[
https://issues.apache.org/jira/browse/STORM-1632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15208024#comment-15208024
]
ASF GitHub Bot commented on STORM-1632:
---------------------------------------
Github user arunmahadevan commented on the pull request:
https://github.com/apache/storm/pull/1217#issuecomment-200233334
Re-ran the exclamation topology with,
- no anchoring/acking
- TestWordSpout emitting a fixed word without any sleep
- spout, bolt parallelism = 1
- worker = 1
- Loglevel = error, conf.setDebug(false)
Throughput I observed :-
566 K tuples/sec (with eventlogger: nil)
569 K tuples/sec (with eventlogger: 0) roughly 0.5% improvement.
I then ran Jprofiler and saw **1.3%** time being spent in
send_to_eventlogger.
The profiler output itself might be offset due to the instrumentation
overhead. Jprofiler detects the following with send_to_eventlogger
`some methods with excessive instrumentation overhead has been detected.
They are called very frequently, their execution times are very short and the
time required for measuring those calls are dispropotional`
>I retired the throughput measurements with ackers=0 .. impact is even
greater ... its 25% faster when event.logger=0
Are you taking the measurements while the profiler is running or with the
debug flag turned on? I don't see this happening otherwise. Are you using the
latest Jprofiler (9.1.1) ? Are you using any extra plugins to instrument the
hashmap lookups (since I see the hashmap keys in your screenshot) ? If so that
itself might be skewing your results.
To avoid the persistentMap lookups, I also tried passing the **storm-id**
and **debug-atom** values as args to send_to_eventlogger and the percentage
reduced from 1.3 % to 1 % . You could try this change and see how it impacts
your topology.
I agree with @HeartSaVioR and don't think we need to set eventlogger to 0.
> Disable event logging by default
> --------------------------------
>
> Key: STORM-1632
> URL: https://issues.apache.org/jira/browse/STORM-1632
> Project: Apache Storm
> Issue Type: Bug
> Components: storm-core
> Reporter: Roshan Naik
> Assignee: Roshan Naik
> Priority: Blocker
> Fix For: 1.0.0
>
>
> EventLogging has performance penalty. For a simple speed of light topology
> with a single instances of a spout and a bolt, disabling event logging
> delivers a 7% to 9% perf improvement (with acker count =1)
> Event logging can be enabled when there is need to do debug, but turned off
> by default.
> **Update:** with acker=0 the observed impact was much higher... **25%**
> faster when event loggers = 0
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
