[
https://issues.apache.org/jira/browse/FLINK-37379?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17932290#comment-17932290
]
Zakelly Lan edited comment on FLINK-37379 at 3/5/25 8:53 AM:
-------------------------------------------------------------
Merged into master : 9f5f13ba ... 66104a3f
2.0: eece5cb0...0cb17adb
was (Author: zakelly):
Merged into master :
9f5f13ba9d4f1175d06ccc9eac8ac6c1979e4110 ...
66104a3fbbc8c3c8a5f6869070a4b44ddee093cc
2.0: in progress
> Lack of state cache load for ForSt after recovery
> -------------------------------------------------
>
> Key: FLINK-37379
> URL: https://issues.apache.org/jira/browse/FLINK-37379
> Project: Flink
> Issue Type: Bug
> Components: Runtime / State Backends
> Affects Versions: 2.0.0
> Reporter: Zakelly Lan
> Assignee: Zakelly Lan
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 2.0.0
>
>
> Currently, the file-based LRU cache lacks of loading files from the
> disaggregated DFS. Meaning that after the job recovery, there will be no
> cache available, hence the performance is low over a period of time. It is
> critical for production availability of disaggregated state.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
