[
https://issues.apache.org/jira/browse/ARTEMIS-2224?focusedWorklogId=185937&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-185937
]
ASF GitHub Bot logged work on ARTEMIS-2224:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 16/Jan/19 18:49
Start Date: 16/Jan/19 18:49
Worklog Time Spent: 10m
Work Description: franz1981 commented on issue #2494: ARTEMIS-2224 Reduce
contention on LivePageCacheImpl
URL: https://github.com/apache/activemq-artemis/pull/2494#issuecomment-454895122
@qihongxu thanks! I need to put a couple do logs to check how is th
behaviour with several consumers and check if live page queries are happening
with single-threaded semantic ie are just limited by the efficiency of the
query and can't scale.
I'm surprised about getValidLastIndex(): it should be the bottleneck (if the
paged message query is the caller) only when the cache is empty (no messages)
or the index is == cache size. If there is a cache hit I expect that getChunkOf
or the volatile get on the message slot to be costy.
About the test is a good idea: having 1/200/400 consumers can help to
understand the scaling, but I suppose that your intuitions are right here.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 185937)
Time Spent: 3h 10m (was: 3h)
> Reduce contention on LivePageCacheImpl
> --------------------------------------
>
> Key: ARTEMIS-2224
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2224
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Components: Broker
> Affects Versions: 2.7.0
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Major
> Time Spent: 3h 10m
> Remaining Estimate: 0h
>
> Has been measured that LIvePageCacheImpl operations are a source of
> contention on producer side while paging.
> This contention decrease the scalability of the broker in an evident way
> while using topics, because the page cache is been accessed concurrently by
> several producers to ack transactions while the messages are being appended.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
