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

Merge feature ecs-anywhere-gpu support to dev #3040

Merged
merged 7 commits into from Sep 28, 2021

Conversation

prateekchaudhry
Copy link
Contributor

Summary

This change adds the ability to handle GPU tasks for external instances. For external instances, GPU support will rely on nvidia-container-runtime package, for which the support has been added in this PR. For internal instances, the behavior is unchanged.

Implementation details

  • For external instance, use DeviceRequest struct to carry GPU IDs for using nvidia-container-runtime
  • For internal instance, the environment variable NVIDIA_VISIBLE_DEVICES will continue to carry GPU IDs (unchanged behavior, continue to use default runtime)

Testing

  • Manually tested changes running GPU tasks on external and internal instances
  • Added unit tests to verify GPU IDs are properly populated for internal and external instances.
  • Verified using 'make test'

New tests cover the changes: yes

Description for the changelog

Add support for GPU tasks for external instances

Licensing

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

@ubhattacharjya ubhattacharjya merged commit 983269c into aws:dev Sep 28, 2021
@fierlion fierlion added this to the 1.55.4 milestone Sep 29, 2021
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.

None yet

5 participants