<div class="gmail_quote">On Thu, Jan 24, 2013 at 3:38 PM, 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">
{queue,exchange}.declare when the resource already exists are quite cheap. However, the latency of the associated network roundtrip can badly affect performance. Unless you set the nowait flags, thus suppressing the response and making the commands asynchronous. Even then though these are not commands you'd want to send for every message.<br>
</blockquote></div><br>Matthias, thanks, that is good to know. Latency isn't the issue in my study. I drove the RabbitMQ servers in a cluster of up to 3 to 100% CPU and throughput was my main metric. It would be interesting to profile the RabbitMQ server to see where it is spending most of it's time.<br>