OpenMp
-
Hi all. I've been trying to use openmp directives to boost preformance. I've managed to make it work, by using stuff like
#pragma omp parallel for. But it doesn't work correctly: instead of improving performance, i get better results without it. I've tried different scenarios without results. Has anyone encountered such problems, or used OpenMP successfully so I may get some tips ?there are no facts, only interpretations
-
Hi all. I've been trying to use openmp directives to boost preformance. I've managed to make it work, by using stuff like
#pragma omp parallel for. But it doesn't work correctly: instead of improving performance, i get better results without it. I've tried different scenarios without results. Has anyone encountered such problems, or used OpenMP successfully so I may get some tips ?there are no facts, only interpretations
Maybe your parallelization strategy for this specific loop is causing cache misses. Try redesigning the loop and see if you can improve cache hits. Its very important to test several variations of the critical loop. Also test both shared and privatization clauses on some variables, you may get better performance through selective privatization. Keep in mind not all nested loops benefit from parallelization. Best Wishes, -David Delaune