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

Ompmultigpu #7

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Ompmultigpu #7

wants to merge 4 commits into from

Conversation

vlkale
Copy link
Contributor

@vlkale vlkale commented Apr 27, 2022

OpenMP multi-GPU parallelization through dynamic-GPUstatic multi-GPU loop scheduling, i.e., thread ID tid_i offloads its assigned chunk from schedule(dynamic) to GPU ID tid_i%numGPUs, where numGPUs is the number of devices on a node of a supercomputer.

Note that the GPU should be generalized to 'devices' for truly heterogeneous parallelism and scheduling on a node.

The variable names use 'devices' rather than 'GPU', but the code is currently experiemented with for GPUs only and not CPUs and GPUs. Experimentation with scheduling across CPUs, GPUs, or any xPUs is left for future work at this time.

Adding OpenMP multi-GPU parallelization for OpenMP offload in the version without function overloading of the star operator.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant