Closed henryleberre closed 1 month ago
Very good. I think we need not hold .csv
data in version control, though everything else looks great on a quick look. Several runners are failing but I have not investigated that either.
Also, I think we need to rename cu_mpi
$\to$ rdma_mpi
or gpu_mpi
or gpudirect
or something (since cu
was for cuda)
Thanks @sbryngelson. The Frontier runner is failing because of a fix for batch execution. I'll get it working with interactive execution, too.
I think I fixed all the issues.
Seems like something isn't quite right. Also you can use the existing plots if you like they are fine, I am just tweaking them a bit for other purposes.
I made this figure but you don't have to use it for anything strong_scaling_frontier.pdf strong_scaling_frontier.pdf
Description
-f
(--flags
) option that became obsolete with the addition of template files for running cases and submitting jobs.Type of change
Scope
How Has This Been Tested?
Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration
cu_mpi
.