[
https://issues.apache.org/jira/browse/SPARK-47225?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
王俊博 updated SPARK-47225:
------------------------
Description:
Different stages execute different codes, and hot spots will be ignored. Task
code executed in stage is same.Stage-Isolated JVM profile analysis is easier to
locate problems.And the entire Executor profile is not necessary, we can sample
it.
# Support stage-level Isolated profile files
# Support partial Executor profile sampling
Or, we can support start Executor JVM profile from UI.
was:
Different stages execute different codes, and hot spots will be ignored. Task
code executed in stage is same.Stage-Isolated JVM profile analysis is easier to
locate problems.And the entire Executor profile is not necessary, we can sample
it.
# Support stage-level Isolated profile files
# Support partial Executor profile sampling
> Support Stage Isolated Executor JVM Profiling
> ----------------------------------------------
>
> Key: SPARK-47225
> URL: https://issues.apache.org/jira/browse/SPARK-47225
> Project: Spark
> Issue Type: Improvement
> Components: Connect
> Affects Versions: 4.0.0
> Reporter: 王俊博
> Priority: Minor
>
> Different stages execute different codes, and hot spots will be ignored. Task
> code executed in stage is same.Stage-Isolated JVM profile analysis is easier
> to locate problems.And the entire Executor profile is not necessary, we can
> sample it.
> # Support stage-level Isolated profile files
> # Support partial Executor profile sampling
> Or, we can support start Executor JVM profile from UI.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
