Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add pre_register_memory option #1134

Merged

Conversation

AlexanderSinn
Copy link
Member

@AlexanderSinn AlexanderSinn commented Jul 4, 2024

On some platforms, such as JUWELS booster, the memory passed into MPI needs to be registered to the network card, which can take a long time. When using this option, all ranks can do this at once in initialization instead of one after another as part of the communication pipeline.

  • Small enough (< few 100s of lines), otherwise it should probably be split into smaller PRs
  • Tested (describe the tests in the PR description)
  • Runs on GPU (basic: the code compiles and run well with the new module)
  • Contains an automated test (checksum and/or comparison with theory)
  • Documented: all elements (classes and their members, functions, namespaces, etc.) are documented
  • Constified (All that can be const is const)
  • Code is clean (no unwanted comments, )
  • Style and code conventions are respected at the bottom of https://github.com/Hi-PACE/hipace
  • Proper label and GitHub project, if applicable

@AlexanderSinn AlexanderSinn added performance optimization, benchmark, profiling, etc. Parallelization Longitudinal and transverse MPI decomposition labels Jul 4, 2024
@MaxThevenet MaxThevenet merged commit 5d8dda3 into Hi-PACE:development Jul 16, 2024
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Parallelization Longitudinal and transverse MPI decomposition performance optimization, benchmark, profiling, etc.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants