Project

General

Profile

Feature #2915

Updated by Alan Gray over 1 year ago

Part of this family of developments involves enablement of direct communications between GPU memory spaces. This issue is to discuss common sub-issues related to the development of these new features.

TODO
* -Investigate Investigate if GPU halo exchange inter-GPU sync in thread MPI case can be made more lightweight with cuStreamWaitValue32 (or similar).- (done with MPI exchange of pointers to events, with remote event enqueued to stream). similar).

Back