lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 34930] - IndexWriter.maybeMergeSegments() takes lots of CPU resources
Date Tue, 17 May 2005 02:19:57 GMT
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG·
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=34930>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND·
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=34930





------- Additional Comments From psmith@apache.org  2005-05-17 04:19 -------
Doug, after more testing, you are correct that tweaking the maxBuffered from
10,000 to 5000 gives slightly better performance.

However I believe this is because with maxBufferedDocs==5000, this loop counter
is reset more frequently, and so suffers less of the CPU drain.  

I ran more tests using the hacked version using both 5000 and 10000, and the
hacked version still runs better.  I'll attach an Excel sheet with the results,
but I only had time to run it over 5 samples each run, which is not exactly
statistically significant, but hopefully still correct.

As you can see, deferring this loop enables larger maxBuffered to gain ahead of
smaller maxBufferedDocs.

As always with these tests there are transient factors that affect things
(getting DB results for one thing).

-- 
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

---------------------------------------------------------------------
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