search

openshmem-org / specification

OpenSHMEM Application Programming Interface
http://www.openshmem.org
49 stars 32 forks source link

Nonblocking Collectives #456

Open manjugv opened 3 years ago

manjugv commented 1 year ago

Feedback from Jim (May Spec Meeting): Should we consider shmem_local_complete ?

jdinan commented 4 months ago

Feedback from Feb. 27, 2024 discussion:

  1. Request handle should be passed by reference to wait/test to allow implementation to reset the request handle's value to SHMEM_REQUEST_INVALID.
  2. We discussed a counter proposal to make "nbi" collectives on the team that can be completed in bulk via shmem_team_sync or other similar operation. This would mirror the contexts behavior for nonblocking RMA/AMO.
  3. Discussion about how to manage internal psync space used by the collectives
    • Wait/test are local operations which prevents freeing symmetric memory in the wait/test. If a pool allocator is used, the pool will become inconsistent across PEs. This can be addressed through periodic synchronization when initiating a collective to "garbage collect" pSync operations.
jdinan commented 1 week ago

Notes from today's discussion -- 06-27-2024 -- NBI Collectives pSync Management.pptx