[
https://issues.apache.org/jira/browse/FLINK-3071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15030053#comment-15030053
]
ASF GitHub Bot commented on FLINK-3071:
---------------------------------------
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1417#issuecomment-160175731
I also tested this on a cluster with Kafka and very large state. The
backups and kafka I/O were running along smoothly and concurrently. The state
was big enough for the checkpoint interval to be too low. But with
@StephanEwen's recent fix I only ever saw one checkpoint in progress.
> Add asynchronous materialization thread
> ---------------------------------------
>
> Key: FLINK-3071
> URL: https://issues.apache.org/jira/browse/FLINK-3071
> Project: Flink
> Issue Type: Sub-task
> Components: Streaming
> Affects Versions: 0.10.0
> Reporter: Stephan Ewen
> Assignee: Aljoscha Krettek
> Fix For: 1.0.0
>
>
> Add a thread to the stream task that handles background materialization and
> acknowledges the checkpoint ones the materialization is complete.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
