[Perf] Add a CPU-based training workload

ray-project / kuberay

A toolkit to run Ray applications on Kubernetes

Apache License 2.0

974 stars 330 forks source link

Closed kevin85421 closed 2 months ago

kevin85421 commented 2 months ago

Our scalability test requires some CPU-based workloads. This PR uses CPU to train a MNIST model. I also can run the RayJob successfully on my devbox.

[ ] I've made sure the tests are passing.
Testing Strategy
- [ ] Unit tests
- [ ] Manual tests
- [ ] This PR is not tested :(

kevin85421 commented 2 months ago

cc @andrewsykim

andrewsykim commented 2 months ago

LGTM! Tested this locally and confirmed it works. Thanks @kevin85421!