Differences

This shows you the differences between two versions of the page.

--- profiling [2013/10/15 15:48] – add cuda profiling info vondele
+++ dev:profiling [2020/08/21 10:15] (current) – external edit 127.0.0.1
@@ Line 70: / Line 70: @@
   - TOTAL TIME: How much time is spent in this subroutine, including time spent in timed subroutines. AVERAGE and MAXIMUM as defined above
-Note that, for the threaded code, only the master thread is instrumented.
+By default, only routines contributing up to 2% of the total runtime are included in the timing report.  To see smaller routines, set a smaller cut-off with the [[http://manual.cp2k.org/trunk/CP2K_INPUT/GLOBAL/TIMINGS.html#desc_THRESHOLD|GLOBAL%TIMINGS%THRESHOLD]] keyword
+Note that, for the threaded code, only the master thread is instrumented.
 ==== Modifying the timing report ====
@@ Line 227: / Line 228: @@
 export PMI_NO_FORK=1
 # no cuda proxy
-# export CRAY_CUDA_PROXY=1
+# export CRAY_CUDA_MPS=1
 # use all cores with OMP
 export OMP_NUM_THREADS=8
 # use aprun in MPMD mode to have only the output from the master rank (here 169 nodes are used)
-aprun -N 1  -n 1 -d ${OMP_NUM_THREADS}  nvprof -o log.nvprof ./cp2k.psmp -i test.inp -o test.out-profile : -N 1  -n 168 -d ${OMP_NUM_THREADS}  ./cp2k.psmp -i test.inp -o test.out-profile
+COMMAND="./cp2k.psmp -i test.inp -o test.out-profile"
+PART1="-N 1  -n 1 -d ${OMP_NUM_THREADS} nvprof -o log.nvprof ${COMMAND}"
+PART2="-N 1  -n 168 -d ${OMP_NUM_THREADS} ${COMMAND}"
+aprun ${PART1} : ${PART2}
 </code>