enable gpu-aware MPI by default #121
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
For the HydroBlast3D problem (with 1 A100: for a scaling efficiency of >99%. |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
GPU tests are failing because |
This was referenced Sep 29, 2022
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
BenWibking
added a commit
that referenced
this pull request
Oct 3, 2022
* enable gpu-aware MPI by default * updated README for GPU-aware MPI * updated batch scripts * change fextract to use device-allocated data * copy reference solution to device when needed * interpolate.h -> interpolate.hpp * initialize on device * avoid LoopConcurrentOnCpu
BenWibking
added a commit
that referenced
this pull request
Oct 3, 2022
This reverts commit 4250108.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
It turns out GPU-aware MPI works great, as long as managed memory is disabled and all CUDA devices are visible from each MPI process. These conditions mean that with OpenMPI, the
cuda_ipctransport will be used.On-node scaling is now ~99% efficient on Delta, Gadi.
Note: this will need to be re-tested for Setonix and Frontier.
For AMD devices, see also: https://docs.amd.com/bundle/AMD-Instinct-MI250-High-Performance-Computing-and-Tuning-Guide-v5.3/page/GPU-Enabled_MPI.html