnet Appl eng chaucer marlowe blake joyc Tsnet Appl chem sparcs rs6k priestley dalton Each of these lines defines a name for a list of nodes The first line defines a list of nodes to be associated with the name sparcs for example When a list name is used as a component in another list its name is preceded by an at sign to indicate resource indirection as in the third line above Up to 16 levels of indirection are allowed
220. ve joined the execution group then execution will begin Otherwise execution will begin as soon as minworkers workers do join or the maxwait interval has expired the latter includes the time in minwait If there are still not maxworkers workers when the maxwait interval ends execution will terminate Specifies the minimum number of workers started for a given application The default is 1 Thus the default minimum shall be a master process and one worker see maxworkers Specifies whether or not workers on a specific node run nice d This resource is node and application specific The default is true When the high resource is set to true this resource is ignored When debug is true its value is overridden to be false Specifies the pathname of the file containing a list of nodes on which this application can run The default is t snet nodes If nodefile and nodelist are both undefined ntsnet shall look for the list of nodes on which to run in the tsnet nodes file in the current working directory This is the default behavior and is backwards compatible with the old tsnet utility The nodefile resource is only used if the nodelist resource is set to nodefile which is the default value See the description of the nodelist resource for more details Specifies a space separated list of nodes on which an application may run The nodelist value can be set to any one or a combination of these items the key word nodefile a node na
221. xt usually commands to the operating system typed by the user Italic type is used for replaceable arguments in operation and command definitions for summaty lines and other descriptions within example code and occasionally for emphasis Boldface sans serif type is used for Linda operation names used in a generic way within normal text for non varying text within syntax definitions and for menu buttons and other controls in The Tuplescope Display Italic sans serif type is used for replaceable arguments within formal syntax definitions and when they are referred to within the text Linda User Guide 3 4 Introduction 1 Overview of Parallel Programming and the Linda Model People always are trying to make programs run faster One way to do so is to divide the work the program must do into pieces that can be worked on at the same time on different processors More formally creating a parallel program depends on finding independent computations that can be executed simultaneously Producing a parallel program generally involves three steps Developing and debugging a sequential program For existing programs this step is already done Transforming the sequential program into a parallel program Optimizing the parallel program Of course if the second step is done perfectly then the third one will be unnecessary but in practice that rarely happens It s usually much easier and quicker in the long run to tr

User Guide - Scientific Computing Associates

Contents

Download Pdf Manuals

Related Search

Related Contents