Christian, A random idea: could you retest performance using the old 2.0.11 package? On the remote chance that something changed in the 2.1 branch that affects Windows performance...
Cheers Pieter On Fri, May 13, 2011 at 10:27 PM, Christian Martinez <[email protected]> wrote: > Thanks Steve! Good to know the docs reflect your experience. > > > > --CM > > > > From: [email protected] > [mailto:[email protected]] On Behalf Of Steven McCoy > Sent: Friday, May 13, 2011 1:24 PM > To: ZeroMQ development list > Subject: Re: [zeromq-dev] Tips on optimization of TCP for Windows 7 and > Windows server > > > > On 14 May 2011 01:03, Christian Martinez <[email protected]> wrote: > > Yes I could look this up internally, but I’d like input from the list on > what settings/tweaks they’ve used in real life that had the most impact on > their performance. > > > > > > Follow the performance guidelines set by Microsoft, the 2008 R2 guide works > pretty well. > > > > http://msdn.microsoft.com/en-us/windows/hardware/gg463392 > > > > I'd love it if you can find the source code for NTttcp as yields rather odd > results at times when compared with the original Test TCP (TTCP), PCAUSA > port PCATTCP, iperf or netperf. Here's my notes for UDP and PGM performance > testing, > > > > http://openpgmdev.blogspot.com/2011/04/wherefore-art-thou-ip-packet-make-haste.html > > > > -- > Steve-o > > _______________________________________________ > zeromq-dev mailing list > [email protected] > http://lists.zeromq.org/mailman/listinfo/zeromq-dev > > _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
