Newsgroups: comp.parallel.pvm
From: thomas@wein01.elektro.uni-wuppertal.de (thomas faerbinger)
Subject: measuring CPU idle-time
Organization: University of Wuppertal
Date: 23 Aug 1994 17:53:05 GMT
Message-ID: <33dd21$4lv@wmwap1.math.Uni-Wuppertal.DE>


In article <Cuo373.Fn2@dnsserv.go.dlr.de>, eckhard@ts.go.dlr.de (Eckhard Rueggeberg) writes:
|> If you measure CPU time (given by getrusage, for example),
|> you won't get the effects of CPU's waiting for communication. We 
|> had such an example here : A programmer measured only CPU time, 
|> and was happy that the sum of two CPU times was only epsilon larger
|> than sequential execution. I had to show him that on two otherwise
|> idling workstations on a quiet ethernet, the total execution time
|> was LONGER than sequential execution on a single machine, because 
|> the efficiency was extremely poor. The processes got only 30% CPU
|> time on their machines, which were idling 70%. This obviously is
|> not what you want, but you wouldn notice if you rely on any formula
|> on slave CPU times.

Can I measure the idling percentage of a CPU ? Or even better the
computing power available at a given time ?
Something like

 get_over_all_rusage()

or

 int get_MFLOPS_now_possible()

I have an algorithm where dynamic load-balancing is not worth
it but I'd like to distribute the problem in a sensible way.
Any hints appreciated,
	Thomas.

