Polyaxon allows to schedule distributed job via the spark-on-k8s-operator natively.
Before you can use the sparkjob runtime, you need to make sure that SparkJob operator and the CRD (custom resource definition) are deployed in your cluster.
To be able to schedule distributed jobs with the SparkJob operator, you need to enable the operator in your deployment config.
You need to enable the operator in Polyaxon CE deployment or Polyaxon Agent deployment:
operators: sparkjob: true
Once you have the SparkJob operator running on a Kubernetes namespace managed by Polyaxon, you can check the specification for creating components with the sparkjob runtime:
version: 1.1 kind: component run: kind: sparkjob ...
For more details about the specification for creating sparkjob runtime, please check please check this section.
Running components with the sparkjob runtime is similar to running any other component:
polyaxon run --hub component -P ...
After running an operation with this component, you can view it on the Dashboard:
polyaxon ops dashboard
polyaxon ops -p [project-name] -uid [run-uuid] dashboard -y
To stop a running operation with this component:
polyaxon ops stop
polyaxon ops -p [project-name] -uid [run-uuid] stop
To run this component using Polyaxon Client:
from polyaxon.client import RunClient client = RunClient(...) client.create_from_polyaxonfile(polyaxon="path/to/file", ...)