You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "Siyao Meng (Jira)" <ji...@apache.org> on 2020/12/09 18:27:00 UTC
[jira] [Updated] (HDDS-4570) Reduce memory footprint of
OMUpdateEventBatch for Recon when processing OM update events
[ https://issues.apache.org/jira/browse/HDDS-4570?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Siyao Meng updated HDDS-4570:
-----------------------------
Summary: Reduce memory footprint of OMUpdateEventBatch for Recon when processing OM update events (was: Reduce memory footprint of OMUpdateEventBatch for Recon)
> Reduce memory footprint of OMUpdateEventBatch for Recon when processing OM update events
> ----------------------------------------------------------------------------------------
>
> Key: HDDS-4570
> URL: https://issues.apache.org/jira/browse/HDDS-4570
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Reporter: Siyao Meng
> Assignee: Siyao Meng
> Priority: Major
>
> We are experiencing OOM when Recon heap size is too small (1GB) and it looks like {{OMUpdateEventBatch#events}} is taking most of them.
> Goal:
> 1. Reduce new object creation when Recon is consuming batches OM events ({{consumeOMEvents}})
> 2. Improve {{OMUpdateEventBatch#filter}} to reduce memory usage. e.g. avoid generating a new list?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org