hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14480) Small optimization in SingleByteBuff
Date Thu, 24 Sep 2015 16:27:04 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14480?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14906594#comment-14906594

stack commented on HBASE-14480:

Patch LGTM.

Is this code path excercised in any tests?

> Small optimization in SingleByteBuff
> ------------------------------------
>                 Key: HBASE-14480
>                 URL: https://issues.apache.org/jira/browse/HBASE-14480
>             Project: HBase
>          Issue Type: Sub-task
>          Components: regionserver, Scanners
>    Affects Versions: 2.0.0
>            Reporter: Anoop Sam John
>            Assignee: Anoop Sam John
>             Fix For: 2.0.0
>         Attachments: HBASE-14480.patch
> We do read from this using Unsafe based reads on underlying BB.  The UnsafeAccess getters
having 2 branches of code for HBB and DBB and exact memory offset calculation varies. We can
do an optimize here.  We can initialize the  reference object and exact memory offset for
Unsafe reads while creating the SBB.  We will do doing the reads on this SBB many a times.
Each time when we do read we can use
> Unsafe.getXXX(Object, long)
> I am able to get a slight adv of 2% on this change. Small change and worth getting this
slight gain.

This message was sent by Atlassian JIRA

View raw message