[
https://issues.apache.org/jira/browse/LUCENE-4515?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13488947#comment-13488947
]
Michael McCandless commented on LUCENE-4515:
--------------------------------------------
{quote}
1. reduce the GC load on MemIndex if it is used extensively - that is why I
moved to ByteBlockPool / IntBlockPool
2. control memory consumption / buffering in a very flexible way. When buffers
are reused and how many of them.
{quote}
I think the current patch, minus passing allocator to MemoryIndex, solves #1
very well?
I think we should open a new Jira issue for #2 since it's apparently somewhat
controversial. In that issue I'd really like to understand "the memory
consumption / buffering in a very flexible way / when buffers are reused and
how many of them" use cases. If it's really an accounting issue (knowing how
many bytes a given MemoryIndex is consuming) maybe we can expose that in other
ways ... eg .sizeInBytes() method, or app passes in a Counter that MemoryIndex
updates w/ how much RAM it thinks it's using, or ... something else?
> Make MemoryIndex more memory efficient
> --------------------------------------
>
> Key: LUCENE-4515
> URL: https://issues.apache.org/jira/browse/LUCENE-4515
> Project: Lucene - Core
> Issue Type: Improvement
> Components: modules/other
> Affects Versions: 4.0, 4.1, 5.0
> Reporter: Simon Willnauer
> Fix For: 4.1, 5.0
>
> Attachments: LUCENE-4515.patch, LUCENE-4515.patch
>
>
> Currently MemoryIndex uses BytesRef objects to represent terms and holds an
> int[] per term per field to represent postings. For highlighting this creates
> a ton of objects for each search that 1. need to be GCed and 2. can't be
> reused.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
