You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "runzhiwang (Jira)" <ji...@apache.org> on 2020/06/06 13:37:00 UTC
[jira] [Updated] (HDDS-3745) Improve OM and SCM performance with
64% by avoid call getServiceInfo in s3g
[ https://issues.apache.org/jira/browse/HDDS-3745?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
runzhiwang updated HDDS-3745:
-----------------------------
Summary: Improve OM and SCM performance with 64% by avoid call getServiceInfo in s3g (was: Improve OM and SCM performance with 63% by avoid call getServiceInfo in s3g)
> Improve OM and SCM performance with 64% by avoid call getServiceInfo in s3g
> ---------------------------------------------------------------------------
>
> Key: HDDS-3745
> URL: https://issues.apache.org/jira/browse/HDDS-3745
> Project: Hadoop Distributed Data Store
> Issue Type: Sub-task
> Reporter: runzhiwang
> Assignee: runzhiwang
> Priority: Major
> Labels: pull-request-available
> Attachments: om.svg, scm.svg, screenshot-1.png, screenshot-2.png
>
>
> *What's the problem ?*
> I start a ozone cluster with 1000 datanodes and 10 s3gateway, and run two weeks with heavy workload, and perf om and scm.
> 1. From om perf, getServiceList cost 63.75% cpu.
> !screenshot-1.png!
> 2. From scm perf, queryNode come from om::getServiceList cost 33.20% cpu
> !screenshot-2.png!
> *What's the reason ?*
> Now s3g create a client for each request. when create each RpcClient, s3g will call ServiceInfoEx serviceInfoEx = ozoneManagerClient.getServiceInfo(), getServiceInfo will call getServiceList. Then om and scm are busy with getServiceList.
> But s3g does not use the List<ServiceInfo> which got from getServiceList at all.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: ozone-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: ozone-issues-help@hadoop.apache.org