geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rainer Jung <rainer.j...@kippdata.de>
Subject Re: Geronimo peak performance
Date Sat, 08 Apr 2006 13:58:09 GMT
When preparing for the test, have a look at the very interesting 
cpustat/cputrack. Without this you won't really be able to judge on the 
cpu usage. Classical tools like sar/vmstat/prstat easily lead to wrong 
results (the amount of work a single strand [hardware thread] is able to 
do is not fixed. It really shares the core with the other strands. If 
they stall, it will execute on every cycle, if all 4 strands are busy 
each one will only execute on every 4th cycle. So the cpu power 
available to a strand is dynamic, but tools like prstat are not able to 
show that.

For memory bus usage busstat is the way to go.

If you need more info, you could start an off topic thread or mail me 
directly.

Rainer

Jeff Genender wrote:
> Yep...each core will handle 4 hardware threads.  It was built for
> multiple threads and a JVM (multithreaded of course) is made for this
> box.  For an appserver under load, I think the numbers should be
> impressive on this box.
> 
> The cool thing about this machine is, each core is viewed as 4 cpus.
> That means this box looks like it has 16 CPUs running (4 cores x 4
> hardware threads)!
> 
> I can't wait to pound on this and see what it can do.  I am going to
> have T2000->T2000 trials going on to maximize the throughput for CPU.
> They come with 4 1 Ghz Ethernet ports each, so I am going to pipe these
> up and hopefully there will be zero bottleneck on IO.

Mime
View raw message