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.

Sign up for GitHub

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

Jump to bottom

Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies (No Shared Dir / Mixin / Sep 3rd) #181

Closed
wants to merge 15 commits into from
Closed

Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies (No Shared Dir / Mixin / Sep 3rd) #181

wants to merge 15 commits into from

Conversation

TaekyungHeo
Copy link
Member

@TaekyungHeo TaekyungHeo commented Aug 21, 2024 •
edited
Loading

Summary

This PR relies on #180

  • Refactoring: Refactored code to separate Kubernetes and Slurm strategies, including output readers and report generation.
  • Kubernetes Support: Added Kubernetes-specific strategies for NCCL tests, including:
    • KubernetesNcclTestGradingStrategy
    • KubernetesNcclTestJobStatusRetrievalStrategy
    • NcclTestKubernetesJsonGenStrategy
    • KubernetesNcclTestReportGenerationStrategy
  • NCCL Test Templates: Removed the "_mpi" suffix from subtest_name in all NCCL test templates.
  • Test Coverage: Added unit tests for new Kubernetes strategies and updated existing tests to accommodate changes.

Test Plan

CI passes.

@TaekyungHeo TaekyungHeo added the Oct24 Oct'24 release feature label Aug 21, 2024
@TaekyungHeo TaekyungHeo changed the title K8s nccl Update NCCL test templates and introduce Kubernetes-specific strategies Aug 21, 2024
@TaekyungHeo TaekyungHeo changed the title Update NCCL test templates and introduce Kubernetes-specific strategies Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies Aug 21, 2024
@TaekyungHeo TaekyungHeo force-pushed the k8s-nccl branch 2 times, most recently from c05f803 to 9e8eae3 Compare August 22, 2024 11:08
@TaekyungHeo TaekyungHeo mentioned this pull request Aug 22, 2024
Closed
TaekyungHeo and others added 7 commits August 29, 2024 16:03
@TaekyungHeo @wpeng102
Add Kubernetes job
92207bc 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add Kubernetes system
7b630a0 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add Kubernetes system parser
37ca92e 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add example Kubernetes system schema
73565af 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add Kubernetes installer
a9effd6 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add Kubernetes runner
e8c1139 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo @wpeng102
Add Kubernetes support for Sleep
c4ec665 
Co-authored-by: Peng Wang <pengwang@nvidia.com>
@TaekyungHeo TaekyungHeo mentioned this pull request Aug 30, 2024
Merged
@TaekyungHeo TaekyungHeo changed the title Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies (Sep 3rd) Sep 3, 2024
@TaekyungHeo TaekyungHeo changed the title Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies (Sep 3rd) Update NCCL Test Templates and Introduce Kubernetes-Specific Strategies (No Shared Dir / Mixin / Sep 3rd) Sep 3, 2024
@TaekyungHeo TaekyungHeo closed this Sep 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@amaslenn amaslenn Awaiting requested review from amaslenn

@artemry-nv artemry-nv Awaiting requested review from artemry-nv

@srivatsankrishnan srivatsankrishnan Awaiting requested review from srivatsankrishnan

@srinivas212 srinivas212 Awaiting requested review from srinivas212

Assignees
No one assigned
Labels
feature Oct24 Oct'24 release feature
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@TaekyungHeo