You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@trafficserver.apache.org by "weijin (JIRA)" <ji...@apache.org> on 2011/08/30 11:55:37 UTC

[jira] [Updated] (TS-931) cluster latency too high, about 24ms

     [ https://issues.apache.org/jira/browse/TS-931?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

weijin updated TS-931:
----------------------

    Attachment: cluster_performance.patch

This patch will reduce the cluster response time. In my test, cluster memory cache hit:
300qps(9.9ms~10.2ms); 600qps(10.5ms~12ms). I did not test the rt under very high pressure.

> cluster latency too high, about 24ms
> ------------------------------------
>
>                 Key: TS-931
>                 URL: https://issues.apache.org/jira/browse/TS-931
>             Project: Traffic Server
>          Issue Type: Improvement
>          Components: Clustering
>    Affects Versions: 3.1.1, 3.0.1
>         Environment: cluster type 1, content hashing
>            Reporter: Zhao Yongming
>            Assignee: weijin
>             Fix For: 3.1.1
>
>         Attachments: cluster_performance.patch
>
>
> when we enable full clustering, the performance is unacceptable, it add about 24ms per request. we'd expect <10ms latency  for each cluster request.
> this issue is for weijin :D

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira