lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Johan Stuyts" <>
Subject RE: [jira] Commented: (LUCENE-639) [PATCH] Slight performance improvement for readVInt() of IndexInput
Date Wed, 02 Aug 2006 11:16:35 GMT
> If people know that a potentially significant performance gain is
> available, and an easy mechanism is provided for them to give 
> you stats
> about their index, i'm sure you'd get more then a few responses.

The performance gain is not significant. I am able to shave of about 4 %
of an algorithm which takes 13 % of the total run time. This results in
a gain of about 0.5 % in my program, which is a prototype and not a real
application. I just focused on 'readVInt()' because my profiler
indicated it was a 'hot spot'.

I thought my isolated test showed what the optimal algorithm was no
matter what the distribution of the vints is. Whether the Lucene team
decides to change the current algorithm to something else or not, it's
fine with me.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message