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

Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2015/04/13 23:06:14 UTC

[jira] [Updated] (HBASE-12116) Hot contention spots; writing

     [ https://issues.apache.org/jira/browse/HBASE-12116?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-12116:
--------------------------
    Attachment: Screen Shot 2015-04-13 at 2.03.05 PM.png
                perf.write3.svg
                perf.write4.svg

A few flame graphs doing sequential writes of 1M by 50 clients run on 5 different machines. Cells are zipfian 0-1024 bytes in size. Source is perf recording.

Mostly GC. Next is CRC and unsafe compareto from memstore.

Attached is associated flight recorder taken just after. It fingers our sending of exceptions back to client as expensive beyond compareto and crc.

> Hot contention spots; writing
> -----------------------------
>
>                 Key: HBASE-12116
>                 URL: https://issues.apache.org/jira/browse/HBASE-12116
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>         Attachments: 12116.checkForReplicas.txt, 12116.stringify.and.cache.scanner.maxsize.txt, 12116.txt, Screen Shot 2014-09-29 at 5.12.51 PM.png, Screen Shot 2014-09-30 at 10.39.34 PM.png, Screen Shot 2015-04-13 at 2.03.05 PM.png, perf.write3.svg, perf.write4.svg
>
>
> Playing with flight recorder, here are some write-time contentious synchronizations/locks (picture coming)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)