incubator-couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jann Horn (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (COUCHDB-1791) Initial index of views is highly fragmented
Date Thu, 16 May 2013 11:01:19 GMT

    [ https://issues.apache.org/jira/browse/COUCHDB-1791?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13659432#comment-13659432
] 

Jann Horn commented on COUCHDB-1791:
------------------------------------

OT: I think you want `==` instead of `=`?

[sorry to the dev mailinglist for the spam, didn't realize that replies to JIRA mails don't
go back into JIRA]
                
> Initial index of views is highly fragmented
> -------------------------------------------
>
>                 Key: COUCHDB-1791
>                 URL: https://issues.apache.org/jira/browse/COUCHDB-1791
>             Project: CouchDB
>          Issue Type: Improvement
>          Components: JavaScript View Server
>            Reporter: James Howe
>
> function(doc) {
>   if (doc.a = 'b') { emit(doc._id, null); }
> }
> This view, running on CouchDB 1.2.1, takes 16 minutes to index from scratch on an inactive
database, creating a 2.5GB file.
> After compaction, the file is then only 157MB.
> Can we do something about this waste of disk I/O, such that the initial index is smaller
to start with?

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

Mime
View raw message