Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Message-ID: <8003996.118371283362196927.JavaMail.jira@thor>
Date: Wed, 1 Sep 2010 13:29:56 -0400 (EDT)
From: "stack (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Subject: [jira] Resolved: (HBASE-2510) [performance] make hbase splits run
 even faster still
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HBASE-2510?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack resolved HBASE-2510.
--------------------------

    Fix Version/s: 0.90.0
       Resolution: Fixed

Fixed by hbase-2692; it does what this issue asks: i.e. open daughters on the parents regionserver.  Will open new issue if other ideas about making splits faster still.

> [performance] make hbase splits run even faster still
> -----------------------------------------------------
>
>                 Key: HBASE-2510
>                 URL: https://issues.apache.org/jira/browse/HBASE-2510
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: stack
>             Fix For: 0.90.0
>
>
> hbase-1892 made splits faster.  This issue is about making them faster still.
> Chatting w/ some bright lads down south, and because Ryan has been going on about it for a year now, on split, we should just open the daughter regions on the parent's RegionServer.  Don't even bother telling master about the split just happened.  Let the balancer rectify uneven region distribution.  Requires smarter load balancer first, one that can at least recognize the special case of a fresh unloaded cluster with bulk upload running.  In this case it would balance differently than it would on a laden cluster.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.