How do I specify/increase CPU usage for a TFjob served on KubeFlow?

6/5/2019

I have a GKE setup running KubeFlow on the latest versions with Kustomize. The master TFJob pulls a Docker image of the full model and runs it. I'm running into a simple issue where I wish to increase the amount of CPU usage but can't seem to do it.

Is this something I need to specify in the Docker model code, in the Kubernetes configuration, in the TFJob yamls, or somewhere else?

I've tried specifying a minimum size & limit for cpu resources for the container. I also tried specifying those for the master pod but it doesn't seem to update.

There also isn't much documentation or many similar questions on this.

-- Anton Zabirko
google-cloud-platform
google-kubernetes-engine
kubeflow
kubernetes
tensorflow

1 Answer

10/25/2019

Just refer to Customizing the TFJob

And yes, it is set in your TFJob yaml. the code snippets:

resources:
          limits:
            cpu: '1'
-- Jason Lee
Source: StackOverflow