You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While working on #257 today I noticed that the current checkpoint code only uses the temperature and pressure data from rank 0 when writing status information to stdout. This seems potentially misleading. Should it be doing reductions over the whole domain instead?
The text was updated successfully, but these errors were encountered:
Correct. The original code which collected field statistics across all ranks was removed in favor of using the discretization's built-in norms. It was my understanding that these norms would be extended on the inside to perform the parallel reductions.
This whole thing may be OBE after #96, as I think that facility may take over responsibility for doing runtime field data stats/reports.
While working on #257 today I noticed that the current checkpoint code only uses the temperature and pressure data from rank 0 when writing status information to stdout. This seems potentially misleading. Should it be doing reductions over the whole domain instead?
The text was updated successfully, but these errors were encountered: