On 4/18/07, Mittler, Nathan <[EMAIL PROTECTED]> wrote:
Interesting ... Certainly possible. Although if running the same test for openwire and stomp, wouldn't you expect the same issue with both?
I just wondered if the TCP-NO-Delay being off would hit OpenWire much harder, since the packets are much smaller so would be likely to sit in the TCP buffers for longer. It might be worth turning it on & off and trying both stomp & openwire to see if it has an effect. (I'd expect it to hurt openwire more than stomp) James
Nate > -----Original Message----- > From: James Strachan [mailto:[EMAIL PROTECTED] > Sent: Wednesday, April 18, 2007 6:51 AM > To: [email protected] > Subject: Re: AMQCPP Openwire much slower than Stomp? > > I wonder if its the effect of TCP-NO DELAY (Nagler) kicking in. i.e. > the small messages of OpenWire may take a while to be flushed > to the socket until the buffer fills or timeouts kick in. > > On 4/18/07, Albert Strasheim <[EMAIL PROTECTED]> wrote: > > Hello all > > > > On Wed, 18 Apr 2007, Nathan Mittler wrote: > > > > > Unfortunately we didn't do a performance analysis, as we've been > > > focused on just getting 2.0 out the door. We did however > notice a > > > bit of slowness in our integration tests. Have you, by > any chance, > > > profiled the openwire connector to see where it's > spending its time? > > > > > > With any luck, we can zero in on the bottlenecks quickly > and cut a 2.1. > > > > I'm going to spend some time on this over the next few days. The > > symptoms are quite strange. Factors that seem to influence the speed > > include: > > > > - whether connection is to a local broker or a remote broker > > - whether the broker is running on Windows or Linux > > - message size possibly in combination with wire format > > > > I'll report back if I figure out anything more useful. > > > > Regards, > > > > Albert > > > > > -- > > James > ------- > http://radio.weblogs.com/0112098/ >
-- James ------- http://radio.weblogs.com/0112098/
