<div dir="ltr">Watch also NUMA stuff on these bigger multi-core systems. I've seen some very interesting effects due to NUMA and how it affects erlang. Look at these posts for some more details:<div><a href="http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/2013-January/024790.html" target="_blank">http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/2013-January/024790.html</a><br>
</div><div><br></div><div>Typically a system with as many cores as you've got requires some more "tweaking" to get good performance out of it. One of the best articles I'd found was this:</div><div><a href="http://www.agoragames.com/blog/2011/06/24/of-penguins-rabbits-and-buses/" target="_blank">http://www.agoragames.com/blog/2011/06/24/of-penguins-rabbits-and-buses/</a><br>
</div><div><br></div><div><br></div><div>I don't THINK this would have an impact on your issues with inter-node communications but I have to toss it out there. Also, another fun thing we saw was long DNS lookups causing some odd issues as well.</div>
<div>Jason</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Apr 17, 2014 at 8:56 AM, Matthias Radestock <span dir="ltr"><<a href="mailto:matthias@rabbitmq.com" target="_blank">matthias@rabbitmq.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On 15/04/14 14:20, joseph rouphael wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">
On Mon, Apr 14, 2014 at 12:00 PM, Matthias Radestock<br></div><div class="">
<<a href="mailto:matthias@rabbitmq.com" target="_blank">matthias@rabbitmq.com</a> <mailto:<a href="mailto:matthias@rabbitmq.com" target="_blank">matthias@rabbitmq.com</a>><u></u>> wrote:<br>
What happens when you run the above test with both nodes on the same<br>
machine?<br>
<br>
The result is almost the same: Fluctuating between 10 and 23KHZ<br>
</div></blockquote>
<br>
Thanks for running all these tests. I've reproduced your observations.<br>
<br>
What we are seeing here is, at least partially, an issue with the Erlang scheduler.<br>
<br>
I noticed that when running the test with PerfTest, single producer, single consumer, two nodes, the consuming node was using 100% CPU, i.e. the equivalent of a single core. That is usually an indication of an Erlang scheduler issue since there is an appreciable degree of parallelism that should be exploitable.<br>
<br>
For this particular test I managed to double performance by supplying the "+swt very_low" or "+sfwi 1" flags to Erlang via RABBITMQ_SERVER_ERL_ARGS.<br>
<br>
So these flags, plus the other +s... flags of the Erlang VM, are worth playing with.<div class="HOEnZb"><div class="h5"><br>
<br>
Matthias.<br>
______________________________<u></u>_________________<br>
rabbitmq-discuss mailing list<br>
<a href="mailto:rabbitmq-discuss@lists.rabbitmq.com" target="_blank">rabbitmq-discuss@lists.<u></u>rabbitmq.com</a><br>
<a href="https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss" target="_blank">https://lists.rabbitmq.com/<u></u>cgi-bin/mailman/listinfo/<u></u>rabbitmq-discuss</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div dir="ltr">Jason McIntosh<br><a href="https://github.com/jasonmcintosh/" target="_blank">https://github.com/jasonmcintosh/</a><br>573-424-7612</div>
</div>