[
https://issues.apache.org/jira/browse/SPARK-54001?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dongjoon Hyun updated SPARK-54001:
----------------------------------
Parent Issue: SPARK-54357 (was: SPARK-51166)
> Reduce memory footprint from cached local relations upon cloning
> ----------------------------------------------------------------
>
> Key: SPARK-54001
> URL: https://issues.apache.org/jira/browse/SPARK-54001
> Project: Spark
> Issue Type: Sub-task
> Components: Connect, Spark Core
> Affects Versions: 4.1.0
> Reporter: Venkata Sai Akhil Gudesa
> Assignee: Pranav Dev
> Priority: Major
> Labels: pull-request-available
> Fix For: 4.1.0
>
>
> Cloning sessions is a common operation in Spark applications (e.g., for
> creating isolated execution contexts). The current approach of duplicating
> cached data can significantly increase memory footprint, especially when:
> * Sessions are cloned frequently
> * Cached relations contain large datasets
> * Multiple clones exist simultaneously
>
> An improvement can be made by implementing reference counting as opposed to
> data replication for the block manager entries that reference cached local
> relations.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
