lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael McCandless (JIRA)" <>
Subject [jira] Commented: (LUCENE-1604) Stop creating huge arrays to represent the absense of field norms
Date Mon, 27 Apr 2009 12:33:30 GMT


Michael McCandless commented on LUCENE-1604:

I tested this change on a Wikipedia index, with query "1", on a field
that has norms.

On Linux, JDK 1.6.0_13, I can see no performance difference (both get
7.2 qps, best of 10 runs).

On Mac OS X 10.5.6, I see some difference (13.0 vs 12.3, best of 10
runs), but given quirkiness I've seen on OS X's results not matching
other platforms, I think we can disgregard this.

Also, given the performance gain one sees when norms are disabled, I
think this is net/net a good change.

We'll leave the default as false (for back compat), but this setting
is deprecated with a comment that in 3.0 it hardwires to true.

> Stop creating huge arrays to represent the absense of field norms
> -----------------------------------------------------------------
>                 Key: LUCENE-1604
>                 URL:
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: Index
>    Affects Versions: 2.9
>            Reporter: Shon Vella
>            Assignee: Michael McCandless
>            Priority: Minor
>             Fix For: 2.9
>         Attachments: LUCENE-1604.patch, LUCENE-1604.patch, LUCENE-1604.patch
> Creating and keeping around huge arrays that hold a constant value is very inefficient
both from a heap usage standpoint and from a localility of reference standpoint. It would
be much more efficient to use null to represent a missing norms table.

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

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message