[
https://issues.apache.org/jira/browse/LUCENE-2662?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Simon Willnauer updated LUCENE-2662:
------------------------------------
Attachment: LUCENE-2662.patch
Next iteration - seems to be very close!
I have applied the following changes:
* introduces a AtomicLong to track bytesUsed in DocumetnsWriter,
TermsHashPerField, ByteRefHash and RecyclingByteBlockAllocator
* Factored out a BytesStartArray class from BytesRefHash that manages the
int[] holding the bytesStart offsets. TermsHashPerField subclasses and manages
the ParallelPostingsArray through it.
* remove remaining no-commits
* made RecyclingbyteBlockAllocator synced by default (we use synchronized
methods for it now)
I run a quick Wikipedia 100k docs benchmark against trunk vs. LUCENE-2662 and
the results are promising.
|version|rec/sec|elapsed sec|avgUsedMem|
|LUCENE-2662|717.30|139.41|536,682,592|
|trunk| 682.66|146.49|546,065,344|
I will run the 10M benchmark once I get back to this.
> BytesHash
> ---------
>
> Key: LUCENE-2662
> URL: https://issues.apache.org/jira/browse/LUCENE-2662
> Project: Lucene - Java
> Issue Type: Improvement
> Components: Index
> Affects Versions: Realtime Branch, 4.0
> Reporter: Jason Rutherglen
> Assignee: Simon Willnauer
> Priority: Minor
> Fix For: Realtime Branch, 4.0
>
> Attachments: LUCENE-2662.patch, LUCENE-2662.patch, LUCENE-2662.patch,
> LUCENE-2662.patch, LUCENE-2662.patch
>
>
> This issue will have the BytesHash separated out from LUCENE-2186
--
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: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
