[rabbitmq-discuss] Performance degrades with increasing queue depth

majek04 majek04 at gmail.com
Mon Sep 14 11:57:54 BST 2009


On Tue, Sep 8, 2009 at 06:00, Matthias Radestock <matthias at lshift.net> wrote:
> I ran the tests and got the following results:
>
> - 20k messages: 500Hz, with beam.smp using 30% of the CPU
>
> - 200k messages: 300Hz, with beam.smp using 45% of the CPU - that kind
> of slowdown is not totally unexpected and probably due to the increased
> cost of gc and the nature of the queue data structure we are using,
>
> - 200k messages: 35Hz, with beam.smp using 95% of the CPU - yes, that's
> right, the same test as the previous one but only 1/9th of the throughput!

I was able to reproduce this results. The slowdown was identified
in basic.get. The real cause is a bit mysterious, it looks like the issue
has something to do with Erlang garbage collection mechanism.

Hopefully, the problematic code was refactored during the work
on new persistor, so RabbitMQ 1.7 will not be affected.

Cheers.
  Marek Majkowski




More information about the rabbitmq-discuss mailing list