[
https://issues.apache.org/jira/browse/STORM-297?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14017428#comment-14017428
]
ASF GitHub Bot commented on STORM-297:
--------------------------------------
Github user Gvain commented on the pull request:
https://github.com/apache/incubator-storm/pull/103#issuecomment-45056355
@clockfly
Thanks for your detailed test. But notice that there is a big difference
between our test results.
In my test, only using 8 workers, i get the throughput peak which is 656K
tps, and adding more workers the throughput drops down.
In your test, adding more worker helps increasing throughput, and the
throughput peak reached 622K tps when using 24 workers. The worker number is
greatly increased and thus the total threads number.
I am wondering why this difference exists.
> Storm Performance cannot be scaled up by adding more CPU cores
> --------------------------------------------------------------
>
> Key: STORM-297
> URL: https://issues.apache.org/jira/browse/STORM-297
> Project: Apache Storm (Incubating)
> Issue Type: Bug
> Reporter: Sean Zhong
> Labels: Performance, netty
> Fix For: 0.9.2-incubating
>
> Attachments: Storm_performance_fix.pdf,
> storm_Netty_receiver_diagram.png, storm_conf.txt,
> storm_performance_fix.patch, worker_throughput_without_storm-297.png
>
>
> We cannot scale up the performance by adding more CPU cores and increasing
> parallelism.
> For a 2 layer topology Spout ---shuffle grouping--> bolt, when message size
> is small (around 100 bytes), we can find in the below picture that neither
> the CPU nor the network is saturated. When message size is 100 bytes, only
> 40% of CPU is used, only 18% of network is used, although we have a high
> parallelism (overall we have 144 executors)
--
This message was sent by Atlassian JIRA
(v6.2#6252)
