[
https://issues.apache.org/jira/browse/HDDS-3743?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
runzhiwang updated HDDS-3743:
-----------------------------
Description:
I start a ozone cluster with 1000 datanodes, and run two weeks with heavy
workload, and perf om.
You can find NetUtils.normalize cost 0.17% cpu, because it needs to do
Matcher.replaceAll. It happen in the critical path, i.e. lookUpKey when read.
we need not to do this when getFromProtoBuf, So we can avoid normalize by check
the last character.
!screenshot-3.png!
!screenshot-4.png!
was:
I start a ozone cluster with 1000 datanodes, and run two weeks with heavy
workload, and perf om.
You can find NetUtils.normalize cost 0.17% cpu, because it needs to do
Matcher.replaceAll, we need not to do this when getFromProtoBuf, So we can
avoid normalize by check the last character.
!screenshot-3.png!
!screenshot-4.png!
> Avoid NetUtils.normalize when get DatanodeDetails from proto
> ------------------------------------------------------------
>
> Key: HDDS-3743
> URL: https://issues.apache.org/jira/browse/HDDS-3743
> Project: Hadoop Distributed Data Store
> Issue Type: Sub-task
> Reporter: runzhiwang
> Assignee: runzhiwang
> Priority: Major
> Attachments: om.svg, screenshot-1.png, screenshot-2.png,
> screenshot-3.png, screenshot-4.png
>
>
> I start a ozone cluster with 1000 datanodes, and run two weeks with heavy
> workload, and perf om.
> You can find NetUtils.normalize cost 0.17% cpu, because it needs to do
> Matcher.replaceAll. It happen in the critical path, i.e. lookUpKey when read.
> we need not to do this when getFromProtoBuf, So we can avoid normalize by
> check the last character.
> !screenshot-3.png!
> !screenshot-4.png!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
