lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zhang, Lisheng" <Lisheng.Zh...@broadvision.com>
Subject Lucene search performance: linear?
Date Tue, 05 Dec 2006 02:49:44 GMT
Hi,

I indexed first 220,000, all with a special keyword, I did a simple
query and only fetched 5 docs, with Hits.length()=220,000.

Then I indexed 440,000 docs, with the same keyword, query it
again and fetched a few docs, with Hits.length(0=440,000.

I found that search time is about linear: 2nd time is about 2 times
longer than 1st query. I would like to understand:

Does the linear relation come from score calculation, since we
have to calculate score one by one? Or other reason?

If we have B-tree index I would naively expect a better scalibility?

Thanks very much for your helps, 

Lisheng

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


Mime
View raw message