Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Thu, 12 Sep 2013 04:00:52 +0000 (UTC)
From: "Lars Hofhansl (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.12667730.1378782393942.120942.1378958452752@arcas>
In-Reply-To: <JIRA.12667730.1378782393942@arcas>
References: <JIRA.12667730.1378782393942@arcas>
Subject: [jira] [Commented] (HBASE-9488) Improve performance for small scan
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


    [ https://issues.apache.org/jira/browse/HBASE-9488?page=3Dcom.atlassian=
.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D1376=
5150#comment-13765150 ]=20

Lars Hofhansl commented on HBASE-9488:
--------------------------------------

I don't see how we can reliably default this. (I have the same problem for =
the parallel scanning stuff). In order to do that we need statistics (we're=
 actually working on this here - on the backburner).

The only scans we know for sure will be small are META scans.
               =20
> Improve performance for small scan
> ----------------------------------
>
>                 Key: HBASE-9488
>                 URL: https://issues.apache.org/jira/browse/HBASE-9488
>             Project: HBase
>          Issue Type: Improvement
>          Components: Client, Performance, Scanners
>            Reporter: chunhui shen
>            Assignee: chunhui shen
>             Fix For: 0.98.0, 0.94.13
>
>         Attachments: hbase-9488-94-v3.patch, HBASE-9488-trunk.patch, HBAS=
E-9488-trunkV2.patch, HBASE-9488-trunkV3.patch, HBASE-9488-trunkV4.patch, H=
BASE-9488-trunkV4.patch, mergeRpcCallForScan.patch, test results.jpg
>
>
> review board:
> https://reviews.apache.org/r/14059/
> *Performance Improvement*
> Test shows about 1.5~3X improvement for small scan where limit<=3D50 unde=
r cache hit ratio=3D100%.
> See more performance test result from the picture attachment
> *Usage:*
> Scan scan =3D new Scan(startRow,stopRow);
> scan.setSmall(true);
> ResultScanner scanner =3D table.getScanner(scan);
> Set the new 'small' attribute as true for scan object, others are the sam=
e
> Now, one scan operation would call 3 RPC at least:
> openScanner();
> next();
> closeScanner();
> I think we could reduce the RPC call to one for small scan to get better =
performance
> Also using pread is better than seek+read for small scan (For this point,=
 see more on HBASE-7266)
> Implements such a small scan as the patch, and take the performance test =
as following:
> a.Environment=EF=BC=9A
> patched on 0.94 version
> one regionserver;=20
> one client with 50 concurrent threads;
> KV size:50/100;
> 100% LRU cache hit ratio;
> Random start row of scan
> b.Results:
> See the picture attachment
> =20

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrato=
rs
For more information on JIRA, see: http://www.atlassian.com/software/jira