Difference between revisions of "Intel Trace Analyzer and Collector"
Jump to navigation
Jump to search
import>Cen1001 |
import>Cen1001 |
||
| Line 1: | Line 1: | ||
This is a thing that came as part of the Intel cluster toolkit on zero but I am still figuring out how to use it, sp these are rough notes. |
This is a thing that came as part of the Intel cluster toolkit on zero but I am still figuring out how to use it, sp these are rough notes. |
||
| + | module add itac |
||
mpicc -trace -o foo foo.c |
mpicc -trace -o foo foo.c |
||
mpicc -tcollect -o foo foo.c # not sure what the difference is |
mpicc -tcollect -o foo foo.c # not sure what the difference is |
||
Revision as of 16:06, 25 June 2008
This is a thing that came as part of the Intel cluster toolkit on zero but I am still figuring out how to use it, sp these are rough notes.
module add itac mpicc -trace -o foo foo.c mpicc -tcollect -o foo foo.c # not sure what the difference is itcpin -- foo mpiexec foo
The output is voluminous and rapidly causes the nodes to swap. You can filter it.
[cen1001@zero tracing]$ cat itac.conf MEM-FLUSHBLOCKS 256 # flush to disk once you have 16Mb of data MEM-MAXBLOCKS 1024 # stop and flush if you have more than 64Mb FLUSH-PREFIX /scratch/cen1001 # where to flush to STOPFILE-NAME /home/cen1001/STOP # this doesn't work unless you use the failsafe libs echo $VT_CONFIG /home/cen1001/src/NMM-MPI/tracing/itac.conf qsub -v PATH,LD_LIBRARY_PATH,VT_CONFIG run.sh
traceanalyse