I meant sum of all cores percent workload divided by 8 for the Intel i7
processor works out to 12% to 14% (it varies a little) when only one Maxima
instance is running the tests. That is a little easier to understand.
With two instances is stays near 25%. In any case it is max() not sum
Rich
-----Original Message-----
From: Richard Hennessy
Sent: Tuesday, December 28, 2010 12:38 AM
To: smh at franz.com ; Richard Fateman
Cc: Stavros Macrakis ; Maxima List
Subject: Re: [Maxima] Multi-threaded Maxima
-----Original Message-----
From: Steve Haflich
Sent: Monday, December 27, 2010 8:44 PM
To: Richard Fateman
Cc: Stavros Macrakis ; Maxima List
Subject: Re: [Maxima] Multi-threaded Maxima
Richard Fateman <fateman at eecs.berkeley.edu> wrote:
Yeah, max, not sum would be right.? But here's a question. Why, if
we are running on a single processor, is it only 50% busy?? Is it?
50% = 100% X (1/2) ? because it is? one processor out of two?? Or
is it 50% of one processor and 0% on the other processor?
That is how I interpreted Hennesey's original comment. His 13% was
suspiciously close to 1/7, i.e. one out of seven processors working.
But both kinds of metering are used by various metering tools.
The Task manager shows 8 cpu's busyness state which does not include system
idle time. When Maxima is running in a single process about half of the
cpu's are parked, which I interpret to mean they are doing nothing. The 13%
is reported as a total busyness state of all 8 threads in all cores, which I
think is the sum of all cores divided by state / 8 * 100 (to give as a
percent you have to multiply by 100), this is also shown in task manager.
_______________________________________________
Maxima mailing list
Maxima at math.utexas.edu
http://www.math.utexas.edu/mailman/listinfo/maxima
_______________________________________________
Maxima mailing list
Maxima at math.utexas.edu
http://www.math.utexas.edu/mailman/listinfo/maxima