| 25 ms is the timeout I think.<br><br>If so, this matches up very well with my experimental results, which was showing that that a single publisher never went over ~37 message/sec. With a 25 ms window, the max theoretical throughput would be 40/sec if I'm understanding you correctly.<br>
<br><br><div class="gmail_quote">On Mon, Feb 20, 2012 at 2:58 PM, Matthew Sackman <span dir="ltr"><<a href="mailto:matthew@rabbitmq.com">matthew@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="im">On Mon, Feb 20, 2012 at 10:22:15AM -0800, Matt Pietrek wrote:<br>
> Ah, thanks so much for this reply Matthew. It's extremely enlightening and<br>
> seems to explain what we're seeing.<br>
><br>
> A follow up question just for my understanding: Are these "windows" dynamic<br>
> in nature (e.g., varying length, varying frequency, somehow tied to current<br>
> load)? Or are they more static (e.g. "Once every 20 milliseconds").<br>
<br>
</div>25 ms is the timeout I think. If the queue goes "idle" before then (i.e.<br>
has no more work to do) then it will issue the fsync immediately. Thus<br>
this helps prevent latency go up too much for the single-publisher case.<br>
<div class="HOEnZb"><div class="h5"><br>
Matthew<br>
_______________________________________________<br>
rabbitmq-discuss mailing list<br>
<a href="mailto:rabbitmq-discuss@lists.rabbitmq.com">rabbitmq-discuss@lists.rabbitmq.com</a><br>
<a href="https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss" target="_blank">https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss</a><br>
</div></div></blockquote></div><br>