hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-16156) In write heavy scenario creating in memory flushes leads to contention
Date Fri, 01 Jul 2016 07:11:10 GMT

    [ https://issues.apache.org/jira/browse/HBASE-16156?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15358534#comment-15358534
] 

Anoop Sam John commented on HBASE-16156:
----------------------------------------

{code}
if (!active.isEmpty()) {
      long delta = DEEP_OVERHEAD_PER_PIPELINE_ITEM - DEEP_OVERHEAD;
      active.setSize(active.getSize() + delta);
      pipeline.pushHead(active);
      resetCellSet();
    }
{code}
Why we need block all the writes for push also? 
We can create new active and only when the new CSLM is assigned to active we need lock.  

> In write heavy scenario creating in memory flushes leads to contention
> ----------------------------------------------------------------------
>
>                 Key: HBASE-16156
>                 URL: https://issues.apache.org/jira/browse/HBASE-16156
>             Project: HBase
>          Issue Type: Sub-task
>    Affects Versions: 2.0.0
>            Reporter: ramkrishna.s.vasudevan
>            Assignee: ramkrishna.s.vasudevan
>             Fix For: 2.0.0
>
>
> In write heavy cases, the inmemory flushes blocks the write because it takes a update
lock. Dumps show that it is causing heavy contention and leads to memstore filling up and
so intern blocks the flush from happening sooner. This JIRA is to discuss optimal settings
and then make suitable changes.



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

Mime
View raw message