[
https://issues.apache.org/jira/browse/LUCENE-5476?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13917165#comment-13917165
]
Shai Erera commented on LUCENE-5476:
------------------------------------
About the patch:
* SampledDocs should be static class, to not carry over the FC reference. We've
been bitten by such things in the past and I don't see a good reason to do it
here.
* The 'sampling' package may be an overkill? if we want to do it, need to add
package.html. But I think you can wrap them all in a SFC now (SampledDocs and
SamplingParams as static classes) as it's not a lot of code.
* Could you take a look at the old tests and see if they can be ported to the
new API?
About the sampling itself, Rob referred me to this link
http://dmurphy747.wordpress.com/2011/03/23/xorshift-vs-random-performance-in-java/
where a simple implementation for PRNG is used, following XOR-Shift method.
Could you take a look and see if it can be applied here instead of using Java's
Random. According to the link, it performs 7 times faster...
> Facet sampling
> --------------
>
> Key: LUCENE-5476
> URL: https://issues.apache.org/jira/browse/LUCENE-5476
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Rob Audenaerde
> Attachments: LUCENE-5476.patch, SamplingFacetsCollector.java
>
>
> With LUCENE-5339 facet sampling disappeared.
> When trying to display facet counts on large datasets (>10M documents)
> counting facets is rather expensive, as all the hits are collected and
> processed.
> Sampling greatly reduced this and thus provided a nice speedup. Could it be
> brought back?
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
