lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steven Rowe (JIRA)" <>
Subject [jira] Commented: (LUCENE-2056) Should NIOFSDir use direct ByteBuffers?
Date Wed, 16 Jun 2010 06:06:24 GMT


Steven Rowe commented on LUCENE-2056:

I used an index built from Reuters line docs, and for the queries, the 92 
English queries from AnswerBus's most recent 100 queries, with question and quotation marks

On Windows Vista 64-bit, 2 CPU cores (Intel Core 2 6600@2.40GHz), Sun JDK 1.6.0_15 64-bit:

|DirectNIOFSDirectory|40|2171|361.73|(5% slower)|

On Windows 7 64-bit, 4 CPU cores (Intel Core i5 750 @ 2.67 GHz), Sun JDK 1.6.0_20 64-bit:

|DirectNIOFSDirectory|40|2658|442.61|(4% slower)|

> Should NIOFSDir use direct ByteBuffers?
> ---------------------------------------
>                 Key: LUCENE-2056
>                 URL:
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: Store
>            Reporter: Michael McCandless
>            Priority: Minor
>         Attachments: LUCENE-2056.patch
> I'm trying to test NRT performance, and noticed when I dump the thread stacks that the
darned threads often seem to be in {{java.nio.Bits.copyToByteArray(Native Method)}}... so
I wondered whether we could/should use direct ByteBuffers, and whether that would gain performance
in general.  We currently just use our own byte[] buffer via BufferedIndexInput.
> It's hard to test since it's likely platform specific, but if it does result in gains
it could be an easy win.

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