[zeromq-dev] epgm performance numbers

Steven McCoy steven.mccoy at miru.hk
Thu Nov 1 18:59:59 CET 2012

On 29 October 2012 16:17, Rohan Bedarkar <r_bedarkar at yahoo.com> wrote:

> Throttling definitely helps. I reduced the HWM to 200 and added a 200 ms
> sleep every 50 samples and that brought down average round-trip times
> massively...
> So why is throttling helping?
The perverse answer is that 0mq is too fast.

0mq basically queues up messages so fast that the system becomes too
saturated to reliably send datagrams as the buffers grow causing expensive
allocs thus slowing everything down.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.zeromq.org/pipermail/zeromq-dev/attachments/20121101/cf757233/attachment.htm>

More information about the zeromq-dev mailing list