[
https://issues.apache.org/jira/browse/SPARK-55120?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ASF GitHub Bot updated SPARK-55120:
-----------------------------------
Labels: pull-request-available (was: )
> Embedding spark-submit
> ----------------------
>
> Key: SPARK-55120
> URL: https://issues.apache.org/jira/browse/SPARK-55120
> Project: Spark
> Issue Type: Improvement
> Components: Kubernetes, Spark Submit
> Affects Versions: 3.1.0
> Reporter: Vincent Janelle
> Priority: Trivial
> Labels: pull-request-available
>
> When combined with the kubeflow spark operator, the spark-submit process
> tends to consume a lot of resources from either concurrent or multiple JVM
> executions. The process could be simplified to a single process (design
> pending), however the spark-submit process statically assigns two UUID values
> for the driver and configmap names.
> To reduce the total implementation (ie to track the upstream spark), it would
> be advantageous to create these UUID values dynamically.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
