[
https://issues.apache.org/jira/browse/HDDS-7622?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17646263#comment-17646263
]
Wei-Chiu Chuang commented on HDDS-7622:
---------------------------------------
This is the doc I put up a few months ago while debugging Impala performance. I
took a few ideas and benchmarked their respective performance.
> Improve container re-replication throughput
> -------------------------------------------
>
> Key: HDDS-7622
> URL: https://issues.apache.org/jira/browse/HDDS-7622
> Project: Apache Ozone
> Issue Type: Task
> Reporter: Wei-Chiu Chuang
> Priority: Major
>
> We identified a few deficiencies in the current container re-replication.
> I had a document to summarize them. Will post it for review.
> Basically:
> 1. we should spread the re-replication traffic across the disk volumes in a
> DN. (This is similar to HDDS-7083)
> 2. re-replication traffic should not be compressed.
> 3. similarly, OM snapshot download traffic should not be compressed.
> There are other things like HDDS-5188 that is worth revisiting.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
