Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Mon, 27 Oct 2014 19:27:34 +0000 (UTC)
From: "Tsz Wo Nicholas Sze (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12736287.1408829728000.346797.1414438054073@Atlassian.JIRA>
In-Reply-To: <JIRA.12736287.1408829728000@Atlassian.JIRA>
References: <JIRA.12736287.1408829728000@Atlassian.JIRA>
 <JIRA.12736287.1408829728790@arcas>
Subject: [jira] [Commented] (HDFS-6934) Move checksum computation off the
 hot path when writing to RAM disk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HDFS-6934?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14185661#comment-14185661 ] 

Tsz Wo Nicholas Sze commented on HDFS-6934:
-------------------------------------------

+1 the 2.6 patch looks good.

> There was a very slight merge conflict on merging to branch-2.6. ...

In such case, we should cherry-pick the trunk commit and resolve the merge manually.

> Move checksum computation off the hot path when writing to RAM disk
> -------------------------------------------------------------------
>
>                 Key: HDFS-6934
>                 URL: https://issues.apache.org/jira/browse/HDFS-6934
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: datanode, hdfs-client
>            Reporter: Arpit Agarwal
>            Assignee: Chris Nauroth
>         Attachments: HDFS-6934-branch-2.6.5.patch, HDFS-6934.3.patch, HDFS-6934.4.patch, HDFS-6934.5.patch, h6934_20141003b.patch, h6934_20141005.patch
>
>
> Since local RAM is considered reliable we can avoid writing checksums on the hot path when replicas are being written to a local RAM disk.
> The checksum can be computed by the lazy writer when moving replicas to disk.


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