comp.lang.ada
 help / color / mirror / Atom feed
From: "Wiljan Derks" <Wiljan.Derks@zonnet.nl>
Subject: Re: puzzled re hyperthreaded performance
Date: Mon, 19 Sep 2005 22:21:47 +0200
Date: 2005-09-19T22:21:47+02:00	[thread overview]
Message-ID: <5bd67$432f1da8$3ea6010b$8314@news.versatel.net> (raw)
In-Reply-To: yqGdnQID8rvOn7TeRVn-jA@comcast.com

I did some performance testing on hyperthreaded pentium 2.8 using windows 
XP.
My conclusion was amazing:
* Hyperhreading works: it is as if things are executed in parallel
* Typically when only one thead is active, it can almost do the same amount 
of work
   compared to the situation with hyperthreading disabled.
* When two threads are active, the total amount of work done is certainly 
not more
   then having one thread is active.
* With hypertreading disabled, there seems to be more total performace.
Thus basically hyperthreading can makes a system more responsive, but not 
faster in cpu performance.
So it is not like a multiprocessor system.
I think the hypertreading is more like multiplexing the cpu over different 
threads without the
cost of taskswitching but t does not give you additional CPU performance.
This effect can make measurements very confusing.

I whould like to know if anyone has a different experience.





  parent reply	other threads:[~2005-09-19 20:21 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-09-15  5:10 puzzled re hyperthreaded performance tmoran
2005-09-15  6:08 ` jtg
2005-09-15 12:13   ` keld_nielsen_4nulspam
2005-09-15 20:44     ` tmoran
2005-09-15 21:43       ` tmoran
2005-09-19 20:21 ` Wiljan Derks [this message]
2005-09-20  0:26   ` tmoran
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox