[
https://issues.apache.org/jira/browse/ARTEMIS-1383?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16151503#comment-16151503
]
ASF GitHub Bot commented on ARTEMIS-1383:
-----------------------------------------
Github user franz1981 commented on the issue:
https://github.com/apache/activemq-artemis/pull/1505
@clebertsuconic Please do not merge it, because:
- need to add an additional test on the priority version
- need to add performance tests to compare it with the linked list one
- need to configure it better (mabe using the window size?) when used into
the core protocol
@tabish121 @gemmellr As far as you know it could be of any use in AMQP too?
Currently, the benchmarks against the latest snapshot show an improvement
for the core client of about ~10% more throughput without producing near to 0
garbage under load.
As soon as I'll provide any JMH benchmark vs the linked list one I'll
update this PR.
> Improved Priority queue
> -----------------------
>
> Key: ARTEMIS-1383
> URL: https://issues.apache.org/jira/browse/ARTEMIS-1383
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Components: Broker
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
>
> The original PriorityLinkedList implementation is based on a double linked
> list implementation that suffer of:
> * fragmentation along the heap
> * pointer chasing due to the presence of nodes
> * allocation heavy (ie each add operation forces allocation of nodes)
> * high hidden (ie the nodes) memory footprint that lead to wrong memory
> estimations
> It is possible to provide a specialized chunked implementation that can
> address all these issues while providing a better performance (throughput,
> latency and memory footprint wise).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
