[
https://issues.apache.org/jira/browse/FELIX-4638?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14140446#comment-14140446
]
Bob Paulin commented on FELIX-4638:
-----------------------------------
Carsten, thank you for the review. I was thinking that if there are multiple
handlers listening to an event even if they ignored blacklisting there might be
a performance gain by executing them in different threads. I did not however
do any testing around this theory :). I expect that there may be some
threshold where it's worth it to send the handlers into the thread pool. This
threshold would likely vary by the number of handlers and the duration of the
handlers. While calculating the number of handlers is straightforward,
calculating duration is tricky so it might be worth looking into adding a flag
to the handlers indicating it should be run in the pool. This might provide
some additional flexibility to your idea of executing the blacklisted or
non-blacklisted with separate approaches. I'll see if I can setup some tests
and get some data. Either way I think it makes sense to add back in the code
to run the handlers that ignore blacklisting in the same thread.
> Event Admin - Less locking on event handler timing
> --------------------------------------------------
>
> Key: FELIX-4638
> URL: https://issues.apache.org/jira/browse/FELIX-4638
> Project: Felix
> Issue Type: Improvement
> Components: Event Admin
> Reporter: Bob Paulin
> Assignee: Carsten Ziegeler
> Priority: Minor
> Fix For: eventadmin-1.4.4
>
> Attachments: FELIX-4638.patch, FELIX-4638b.patch
>
>
> The locking that is done for the blacklist timing seems to degrade
> performance significantly Felix is under stress with multiple firing handler
> callbacks for each event. I'd like to discuss an alternative approach with
> less locking that still guarantees proper event ordering per the OSGi spec.
> Basically instead of using the CyclicBarriers (Rendezvous) on a per handler
> basis we could use a count down latch to only await after all handlers are
> complete. Then instead of using a stopwatch based timer the JMX Current
> Thread Cpu Time which counts CPU time for the application code and any IO
> performed on it's behalf filtering out time context switching between threads
> to provide proper blacklisting.
> Here are my test results.
> Baseline(Event Admin 1.4.2):
> 15 Threads
> 100000 Async Events per Thread
> 7 Active Handlers per Event
> For a total of 10500000 Handler Events Executed in 40000 - 45000ms
> With the same parameters above but a CountDownLatch I see the execution time
> drop to around 25000ms. The improvement is noticeable because the stress
> test includes 7 active handlers per event. The improvement is less
> noticeable with applications that only register one or 2 handlers for an
> active event such as in the PerformanceTestIT. Thoughts on changing how this
> locking occurs? Concerns with using the JMX timings?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
