Mailing-List: contact dev-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@couchdb.apache.org
Date: Wed, 9 Mar 2011 04:45:59 +0000 (UTC)
From: "Jan Lehnardt (JIRA)" <jira@apache.org>
To: dev@couchdb.apache.org
Message-ID: 
 <330192825.7065.1299645959519.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <823419600.3437.1299550079623.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] Commented: (COUCHDB-1084) Remove unnecessary btree lookup
 inside couch_db_updater
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/COUCHDB-1084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13004366#comment-13004366 ] 

Jan Lehnardt commented on COUCHDB-1084:
---------------------------------------

Definitely more testing :)

Randal, if you are comparing numbers from my tests note that the first was 300s the other 200s (IIRC).

I think though your description is apt, Damien says on SSDs fsyncs are faster and cheaper, so we see the benefits of the patch earlier. On spinning media, the waiting-for-fsync trumps most of the time so the improvements in the now relatively smaller btree update code path are not as significant (for the small data set until we actually save seeks in the large data set)

> Remove unnecessary btree lookup inside couch_db_updater
> -------------------------------------------------------
>
>                 Key: COUCHDB-1084
>                 URL: https://issues.apache.org/jira/browse/COUCHDB-1084
>             Project: CouchDB
>          Issue Type: Improvement
>          Components: Database Core
>    Affects Versions: 1.2
>            Reporter: Damien Katz
>            Assignee: Damien Katz
>         Attachments: remove_btree_lookup.patch
>
>
> The CouchDB update process has an unnecessary btree lookup, where it reads the values in bulks, checks for conflicts, writes the docs to disk, updates the values appropriately and writes them to the btree out in a second step. It's possible to avoid this second step, and instead do all the checking, doc writing and value transformation in a single btree lookup, thereby reducing the number of btree traversals and disk IO.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira