Re: gfortran, g95, and dual-core

On Sat, 29 Sep 2007 14:38:17 -0600, Charles Russell <NOSPAM@xxxxxxxxxxxxx>
wrote in <xtyLi.83797$Lu.51133@xxxxxxxxxxxxxxxxxxxxxx>:

So computing time is not simply a matter of counting "flops" (defined in
Golub & Van Loan "Matrix Computations" (1983) as the cost of "a floating
point add, a floating point multiply, and a little subscripting")
because of the complicating factor of cache. The simple model was so
much nicer.

For an analysis of a fairly complex piece of software (in C++
rather than Fortran, but that's largely irrelevant) look at the paper
and/or slides here:

http://indico.cern.ch/contributionDisplay.py?contribId=259&amp;sessionId=28&amp;confId=3580
http://indico.cern.ch/contributionDisplay.py?contribId=490&sessionId=13&confId=3580

For several reasons we're nowhere near to using the full capabilities
of our CPUs.

--
Ivan Reid, School of Engineering & Design, _____________ CMS Collaboration,
Brunel University. Ivan.Reid@[brunel.ac.uk|cern.ch] Room 40-1-B12, CERN
KotPT -- "for stupidity above and beyond the call of duty".
.