ov2txt - User`s manual
Contents
1. edious task specially when there are a great number of statistics and simulations This tool is intended for extracting data from the binary vectorial file ov gen erated by OPNET Modeler as a result of every simulation run without any user interaction with the OPNET graphical interface The data are written to a column separated text file suitable for being processed by R 2 or MS Excel In addition the program can compute scalar data from the vectorial data processed without process the os file Finally this program can be executed in a cluster environment like Condor 3 In this way it is possible to automate the run of a set of simulations and extract the corresponding data with a simple script invoking first to op_runsim for running the simulation and then to ov2txt for extracting the data The set of programs simul_xxx contains the following command line tools e simul_geoloc used for generating network scenarios automatically computing the partition grade according to the transmission range e simul_ah2r used for converting the binary animation file ah to a text format as e simul_run used for program sets of simulations with OPNET Modeler without using the Graphical User Interface e simul_data used for extracting the data obtained with the simulations This program calls to ov2txt for each of the simulations run e ov2txt invoked from the simul_data for extracting data from the binary vec torial fil2. eler The way to specify this method is with the character preceding the file name specified gt Example x vars txt The vars txt file contains the names of the statistics to be xported one per lin 3 3 Files 3 3 1 Input files e The input file is the name of an OPNET generated ov file mandatory e The file name should be specified without the extension ov at the end e If it contains blank spaces it should be specified within quotation marks e The absolute path should be included in mod_dirs preference within OPNET The ov file will be searched by following the same order as in mod_dirs It is possible to specify a path for the specific ov to consider i which sets temporarily the mod_dirs preference to that path 3 3 2 Output files e Two text files are generated in the working path 1 Indexed list of all variables statistics included in the input ov file file ov gt file std 2 Selected data extracted from the ov file file ov gt file xxx where xxx suffix according to the selected output format f Excel xls R data C csv e If an optional suffix is specified with the s option it will appear at the end of both output file names e g lt file sss std gt 0U2txt User s manual 3 4 Examples The following examples show how one can extract data from the ov files Let us suppose that the input file is mymodel ov Usual
3. es ov It can also be invoked for the user for extracting data for a particular simulation as described in this manual e simul_graphs used for generating graphs from the extracted data R or gnu plot 0U2txt User s manual e simul_reports used for generating a report containing the previously generated data and graph These programs have been developed using the C programming language so it is easy to migrate them to any platform currently Windows and Unix like systems as Linux Chapter 2 Installation For the instalation of the ov2txt program you only have to copy the binary to any directory and set some environment variables PATH and LD_LIBRARY_PATH In the following sections how to set these environment variables is explained 2 1 Requeriments The program ov2txt requires that OPNET Modeler be installed as it makes use of the API External Model Access EMA Moreover the OPNET Modeler version installed should match with the one used for compiling this program which is specified in the name of the executable e g ov2txt 16 0 A bin 2 2 Windows Let us suppose that the directory choosen for copying the ov2txt program is C Program files utils It is recommended to modify the PATH environment path This can be do as explained in the following e Windows 98 or previous edit the C AUTOEXEC BAT file adding the follow ing line at the end It would be possible that the path needs to be e
4. index or by name Then we can execute a second time in order to extract data by specifying the x option For example ov2txt sh m mymodel x 0 1 2 3 ov2txt sh m mymodel x names txt In the first example we are using the indexes of the statistics as shown in the previously generated text file In the second one we are using a text file called names txt with the full referenced names of the statistics one per line ex tracted from the same previously generated text file or from the OPNET Modeler graphical interface The second method is used when the character is found at the beginning of the file name next to the x option In both cases the same mymodel data file is generated The q option causes to force the statistics test file be generated too which is rarely useful The file names txt is a plain text file that could be similar to the following time sec IP Number of Hops Wireless LAN Load APP bits sec Wireless LAN Throughput APP bits sec Wireless LAN Delay sec Wireless LAN Data Dropped Buffer Overflow bits sec 10 0U2txt User s manual By default the output file s is written in the current directory that is the direc tory from the ov2txt program was invoked The output directory can be changed with either the o or w options Both are equivalent although they work in a dif ferent way So the following commands are similar in practice i res
5. ly the first thing one should do is to know the name of the statistics variables that the ov file contains So the following command should be issued ov2txt sh m mymodel Please note that the ov suffix should not be specified at the end because we are not specifying the file name but the corresponding name model By default the final ov file used is determined by searching throught the paths listed in the mod_dirs preference within OPNET Modeler exactly in that order However you can force the file to be used by using the i opcion in order to specify the unique directory where to look for the ov file This directory can be either an obsolute or relative path to the working directory Let us suppose that you have run two sets of simulations with the same network model but with different parameters each one and the resulting ov files are in the directories result v1 and result v2 respectively So there should exists two files exactly with the same name but in different directories If we want to extract data from the last simulation result v2 we can modify the mod_di rs preference or specify the i option ov2txt sh m mymodel i result v2 By executing the program without the x option only a text file is generated containing the name of all the statistics variables included in the mymodel ov file but no data are extracted In this way we can take a look to that file and choose the statistics to export either by
6. n use several functions that allow to obtain scalar data similar to the os files generated by OPNET but computed by the ov2txt program from the vectorial data ov Hd A LAST n gt Last value column n MEAN n gt Arithmetic mean column n AVG n gt Temporal mean column n VAR n gt Variance sample column n MIN n gt Minimum value column n MAX n gt Maximum value column n SUM n gt Sum column n ALL n gt All scalar data column n gt Example x 0 14 17 20 MEAN 14 MAX 17 ov2txt User s manual Method 2 by a file containing the index of each selected variable one per line This method is very similar to the previous one This could be useful when the statistics to b xported ar determined with other program which could write the indexes o a text file he way to specify this method is with the K character preceding he file name specified Example x Kvars txt he vars txt file contains the indexes of the statistics to be xported one per lin gt With this method the previous functions can also be used ct 7 v Fl ct E Method 3 by a file containing the name completely referenced of the selected variables one per line The names can be obtained from the output file containing the listed statistics or from the graphical interface within OPNET Mod
7. nation directory for output files std and data This directory can be relative to the current directory e g data or absolute e g home user op_models data By default the working directory is the current directory from the program ov2txt has been executed It is equivalent to specify the output directory o option The w option is necessary when executing ov2txt from a cluster lt suffix gt suffix to be appended to the output file names lt format gt format for the output data file where lt format gt could be f e h Excel f r h R package default h h F G to be read from a C program If the is specified a header with the name of the variables will be exported too first line for vectorial variables and first column for scalar data The default format is f rh that is R and with headers r lt data gt statistic variables to be exported to the data file Data will be written in the format specified with f R by default The statistic variables can be specified by mean of 3 different ways Me gt hod 1 by a indexes list gt 0 xample x 0 14 17 20 gt The corresponding index to each variable can be obtained from the statistics output file by executing once the program specifing only the input file m lt file gt without any other option also with q gt The index 0 is the simulation time abscissa axis x in graphs gt In addition you ca
8. nclosed within quotes if it is more than 8 characters length or it contains blanck spaces SET PATH PATH C Program files utils e Windows NT 2000 XP within the Control Panel choose System gt Ad vanced gt press the button Environment variables Then the path could be added to the PATH environment variable separated by a semicolon Alter natively the Autoexec NT file can be modified in a similar way as explained for the Autoexec bat file 0U2txt User s manual 2 3 Linux Let us suppose that the directory choosen for copying the ov2txt program is SHOME bin Under Linux it is necessary to set the PATH and LD_LIBRARY_PATH environ ment variables In the later we should to add the path for the OPNET libraries as it is necessary in order to run programs compiled with the API EMA built with op_mkema This path depends on the home directory for OPNET and version The way to do that depends on the shell or command line interpreter used bash ksh oe e Linux shells bash ksh zsh sh HOME bash_profile or HOME profile PATH SPATH SHOME bin LD_LIBRARY_PATH usr opnet 16 0 A sys pc_intel_linux lib export PATH export LD_LIBRARY_PATH e Linux shells csh tesh SHOME login setenv PATH PATH SHOME bin setenv LD_LIBRARY_PATH usr opnet 16 0 A sys pc_intel_linux lib Some Linux distributions have problems with the LD_LIBRARY_PATH and user cannot set the value This i
9. ovu2txt User s manual Miguel Hernandez University 1 April 19 2011 1Copyright c 2008 P Pablo Garrido Abenza All rights reserved CL Abstract In this manual the installation and usage of the ov2txt program is described which is included within the simul_xxxset of programs It has been developed by P Pablo Garrido Abenza pgarrido umh es as a member of the Arquitectura y Tecnologia de COMputadores Group GATCOM at the Miguel Hernandez University Contents 1 Introduction 3 2 Installation 2I REQUENA i s See e 16 dos Sie aah ec sony Ge iat ee Wow Ge a Beak ae Acme Hee 2 2 Windows 5 scsi ies da bea ee beeda eas HN a Aa a 3 Usage of the program A os be he Sa abe cay og ae omy Siete ey Se ahs He woah a en HE ah SN aoe ye my fac es S 3 2 Description iu sie S ee Ge wae WP wR HOR WR REE eai Be IGS 3 eu csaka esas oe les cdr es de Qe ge Bede By ee Be ee ae Ree a 3 3 1 MOU Mes ode oo4 24 oek bow hate eeu ae 3 3 22 Output files sis 45 eee be eee Ee Ee ER BES ES 34 EXO ples hao a ee ee oe he ee hh he Oe San eae y Bk 10 Oo OO ON NN Chapter 1 Introduction When OPNET Modeler 1 run a simulation results are written to two kind of files vectorial data ov and scalar data os These files have a binary format and the tipical way for extracting the data in order to be processed by external tools is by using the mouse within the OPNET Graphical User Interface This is a t
10. s why it is necessary to execute that each time that either you are going to execute the program or you open the console In order to simplify the process you can write a little script called ov2txt sh simi lar to the following and put it together to the binary In order to pass the arguments from the script to the executable you can use with the quotes instead of too bin sh export LD_LIBRARY_PATH usr opnet 16 0 A sys pc_intel_linux lib exec ov2txt en 16 0 A x86 bin Sx Chapter 3 Usage of the program In this chapter the usage of the ov2txt program is explained It is supposed that we have written the ov2txt sh script explained in the previous chapter 3 1 Syntax ov2txt sh ov2txt sh m input file i input path o output path w working path v q x data to export f output format s suffix 3 2 Description ov2txt is used to extract data from the ov binary files It extracts vectorial data that is complete columns corresponding to the selected statistic variables More over it also can compute scalar values from the vectorial data namely last value arithmetic mean average or temporal mean variance sample minimum value maximum value and sum The program does not run interactively Instead the program begins to extract data immediately upon invocation and goes on until the process finishes without interruption Because of this it is necessary to specify some options
11. ult v2 o data v2 i result v2 w data v2 ov2txt sh m mymodel x ov2txt sh m mymodel x In this way we can export the result data v1 and v2 to different directories Another option could be to adding a suffix to the end of the output file name with the s option ov2txt sh m mymodel x 0 1 2 3 i result vl s vl ov2txt sh m mymodel x 0 1 2 3 i result v2 s v2 In these cases all the generated data files will be written in the same directory but the names will be different mymodel vl data and mymodel v2 data respectively By default the format of the output file is suitable for the R package data That can be changed with the f option For example if we are going to process the data with MS Excel xls we can execute the following command ov2txt sh m mymodel x 0 1 2 3 f eh Now the generated output data file will be mymodel x1s and will include a header as a first line 11 Bibliography 1 I OPNET Technologies Opnet modeler Online Available http www 2 R Development Core Team R A language and environment for statistical computing 2007 ISBN 3 900051 07 0 Online Available http www R project org 3 T C Team Condor Online Available http www cs wisc edu condor 12
12. when invoking the program These are the options to 0v2txt shows this help v verbose mod q force to generating the list of statistics out file std It is also generated automatically when no data is selected to be exported By default the std file is not generated when exporting data x option in order to reduce the number of files generated unless the q option be used i lt dir gt input directory for the ov file If this option is not used the ov file will be searched in the paths specified in the mod_dirs OPNET preference in the 7 ou2txt User s manual same order The use of the i option has the advantage that the Ov processed will be the desired one as it could exists a ov file with the same name but in different path previously defined in the mod_dirs preference lt dir gt output directory Is the destination directory for output files std and data This directory can be relative to the current directory e g data or absolute e g home user op_models data If this option is not used the current path or the working directory specified with the w option will be considered The options o and w are equivalents i e both produce the same results The difference between them is that w change temporarily the directory and with o the path is used when opening the files lt dir gt change the working directory Is the desti
Download Pdf Manuals
Related Search