lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Shai Erera (JIRA)" <j...@apache.org>
Subject [jira] Commented: (LUCENE-2293) IndexWriter has hard limit on max concurrency
Date Thu, 04 Mar 2010 09:24:27 GMT

    [ https://issues.apache.org/jira/browse/LUCENE-2293?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12841140#action_12841140
] 

Shai Erera commented on LUCENE-2293:
------------------------------------

Ok so I think I understand now. You propose to change IW to bind a Thread to a DW, instead
of that being done inside DW. And therefore it will simplify DW's code ... I wonder if that
won't complicate IW code in return? Perhaps we'll gain a lot of simplification on DW, so a
bit of complexity on IW will be ok.

If we do that .. why not renaming DW to SegmentWriter? If each DW will eventually flush its
own Segment, the name would make more sense?

BTW, I was thinking that an application can emulate this sort of thing even today (well ...
to some extent - w/o deletes). It can create an IW for each indexing thread and at the end
call addIndexes. What we'd need to introduce on IW to make it efficient though is something
like addRawIndexes, which will just update the segments file about the new segments, but won't
attempt to merge them and clean deletes out of them.
I think I want this API anyway for being able to add segments faster to an index, if e.g.
you don't care about the merges at the moment ... but that is separate issue.

Then I think what I proposed is more or less the same as you propose, therefore I'm fine with
that approach. When a DW/SW realizes it exhausted its memory pool, it just flushes and new
threads will bind to other DW/SW.

Thanks for the explanation on WaitQueue.

> IndexWriter has hard limit on max concurrency
> ---------------------------------------------
>
>                 Key: LUCENE-2293
>                 URL: https://issues.apache.org/jira/browse/LUCENE-2293
>             Project: Lucene - Java
>          Issue Type: Bug
>          Components: Index
>            Reporter: Michael McCandless
>            Assignee: Michael McCandless
>             Fix For: 3.1
>
>
> DocumentsWriter has this nasty hardwired constant:
> {code}
> private final static int MAX_THREAD_STATE = 5;
> {code}
> which probably I should have attached a //nocommit to the moment I
> wrote it ;)
> That constant sets the max number of thread states to 5.  This means,
> if more than 5 threads enter IndexWriter at once, they will "share"
> only 5 thread states, meaning we gate CPU concurrency to 5 running
> threads inside IW (each thread must first wait for the last thread to
> finish using the thread state before grabbing it).
> This is bad because modern hardware can make use of more than 5
> threads.  So I think an immediate fix is to make this settable
> (expert), and increase the default (8?).
> It's tricky, though, because the more thread states, the less RAM
> efficiency you have, meaning the worse indexing throughput.  So you
> shouldn't up and set this to 50: you'll be flushing too often.
> But... I think a better fix is to re-think how threads write state
> into DocumentsWriter.  Today, a single docID stream is assigned across
> threads (eg one thread gets docID=0, next one docID=1, etc.), and each
> thread writes to a private RAM buffer (living in the thread state),
> and then on flush we do a merge sort.  The merge sort is inefficient
> (does not currently use a PQ)... and, wasteful because we must
> re-decode every posting byte.
> I think we could change this, so that threads write to private RAM
> buffers, with a private docID stream, but then instead of merging on
> flush, we directly flush each thread as its own segment (and, allocate
> private docIDs to each thread).  We can then leave merging to CMS
> which can already run merges in the BG without blocking ongoing
> indexing (unlike the merge we do in flush, today).
> This would also allow us to separately flush thread states.  Ie, we
> need not flush all thread states at once -- we can flush one when it
> gets too big, and then let the others keep running.  This should be a
> good concurrency gain since is uses IO & CPU resources "throughout"
> indexing instead of "big burst of CPU only" then "big burst of IO
> only" that we have today (flush today "stops the world").
> One downside I can think of is... docIDs would now be "less
> monotonic", meaning if N threads are indexing, you'll roughly get
> in-time-order assignment of docIDs.  But with this change, all of one
> thread state would get 0..N docIDs, the next thread state'd get
> N+1...M docIDs, etc.  However, a single thread would still get
> monotonic assignment of docIDs.

-- 
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: java-dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-dev-help@lucene.apache.org


Mime
View raw message