<div class="gmail_quote">On Wed, Feb 20, 2013 at 1:29 PM, Daniel P. Berrange <span dir="ltr"><<a href="mailto:berrange@redhat.com" target="_blank">berrange@redhat.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="im">
</div>IIUC this '10% hit' is a micro-benchmark result - it isn't considering<br>
the RPC overhead in the context of the other things in OpenStack that<br>
consume CPU resources. If RPC only accounts for, say, 5% of CPU usage<br>
across openstack services, then a 10% hit would only adding 0.5% overhead<br>
to an OpenStack deployment as a whole.<br>
<br>
So 10% sounds like alot of overhead, but my gut feeling is that it<br>
may well be lost in the noise once you consider all the other things<br>
going on in OpenStack services<span class="HOEnZb"></span><br></blockquote><div><br>Excellent observations and yes, it was a measurement of OpenStack RPC with no real service on top. I need to move on to API level testing. <br>
</div></div>