[GitHub] marcoabreu commented on a change in pull request #13602: Fix for import mxnet taking long time if multiple process launched

[GitBox <gi...@apache.org>]

marcoabreu commented on a change in pull request #13602: Fix for import mxnet taking long time if multiple process launched
URL: https://github.com/apache/incubator-mxnet/pull/13602#discussion_r240414882
 
 

 ##########
 File path: src/operator/operator_tune-inl.h
 ##########
 @@ -356,7 +356,9 @@ class OperatorTune : public OperatorTuneByType<DType> {
   static duration_t GetOMPLoopOverhead() {
     // It was found empirically that OMP times was not heavily tied to number of cores,
     // so take an average across all core counts
-    const auto max_cores = static_cast<size_t>(omp_get_num_procs()) >> 1;
+    const auto max_cores_default = static_cast<size_t>(omp_get_num_procs()) >> 1;
 
 Review comment:
   That's just diving it by two, but according to your statement and the comment, this process does not benefit of many codes. Should we maybe instead use max (4, num_cores) as default?
   
   Otherwise, the default behaviour would not change and users would still face the same challenges. I'm not a big fan of having to know a "secret" environment variable. Of course, we can have this variable to offer users the possibility to set it specifically, but generally we should choose a good default

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services