[
https://issues.apache.org/jira/browse/ARTEMIS-3449?focusedWorklogId=643211&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-643211
]
ASF GitHub Bot logged work on ARTEMIS-3449:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 29/Aug/21 17:09
Start Date: 29/Aug/21 17:09
Worklog Time Spent: 10m
Work Description: franz1981 edited a comment on pull request #3711:
URL: https://github.com/apache/activemq-artemis/pull/3711#issuecomment-907827748
@clebertsuconic this is re-addressing
https://issues.apache.org/jira/browse/ARTEMIS-3026 without any exceptional code
path (by accident, because Netty allow to enlarge buffers automatically, pooled
too), although I would like to log some warning, given that handling the
exception case of header > frameSize can cause a severe performance degradation
Another note is about using NettyWritable/NettyReadeable (for @gemmellr
too): Netty::writerIndex/readerIndex used under the hood by such wrappers isn't
free, due to the `ByteBuf::ensureAccessible` checks happening under the hood,
to validate leaks/stale accesses to Netty buffers...
I'm temped to use the ByteBuffer reader variant (that's way more
lightweight) although not sure how much the code quality/readability will be
hit; this is already a very tricky method...
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 643211)
Time Spent: 40m (was: 0.5h)
> Speedup AMQP large message streaming
> ------------------------------------
>
> Key: ARTEMIS-3449
> URL: https://issues.apache.org/jira/browse/ARTEMIS-3449
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Major
> Time Spent: 40m
> Remaining Estimate: 0h
>
> AMQP is using heap ByteBuffer(s) to stream AMQP large messages: given that
> the underline NIO sequential file can both use FileChannel or
> RandomAccessFile (depending if the ByteBuffer used is direct/heap based),
> both approaches would benefit from using direct ByteBuffers pooled from Netty
> and save additional copies (performed by RandomAccessFile) to happen,
> reducing GC too.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
