[email protected] said: > My device thread is running at SCHED_RR pri=66, I have approximately > 256 client threads, the time from last packet send() on REQ to last > device recv() on XREP is averaging around 40ms with a stddev of about > 36-40 (10,000 samples) > > There's limited or no variation reordering poll upcalls (reactors) > > The other direction (device XREP >> user REQ) is almost always nil. > > I haven't declared any HWMs so presumably they are infinite.
They would be infinite by default, yes. Can you define your workload a bit more? With infinite HWM the most obvious thing I could see happening is that the device is just not keeping up with the load, so the large latency may be due to queueing time (i.e. large queue length on client->device path). > > > What platform are you running on? > > > > Linux dev2201.corp.xplusone.com 2.6.18-194.8.1.el5xen #1 SMP Thu Jul 1 > 19:41:05 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux Xen... is the hardware shared with multiple VMs? That'd also be something to check; do you get the same problems on bare hardware? -mato _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
