DIET Tutorial
Contents
1. That way you will be able to verify if the answer is correct or not You will have to remember that the profile used in the client must match exactly the server profile DIET Tutorial GridUse 2004 87 3 4 Setting up and testing the client server The file env_vars bash in for bash shell respectively env_vars csh in for tcsh shell contains all the environment variables needed for the execution of the programs Verify the values of those variables then load this file using the following method gt source env_vars bash When done with this operation you need to start the name server of omniORB omniNames To do that you must give a port number with the start option on which the service will be opened and on which the server listens omniNames start lt port gt Tue Dec 11 14 10 28 2003 Starting omniNames for the first time Wrote initial log file Read log file successfully Root context is TOR 010000002b00000049444c3a6f6d672e6 72672 436f734e616d696e672 4e616d696e674 36 6e746578744578743a312e300000010000000000000060000000010102000d000000313430 2 37372e31332e36310000554f0b0000004e616d6553657276696365000200000000000000080 000000100000000545441010000001c0000000100000001000100010000000100010509010100 0100000009010100 Checkpointing Phase 1 Prepare Checkpointing Phase 2 Commit Checkpointing completed Then you have to copy this port number in the omniORB configuration file the name and location2. of this file is given by the environment variable OMNIORB_CONFIG which is defined in the env_vars bash file Using the examples configuration files found in SDIET_HOME src examples cfgs write your own configu ration files suggestion place them in a cfgs directory You will want to create a hierarchy of agents to make it interesting This hierarchy should contain at least one MA and one LA Compile server and client with the Makefile then launch the server and the client Finally launch several servers in different windows you can use same or different distributed computers With different windows you will see which one is activated You should experiment with different hierarchies 3 5 Another version of the service In this part you will modify the server to make it support a slightly different version of the scalar by matrix product The matrix will be transmitted as a file and not anymore in memory DIET doesn t impose anything about the data format of files but it would be a good idea to facilitate your work to use the data format used in the skeleton files This format is just simple text the file contains a series of numbers separated by space characters The meaning of the numbers is as follows matrix dimensions number of rows number of columns matrix values Create a file containing a matrix then implement a new service smprod_file with the following parameters parameter typ
3. BLAS do matrix vector opera tions and Level 3 BLAS do matrix matrix operations Because the BLAS are efficient portable and widely available they re commonly used in the development of high quality linear algebra software with BLAS DIR specify the root installation directory of BLAS Basic Linear Algebric Subroutines with BLAS includes DIR specify exact header directory for BLAS with BLAS libraries DIR specify exact library directory for BLAS with BLAS extra ARG ARG1 ARG2 specify extra conftest c o conftest for the linker to find the BLAS libraries use in case of several conftest c o conftest This group of options lets the user define all necessary paths to compile with the BLAS libraries Generally with BLAS DIR should be enough and the other options are provided for difficult installation of the BLAS NB these options have no effect if the module example and or its sub module BLAS are disabled ScaLAPACK The ScaLAPACK or Scalable LAPACK library includes a subset of LAPACK routines redesigned for distributed memory MIMD parallel computers It is currently written in a Single Program Multiple Data style using explicit message passing for interprocessor communication It assumes matrices are laid out in a two dimensional block cyclic decomposition Ihttp www netlib org blas 2nttp www netlib org scalapack DIET Tutorial GridUse 2004 85 Like LAPACK the ScaLAPACK routines are base
4. DIET Tutorial Rapha l Bolze Eddy Caron Philippe Combes Holly Dail and Christophe Pera INRIA Rhone Alpes 1 Introduction This is a tutorial for DIET Distributive Interactive Engineering Toolbox It comes with three directories xercise2 exercise3 and solutions The exercise directories contain the skeletons for the programs you will have to write solutions contains the solutions to the exercises 2 Exercise 1 Installing and compiling DIET The installation process is described in full in the User s Manual In the following section we provide a quick start guide Install DIET in the directory of your choice for instance HOME DIET For this tutorial FAST will not be used and as regards omniORB please add the OMNIORB_HOME bin directory in your PATH or give the following option to the configure script with omniORB OMNIORB_HOME 2 1 Dependencies 2 1 1 Hardware dependencies DIET has fully tested on Linux i386 and i686 platforms and has tested on Solaris Sparc Linux Sparc Linux i64 Linux amd64 and Linux Alpha platform and seems to be supported If you found a bug in DIET please submit a bug report at http graal ens lyon fr bugzilla If you have multiple bugs to report please make multiple submissions rather than submitting multiple bugs in one report 2 1 2 Software dependencies CORBA is used for all communications inside the platform So all of the three parts depend on it The implem
5. d on block partitioned algorithms in order to minimize the fre quency of data movements between different levels of the memory hierarchy For such machines the memory hie rarchy includes the off processor memory of other processors in addition to the hierarchy of registers cache and local memory on each processor The fundamental building blocks of the ScaLAPACK library are distributed memory versions PBLAS of the Level 1 2 and 3 BLAS and a set of Basic Linear Algebra Communication Subprograms BLACS for communication tasks that arise frequently in parallel linear algebra computations In the ScaLAPACK routines all interprocessor communication occurs within the PBLAS and the BLACS One of the design goals of ScaLAPACK was to have the ScaLAPACK routines resemble their LAPACK equivalents as much as possible with ScaLAPACK DIR specify the root installation directory of ScaLAPACK parallel version of the LAPACK library with ScaLAPACK includes DIR specify exact header directory for ScaLAPACK with ScaLAPACK libraries DIR specify exact library directory for ScaLAPACK with ScaLAPACK extra ARG ARG1 ARG2 specify extra conftest c o conftest for the linker to find the ScaLAPACK libraries use in case of several conftest c o conftest This group of options lets the user define all necessary paths for compilation with the ScaLAPACK libraries Normally with ScaLAPACK DIR could be enough but the other options are provi
6. ded because the instal lation of the ScaLAPACK libraries is often difficult and thus difficult to detect automatically For instance the with ScaLAPACK extra option is useful to integrate BLACS and MPI libraries which are useful for Sca LAPACK NB these options have no effect if the module example and or its sub module ScaLAPACK are disabled 2 2 3 From configuration to compilation An important option for configure scripts is prefix which specifies where binary files documents and confi guration files will be installed It is important to set this option it otherwise defaults to DIET install The configuration will return with an error if no ORB was found So please help the configure script to find the ORB with the with omniORB options If everything went OK the configuration ends with a summary of the options that were selected and what it was possible to get The output will look like DIET gt configure enable doc enable BLAS DIET successfully configured as follows documents yes examples dmat_manips file_transfer scalars FAST found home username DIET FAST bin fast config ORB omniORB 4 in usr prefix home username DIET install multi MA no LogService no target platform i1686 pc linux gnu Please run make help to get compilation instructions DIET gt make help Usage make help shows this help screen make agent builds DIET agent executable
7. e a scalar double a file containing matrix to be multiplied double the time needed for the product to compute float The file is overwritten by the result 4 Exercise 3 yet another service The BLAS dgemm To compile programs of this exercise the BLAS library Basic Linear Algebric Subroutines is required You also should have configured DIET with enable BLAS option 88 GridUse 2004 Bolze Caron Combes Dail and Pera DGEMM perform one of the matrix matrix operations C alpha op A op B beta C The dgemm_ function is part of the BLAS It performs the following matrix matrix computation C aAB BC o and are scalars and A B and C are matrices with op A an m by k matrix op B ak by n matrix and C an m by n matrix This exercise aims at adding a new service in a DIET platform that performs the dgemm_ computation The idea is to interface the existing dgemm_ function to a DIET SeD Here is its prototype void dgemm_ char tA char et By int m int ny int k double alpha double A int ida double B int ldb double beta double C int ldc All parameters are given by address Parameters alpha beta m n k A B and C correspond exactly to their respective roles in the computation 1da 1db and ldc are the leading dimensions of the corresponding matrices Since matrices are stored in a classical one dimension array it is important to specify if they are stored b
8. enta tions of CORBA currently supported in DIET are omniORB 3 which depends on Python 1 5 omniORB 4 which depends on Python 2 1 or later and on OpenSSL if you would like your DIET platform to be secure We strongly recommend omniORB 4 as it is easier to install and provides SSL support So far only one CORBA service is needed the Naming Service This service is provided by a name server that must be launched before all DIET entities Then DIET entities must be passed a reference to the lt host port gt of your name server Later you will have to define the following environment variables SOMNIORB_CONFIG and SOMNINAMES_LOGDIR and or for omniORB 4 only at compile time with the with omniORB config and with omniNames logdir options NB We have noticed that some problems occur with Python 2 3 the C code generated could not be compiled It has been patched in DIET but some warnings still appear 2004 by the author s GridUse 2004 Sup lec France DIET Tutorial GridUse 2004 83 Since omniORB needs a thread safe management of exception handling compiling DIET with gcc requires at least gcc3 Some examples provided in the DIET sources depend on the BLAS and ScaLAPACK libraries However theses examples do not have to be compiled 2 2 Compiling the platform Once all dependencies are satisfied untar the DIET archive and change to its root directory A configure script will prepare DIET for compiling its ma
9. er directory for omniORB with omniORB libraries DIR specify exact library directory for omniORB with omniORB extra ARG ARG1 ARG2 specify extra conftest c o conftest for the linker to find the omniORB libraries use in case of several conftest c o conftest This group of options lets the user define all necessary paths to compile with omniORB Generally with omniORB DIR should be enough and the other options are provided for ugly installations of omniORB NB having the executable omniid1 in the PATH environment variable should be enough in most cases FAST with FAST DIR installation root directory for FAST optional with FAST bin DIR installation directory for fast config optional This group of options lets the user define all necessary paths to compile with FAST Generally with FAST DIR should be enough and the other options are provided for difficult installations of FAST There is no need to specify includes libraries and extra arguments since FAST provides a tool fast config that does this job automatically NB1 having the executable fast config in the PATH environment variable should be enough in most cases NB2 it is possible to specify without FAST which overrides fast config detection BLAS The BLAS Basic Linear Algebra Subprograms are high quality building block routines for performing basic vector and matrix operations Level 1 BLAS do vector vector operations Level 2
10. in options are described below but please run configure help to get an up to date and complete usage description gt tar xzf DIET_I 1 tgz gt cd DIET DIET gt configure help short Configuration of DIET 1 1 2 2 1 Optional features for configuration disable FEATURE do not include FEATURE same as enable FEATURE no enable FEATURE ARG include FEATURE ARG yes enable maintainer mode enable make rules and dependencies not useful and sometimes confusing to the casual installer enable doc enable the module doc documentation about DIET This option activates the compilation and installation of the DIET documents which is disabled by default because it is very sensitive to the version of your ATEX compiler The output postscript files are provided in the archive disable examples disable the module examples basic DIET examples This option deactivates the compilation of the DIET examples which is enabled by default enable BLAS enable the module BLAS an example for calling BLAS functions through DIET This option activates the compilation of the DIET BLAS examples as a sub module of examples which means that this option has no effect if examples are disabled disabled by default enable logservice enable monitoring through LogService This option enables the connection from DIET to the LogService monitoring software It will enable DIET to generate monitoring data and send it to the L
11. make SeD builds DIET SeD library make client builds DIET client library make examples builds basic examples make builds everything configured make install copy files into home username DIET install This helps the user to choose the DIET parts for installation and compilation It is recommended for beginners to compile the client the SeD and the agent in one single step with make all But please pay attention to the fact that make all does not install DIET in the prefix provided at configuration time To do this run make install 86 GridUse 2004 Bolze Caron Combes Dail and Pera make install will run the compilation for all DIET entities before the installation itself Thus if the user wants to compile only the agent for instance and install it he must run DIET gt cd src agent DIET src agent gt make install 3 Exercise 2 An example of matrix computation With this example we learn how to program a simple client server application that uses DIET We will use the context of matrix computation to make this program more realistic We will implement a basic scalar by matrix product Then we will test this program in different execution schemes 3 1 File skeletons The exercise2 directory located in your home directory contains all the skeleton files needed for a quick start Useful pieces of software are also included in them This directory contains the following files Makefile management of dependencies be
12. ogService Note that the support of LogService is built in You do not have to install the LogService package to compile with this option enable ScaLAPACK enable the module ScaLAPACK an example for calling ScaLAPACK functions through DIET This option activates the compilation of the DIET ScaLAPACK examples as a sub module of examples which means that this option has no effect if examples are disabled disabled by default enable Cichlid enable generation of communication logs for Cichlid enable stats enable generation of statistics logs enable multi MA enable multi MA architecture disable dependency tracking Speeds up one time builds enable dependency tracking Do not reject slow dependency extractors enable shared PKGS build shared libraries default yes enable static PKGS build static libraries default yes enable fast install PKGS optimize for fast installation default yes disable libtool lock avoid locking might break parallel builds 84 GridUse 2004 Bolze Caron Combes Dail and Pera 2 2 2 Optional packages for configuration with PACKAGE ARG use PACKAGE ARG yes without PACKAGE do not use PACKAGE same as with PACKAGE no with gnu ld assume the C compiler uses GNU ld default no with pic try to use only PIC non PIC objects default use both omniORB with omniORB DIR specify the root installation directory of omniORB with omniORB includes DIR specify exact head
13. tween source and compiled files server c program implementing the service scalar by matrix product client_smprod c program using the service defined in server c the matrix is stored in memory client_smprod_file c same program as client_smprod c except that the matrix is passed as a file to the server 3 2 Server side implementation Using the skeleton of program server c write a service of scalar by matrix product This service will have the following parameters parameter type a scalar double a matrix to be multiplied double the time needed for the product to compute float The intial matrix is overwritten by the result The matrix will be stored in memory To start try to define a detailed interface for the service i e a precise definition of the service profile To do so look for in inout and out parameters Next program the solve function solve_smprod and also the initialization of the service in the main function int solve_smprod The following function is given to help you int scal_mat_prod double alpha double M int nb_rows int nb_cols float time It mutliplies the scalar alpha by the matrix Ma nb_rows by nb_cols matrix The results are the matrix alpha x Mand the time of this operation in seconds 3 3 Client side implementation Using the client_smprod c skeleton file write a client for the service defined above You will need to initialize a matrix and a scalar with known values
14. y rows or by columns tA and tB are characters which have the following semantics tA Storage order of A m k T row major row 1 row 2 row m N column major col 1 col 2 col k For this exercise there is no need to explore all possibilities offered by the storage order or the leading dimension Just set tA and tB to N and 1da 1db and 1dc to the number of rows of the corresponding matrix Once you have specified the profile of the service program a server that implements this service and a test client using the file skeletons in the exercise3 directory Matrices will be stored in memory Eventually test the client server architecture through DIET in different contexts of execution
Download Pdf Manuals
Related Search