[
https://issues.apache.org/jira/browse/LUCENE-2324?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13005177#comment-13005177
]
Jason Rutherglen commented on LUCENE-2324:
------------------------------------------
{quote}Because 1) the RAM efficiency ought to scale up very well, as you see a
given term in more and more docs (hmm, though, maybe not, because from Zipf's
law, half your terms will be singletons no matter how many docs you index), and
2) less merging is required.{quote}
I'm not sure how we handled concurrency on the terms hash before, however with
DWPTs there won't be contention regardless. It'd be nice if we could build 1-2
GB segment's in RAM, I think that would greatly reduce the number merges that
are required downstream. Eg, then there's less need for merging by size, and
most merges would be caused by the number/percentage of deletes. If it turns
out the low DF terms are causing the slowdown, maybe there is a different
hashing system that could be used.
> Per thread DocumentsWriters that write their own private segments
> -----------------------------------------------------------------
>
> Key: LUCENE-2324
> URL: https://issues.apache.org/jira/browse/LUCENE-2324
> Project: Lucene - Java
> Issue Type: Improvement
> Components: Index
> Reporter: Michael Busch
> Assignee: Michael Busch
> Priority: Minor
> Fix For: Realtime Branch
>
> Attachments: LUCENE-2324-SMALL.patch, LUCENE-2324-SMALL.patch,
> LUCENE-2324-SMALL.patch, LUCENE-2324-SMALL.patch, LUCENE-2324-SMALL.patch,
> LUCENE-2324.patch, LUCENE-2324.patch, LUCENE-2324.patch, LUCENE-2324.patch,
> lucene-2324.patch, lucene-2324.patch, test.out, test.out, test.out, test.out
>
>
> See LUCENE-2293 for motivation and more details.
> I'm copying here Mike's summary he posted on 2293:
> Change the approach for how we buffer in RAM to a more isolated
> approach, whereby IW has N fully independent RAM segments
> in-process and when a doc needs to be indexed it's added to one of
> them. Each segment would also write its own doc stores and
> "normal" segment merging (not the inefficient merge we now do on
> flush) would merge them. This should be a good simplification in
> the chain (eg maybe we can remove the *PerThread classes). The
> segments can flush independently, letting us make much better
> concurrent use of IO & CPU.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
