lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adrien Grand (JIRA)" <>
Subject [jira] [Commented] (SOLR-2548) Multithreaded faceting
Date Tue, 14 Jun 2011 21:48:47 GMT


Adrien Grand commented on SOLR-2548:

Hoss Man,

Regarding the best value of the number of threads to spawn based on the number of CPUs and
the traffic, one could imagine to decide whether to spawn a new thread or to run the task
in the current thread based on the load of the server. This way, servers under high traffic
would run every request in a single thread (maximizing throughput) whereas servers under low
traffic would be able to use every processor in order to minimize response time. The load
is easily retrievable in Java 6 using OperatingSystemMXBean, I don't know if it is possible
in a non OS-specific way in Java 5.

I don't really understand what you mean by "if you really care about parallelizing faceting,
you probably wouldn't want some other intensive component starving out the thread pool". Do
you mean that you would expect some requests to be run slower with every component using a
global thread pool than with a single thread pool dedicated to facets?

Yonik, why would you want to limit the number of threads on a per-request basis, if enough
CPUs are available?

> Multithreaded faceting
> ----------------------
>                 Key: SOLR-2548
>                 URL:
>             Project: Solr
>          Issue Type: Improvement
>          Components: search
>    Affects Versions: 3.1
>            Reporter: Janne Majaranta
>            Priority: Minor
>              Labels: facet
>         Attachments: SOLR-2548.patch, SOLR-2548_for_31x.patch
> Add multithreading support for faceting.

This message is automatically generated by JIRA.
For more information on JIRA, see:


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

View raw message