lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael McCandless (JIRA)" <>
Subject [jira] Commented: (LUCENE-1593) Optimizations to TopScoreDocCollector and TopFieldCollector
Date Fri, 24 Apr 2009 19:44:30 GMT


Michael McCandless commented on LUCENE-1593:

bq. If it equals, it's also rejected, since now that we move to returning docs in order, it
is assumed that this doc's doc Id is greater than whatever is in the queue, and so it's rejected.

Argh, you are right... so the approach will fail if any of the first topN (eg 10) hits have
a field value equal to the sentinel value.  I guess we could do two separate passes: a "startup"
pass (while queue is filling up) and a "the rest" pass that knows the queue is full.  But
that's getting rather ugly; probably we should leave this optimization to source code specialization.

bq.  I can do the same for the FieldDoc.fields() value, in case one of the fields is FIELD_DOC.

Excellent!  Let's just do this as part of this issue.

bq. I must say I still didn't fully understand what do you mean here.

I also did not understand how BooleanScorer works until Doug explained it (see the comment
at the top).

Right now it gathers hits of each clause in a reversed linked list, which it then makes a
2nd pass to collect.  So the Collector will see docIDs in reverse order for that clause. 
I thought we could simply fix the linking to be forward and we'd have docIDs in order.  But
that isn't quite right because any new docIDs hit by the 2nd clause will be inserted at the
end of the linked list, out of order.

> Optimizations to TopScoreDocCollector and TopFieldCollector
> -----------------------------------------------------------
>                 Key: LUCENE-1593
>                 URL:
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: Search
>            Reporter: Shai Erera
>             Fix For: 2.9
> This is a spin-off of LUCENE-1575 and proposes to optimize TSDC and TFC code to remove
unnecessary checks. The plan is:
> # Ensure that IndexSearcher returns segements in increasing doc Id order, instead of
> # Change TSDC and TFC's code to not use the doc id as a tie breaker. New docs will always
have larger ids and therefore cannot compete.
> # Pre-populate HitQueue with sentinel values in TSDC (score = Float.NEG_INF) and remove
the check if reusableSD == null.
> # Also move to use "changing top" and then call adjustTop(), in case we update the queue.
> # some methods in Sort explicitly add SortField.FIELD_DOC as a "tie breaker" for the
last SortField. But, doing so should not be necessary (since we already break ties by docID),
and is in fact less efficient (once the above optimization is in).
> # Investigate PQ - can we deprecate insert() and have only insertWithOverflow()? Add
a addDummyObjects method which will populate the queue without "arranging" it, just store
the objects in the array (this can be used to pre-populate sentinel values)?
> I will post a patch as well as some perf measurements as soon as I have them.

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:
For additional commands, e-mail:

View raw message