[
https://issues.apache.org/jira/browse/SOLR-17841?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18094021#comment-18094021
]
Renato Haeberli commented on SOLR-17841:
----------------------------------------
Hi [~arupchauhan]
I was running your modified test locally, judging by the results I am not sure
if it actually runs multi threaded. Have a look at my version:
[https://github.com/renatoh/solr/blob/SOLR-1784-Investigate_multithreade_search_performance_bottlenecks/solr/benchmark/src/java/org/apache/solr/bench/search/ExitableDirectoryReaderSearch.java]
I belief you need to set to a number higher than 1, e.g. 4
System.setProperty("solr.searchThreads", String.valueOf(searchThreads));
Here my JMH results:
Benchmark (multiThreaded) Mode Cnt
Score Error Units
ExitableDirectoryReaderSearch.testLongQuery false avgt 4
45.102 ± 0.333 ms/op
ExitableDirectoryReaderSearch.testLongQuery true avgt 4
45.228 ± 1.264 ms/op
ExitableDirectoryReaderSearch.testRangeQuery false avgt 4
187.133 ± 13.034 ms/op
ExitableDirectoryReaderSearch.testRangeQuery true avgt 4
202.017 ± 7.411 ms/op
ExitableDirectoryReaderSearch.testShortQuery false avgt 4
3.877 ± 0.304 ms/op
ExitableDirectoryReaderSearch.testShortQuery true avgt 4
3.753 ± 0.290 ms/op
> Investigate multithreaded search performance bottlenecks
> --------------------------------------------------------
>
> Key: SOLR-17841
> URL: https://issues.apache.org/jira/browse/SOLR-17841
> Project: Solr
> Issue Type: Improvement
> Reporter: Ishan Chattopadhyaya
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 10.0
>
> Attachments: profile.html
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Ever since multithreaded search (SOLR-13350) was introduced, there has been
> not much effort to benchmark the performance benefit or investigate the
> performance bottlenecks. Anecdotally, it is not faster than single threaded
> execution.
> Thanks to [~kevinliang01] [~liangkaiwen] [~liangkw16] and [~dsmiley] for
> bringing this up, yesterday on a dense vector working group meeting.
> I'm marking this as a release 10.0 blocker, because it is embarrassing to be
> releasing with such a gaping hole.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
