[
https://issues.apache.org/jira/browse/FLINK-30818?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
chenyuzhi updated FLINK-30818:
------------------------------
Description:
When recovering a job which using window operator, taskmanager will need more
resouce(cpu/memory) to process incoming data because there are too many created
window on the cenario of data backlog.
Thus, maybe we should add metrics to measure the count of window.
was:
when recovering a job which using window operator, taskmanager will need more
resouce(cpu/memory) to process incoming data because there are too many created
window on the cenario of data backlog.
Thus, maybe we should add metrics to measure the count of window.
> add metrics to measure the count of window
> ------------------------------------------
>
> Key: FLINK-30818
> URL: https://issues.apache.org/jira/browse/FLINK-30818
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Metrics
> Reporter: chenyuzhi
> Priority: Major
>
> When recovering a job which using window operator, taskmanager will need
> more resouce(cpu/memory) to process incoming data because there are too many
> created window on the cenario of data backlog.
> Thus, maybe we should add metrics to measure the count of window.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
