jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ard Schrijvers (JIRA)" <j...@apache.org>
Subject [jira] Commented: (JCR-1222) Index nodes in parallel
Date Tue, 20 Nov 2007 14:11:43 GMT

    [ https://issues.apache.org/jira/browse/JCR-1222?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12543896
] 

Ard Schrijvers commented on JCR-1222:
-------------------------------------

This is great Marcel! An impressive performance gain. 

FYI : I lately tried searching in the multiIndex structure with the lucene ParallelMultiSearcher
but I merely found a decrease in speed (even on large indexes). I think it was because of
filesystem access which might get worse (random access) in using parallel searching on indexes
on a single FS (must admit I am guessing a little)

> Index nodes in parallel
> -----------------------
>
>                 Key: JCR-1222
>                 URL: https://issues.apache.org/jira/browse/JCR-1222
>             Project: Jackrabbit
>          Issue Type: Improvement
>          Components: indexing
>            Reporter: Marcel Reutegger
>            Priority: Minor
>             Fix For: 1.4
>
>         Attachments: diagram.jpg
>
>
> CPUs with multiple cores are now standard and Jackrabbit should make use of it where
it makes sense. Analyzing content while a node is indexed is quite costly, but can be broken
easily into task that can be executed in parallel. E.g. index multiple nodes in parallel.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message