[
https://issues.apache.org/jira/browse/LUCENE-5760?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14031250#comment-14031250
]
ASF subversion and git services commented on LUCENE-5760:
---------------------------------------------------------
Commit 1602531 from [~rcmuir] in branch 'dev/branches/branch_4x'
[ https://svn.apache.org/r1602531 ]
LUCENE-5760: Speed up BufferedIndexInput.randomAccessSlice
> Speed up BufferedIndexInput.randomAccessSlice
> ---------------------------------------------
>
> Key: LUCENE-5760
> URL: https://issues.apache.org/jira/browse/LUCENE-5760
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Robert Muir
> Fix For: 4.9, 5.0
>
> Attachments: LUCENE-5760.patch
>
>
> Today this uses the default implementation, e.g. for readInt(pos):
> {code}
> @Override
> public int readInt(long pos) throws IOException {
> slice.seek(pos);
> return slice.readInt();
> }
> {code}
> But this causes the bounds to be checked twice. Just like we did for MMap, we
> can provide a faster implementation that only checks once: yields ~30%
> speedup.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
