[Spread-users] Java API Performance

[Mike Perik]

Can't seem to reproduce that with my simple test publisher and subscriber.

Setting TcpNoDelay seems to alleviate the latency.

Mike


--- On Wed, 4/23/08, John Keating <jkeating at ca.afilias.info> wrote:

> From: John Keating <jkeating at ca.afilias.info>
> Subject: Re: [Spread-users] Java API Performance
> To: michaelperik at yahoo.com
> Cc: spread-users at lists.spread.org
> Date: Wednesday, April 23, 2008, 12:49 PM
> Earlier, you wrote that when you no longer have your sender
> doing a 
> receive(), the latency dropped to an acceptable level. Did
> you put the 
> receive() call back into the sender?
> 
> Mike Perik wrote:
> > Everything is running on the same box.  It's a
> linux machine using java 1.5.0_07. I have one machine in
> the configuration and all of the applications connect
> locally to the spread daemon.
> >
> > Like I said the C implementation does not have this
> latency.
> >
> > Mike
> >


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ