[zeromq-dev] debugging performance problems
andrew at research.att.com
Mon May 16 16:29:07 CEST 2011
my reading of teh manual pages is that HWM has no effect on sending a message
(only if hwm is set for teh receivers).
on the other hand, when i set it in an experiment, it seemed to have an effect.
on the other other hand, i normally and doing fan-in fan-out on both sides
of my push/pull links, so an hwm on the sending side is problematic.
On May 16, 2011, at 7:22 AM, Ilja Golshtein wrote:
> just to verify, to set hwm limits on a push pull, i set it just on the pull end,
> after the zmq_socket call and before the zmq_connect call. correct?
> Sometimes yes, but generally not .
> If sender is faster than receiver (otherwise everything is fine) queue would grow up at sender side if it is limited at receiver side.
> Assuming the goal is to block, you need HWM at both sides.
> [As far as I understand - hope guru correct me if I am wrong]
> Best regards,
> Ilja Golshtein.
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
Andrew Hume (best -> Telework) +1 623-551-2845
andrew at research.att.com (Work) +1 973-236-2014
AT&T Labs - Research; member of USENIX and LOPSA
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the zeromq-dev