tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nikola Milutinovic <>
Subject Re: dramatic performance differences on development machines
Date Fri, 16 Jan 2004 11:53:00 GMT
Johan Coens wrote:

> Hello all,
> I tried:
> - isolating code in a java class and running it on both machines, about same
> performance

So, the machine itself is OK.

> - running tomcat 4.1.29, bit faster but still ~15 sec.
> - changing network connection to full duplex, 2*faster but stil ~6 sec
> (should be < 1 sec). indicates network traffic could be a bottleneck

Indeed. Try sniffing the traffic, see what actually goes in and out of the box 
during one request.

> - running websphere, performance ~600 ms.

Reasonable, I guess.

> I see tomcat running high on processor performance, peeks in processor
> performance are the same as peeks in the network traffic. But, when running
> everything on one box, it is also slow (so network should not be an issue).

Network IS the problem, since changing to "full duplex" doubled the performance. 
It is not your request (HTTP request is very small). Nor is it server's response 
(I mean, it is a HTML page, it is not *that* big). I bet there is some other 
network traffic, that gets triggered as a side effect of a request. Like DNS 
query (reverse lookup) or some such thing.

> Also, the developer on the fast machine noticed that when getting code from
> CVS from other developers, his system slowed down signifficant too, but I
> have no clue what too look for with this indication.

Sniff the traffic, find out what is actually going on. Leave guessing and 
crystall balls for later. You've tried the obvious and less obvious stuff. Time 
to dig in. You could also employ heavy artilery, some Java Profiler, and run 
Tomcat under (or with) it.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message