[
https://issues.apache.org/jira/browse/SOLR-3583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14056433#comment-14056433
]
Brett Lucey commented on SOLR-3583:
-----------------------------------
bq. how much performance improvement is there with specifying the start index
for that lookup? It seems error-prove as the list is getting populated
elsewhere and no real control on the order is imposed.
I think the hope was that it would turn going to the NamedList into an O(1)
operation instead of an O(n) operation. Some of the queries we are using hit
100+ shards. Before the SOLR-2894 rewrite, these queries could take over a
minute. Now we have it down to around a second. I definitely agree that it
could be error-prone, especially when combining multiple patches. There's
currently ongoing discussion on the other Jira issue as we get closer and
closer to getting SOLR-2894 committed to the main source tree. If you feel
this is especially concerning, it would be worth voicing over there to see if
someone has some free time to look at making a change to that. I definitely
think we want to keep it O(1) if possible to avoid making it slower, the
question is just what the best way to do it is.
> Percentiles for facets, pivot facets, and distributed pivot facets
> ------------------------------------------------------------------
>
> Key: SOLR-3583
> URL: https://issues.apache.org/jira/browse/SOLR-3583
> Project: Solr
> Issue Type: Improvement
> Reporter: Chris Russell
> Priority: Minor
> Labels: newbie, patch
> Fix For: 4.9, 5.0
>
> Attachments: SOLR-3583.patch, SOLR-3583.patch, SOLR-3583.patch,
> SOLR-3583.patch, SOLR-3583.patch, SOLR-3583.patch, SOLR-3583.patch
>
>
> Built on top of SOLR-2894, this patch adds percentiles and averages to
> facets, pivot facets, and distributed pivot facets by making use of range
> facet internals.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
