Super-scaling

[Daniel Kidger]

Simon,

   Superscaling is seen most often when you run the old-fashioned style of
scalability test: one is which the overall problem size stays constant even
though you are increasing the number of CPUs you are running on.
In this case the problem size per node falls as the node count increases
such that more and more of the data will reside in cache.
Such problems are considered 'memory-bandwidth'-bound rather than CPU or
interconnect-bound.


Yours,
Daniel.

--------------------------------------------------------------
Dr. Dan Kidger, Quadrics Ltd.      daniel.kidger at quadrics.com
One Bridewell St., Bristol, BS1 2AA, UK         0117 915 5505
----------------------- www.quadrics.com --------------------


-----Original Message-----
From: Simon Hogg [mailto:seth at hogg.org]
Sent: 19 February 2003 13:30
To: Beowulf mailing list
Subject: Super-scaling


Is there a rough rule of thumb which dictates when a program (if ever) 
shows superscaling with number of nodes.  Of course, I would not expect 
this to carry on ad infinitum, but does anyone see superscalar behaviour up 
to, a certain number of nodes.

What would be the conditions for this to occur?

	Simon

_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf