Potential performance impact from saving MPI rank during save solution callback #2171
Labels
enhancement
New feature or request
parallelization
Related to MPI, threading, tasks etc.
performance
We are greedy
possible student project
In #2132, the MPI rank was included in
element_variables
during each call to the save solution callback for parallel simulations. It would be nice to evaluate whether this change impacts performance, and if that is the case, then introduce an option to deactivate or manually enable this functionality.The text was updated successfully, but these errors were encountered: