activemq-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Francesco Nigro (JIRA)" <>
Subject [jira] [Commented] (ARTEMIS-905) JCtools ConcurrentMap replacement
Date Thu, 05 Jan 2017 06:17:58 GMT


Francesco Nigro commented on ARTEMIS-905:

Good libraries, but no one of them provide a concurrent map replacement AFAIK, this is one
of the reasons behind the choice of JCtools.
In addition to that, JCtools has the fastest implementations of lock-free/wait-free queues;
the next step IMHO will be to test if all the ConcurrentLinkedQueues/ArrayBlockingQueues used
in Artemis could be replaced by these ones, hence we'll kill two birds with one stone (poor
birds!) :)

> JCtools ConcurrentMap replacement
> ---------------------------------
>                 Key: ARTEMIS-905
>                 URL:
>             Project: ActiveMQ Artemis
>          Issue Type: Improvement
>            Reporter: Francesco Nigro
>            Priority: Minor
> The OSS project JCTools (already used in Netty core) provides primitive (long) and normal
implementations of ConcurrentMap with better scalability, lower memory footprint (garbage
collector wise too) and equals to better performances than the vanilla's ConcurrentHashMap.
> It worth to replace every ConcurrentHashMap usage with the JCtools version.
> [Here|] there
is an old but valid overview of the map and on JCtools are provided [JMH benchmarks|]
to evaluate the performance of the implementation.

This message was sent by Atlassian JIRA

View raw message