Changes between Version 47 and Version 48 of Public/User_Guide/OmpSs-2
- Timestamp:
- Jun 17, 2019, 12:06:59 PM (6 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Public/User_Guide/OmpSs-2
v47 v48 189 189 `ssh -X USERNAME@deep.fz-juelich.de` 190 190 191 and from there open the ''multisaxpy'' 191 and from there open the ''multisaxpy'' folder: 192 192 193 193 `cd /work/cdeep/USERNAME/multisaxpy` 194 194 195 torequest an interactive cluster module (CM) node in order to use all the available 48 threads to run a pure !OmpSs-2 application:195 and request an interactive cluster module (CM) node in order to use all the available 48 threads to run a pure !OmpSs-2 application: 196 196 197 197 `srun -p dp-cn -N 1 -n 1 -c 48 --pty /bin/bash -i` … … 300 300 == Creating a Dependency Graph == 301 301 302 Let's continue with the same example used above and create a dependency graph using only 12 threads of one socket (e.g. the second), which demonstrated to be the affinity giving the best results. Furthermore, we are not longer interested in running 100 iterations (nor using a large number of elements) to benchmark this exampleand hence only one iteration will suffice to generate a complete graph of this application. Run the following command:302 Let's continue with the same example used above and create a dependency graph using only 12 threads of one socket (e.g. the second), which demonstrated to be the affinity giving the best results. Furthermore, we are not longer interested in running 100 iterations (nor using a large number of elements) for graph purposes and hence only one iteration will suffice to generate a complete graph of this application. Run the following command: 303 303 304 304 `NANOS6=graph taskset -c 12-23 ./03.multisaxpy_task 196608 8192 1` 305 305 306 This command can take some time. Once it has finished it should have created a script with the name ''graph-XXXXX-YYYYYYYYY-script.sh'' and a directory ''graph-XXXXX-YYYYYYYYY-components''. Execute the script by typing (note that it requires the tool `dot`):306 Once it has finished it should have created a script with the name ''graph-XXXXX-YYYYYYYYY-script.sh'' and a directory ''graph-XXXXX-YYYYYYYYY-components''. Execute said script by typing the following (note that it requires the tool `dot`): 307 307 308 308 `bash graph-XXXXX-YYYYYYYYY-script.sh` … … 317 317 == Obtaining statistics == 318 318 319 The visual execution of tasks can be further complemented with statistics. Runthe following command:319 The visual execution of tasks can be further complemented with statistics. Executing the following command: 320 320 321 321 `NANOS6=stats taskset -c 12-23 ./03.multisaxpy_task 196608 8192 1` … … 390 390 `NANOS6=stats-papi taskset -c 12-23 ./03.multisaxpy_task 196608 8192 1` 391 391 392 to get the following information:392 to get statistics: 393 393 {{{ 394 394 $ NANOS6=stats-papi taskset -c 12-23 ./03.multisaxpy_task 196608 8192 1 … … 513 513 == Tracing with Extrae == 514 514 515 **THIS SECTION IS WORK IN PROGRESS, PLEASE IGNORE IT** 516 515 517 To get traces of this benchmark using Extrae firstly load the corresponding module: 516 518