[
https://issues.apache.org/jira/browse/ARTEMIS-1185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16068938#comment-16068938
]
ASF GitHub Bot commented on ARTEMIS-1185:
-----------------------------------------
Github user franz1981 commented on the issue:
https://github.com/apache/activemq-artemis/pull/1295
@clebertsuconic It is able to capture samples from a running live server,
but the server must be first started with the proper configuration.
It is simply not possible to make it light as it is with other options:
that's why I've build the MulticastBuffer to not backpressure the broker when
the sampling profiler is not reading data.
Making it differently will at least add branch mispredictions and JVM
compile deoptimizations in the hot paths.
Hence you can let a broker be configured to emit performance counters data
with no performance impact, making it suitable to be used in production too.
It is really like most perf profilers work: if you want profiling you need
at least some debug info to capture meaningful data.
> Inter-Process Journal Sampler Profiler + CLI command
> ----------------------------------------------------
>
> Key: ARTEMIS-1185
> URL: https://issues.apache.org/jira/browse/ARTEMIS-1185
> Project: ActiveMQ Artemis
> Issue Type: New Feature
> Components: Broker
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Minor
>
> It provides a sampling profiler on buffered ASYNCIO/NIO based journals.
> The profiling has a minimal cost in term of CPU time for each sample (the
> dominant costs are System.nanoTime() and a single cache line invalidation)
> and total memory footprint (~OS page size in bytes).
> A proper CLI command activates a sampler to collect (ie CSV) the profiled
> data, showing the precision of the sampling: data loss is not considered a
> failure condition.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
