RE: Workstation Computers for FLUKA

Hi Chris,

I follow this discussion that diverges from the original quetion, but I
would like to return a bit to the original question. Are we sure that
Fluka is only, or at least major, dependant on the double precision
performance of the (C,G)PU?
I faintly remember a presentation that was stateing that Geant4
performance was following more closely the SPECint numbers than the
SPECfp numbers. With the explanation that the disicions that an MC code
has to make are as expensive as the actual calculations. I wonder if
this is correct also for Fluka.

Is anyone out there having actual numbers?

Thank you,
Makis

________________________________________
From: owner-fluka-discuss_at_mi.infn.it [owner-fluka-discuss_at_mi.infn.it] on
behalf of Chris Theis [Christian.Theis_at_cern.ch]
Sent: 26 November 2010 16:31
To: Nicholas Bolibruch; turgaykorkut_at_hotmail.com; fluka-discuss_at_fluka.org
Subject: RE: Workstation Computers for FLUKA

Hi Nicholas,

there is a lot of hype recently regarding GPUs (some of it certainly
justified), but there are some subtle issues especially
in connection with MC simulations, which I would like to comment on.

> I personally would love to see a version of Fluka take advantage of modern
> graphics processors that have a large number of cores, and are now showing
> good performance for double precision operations.

It is true that support for double precision has become a lot better lately.
However, in comparison to native single precision calculations the
performance
benefit is currently significantly lower. The benchmarks GPU vs. CPU
usually shown are based on single precision!
The probably more important point is that different vendors have allowed
th=emselves a bit of freedom with
respect to the implementation of the IEEE floating point standard. It is
not guaranteed that the
GPU that you are running your program on will yield the same result as
the one of your colleague because it's from a different
vendor. In terms of debugging& quality control this is a very important
aspect that is commonly overlooked!

Furthermore, using proprietary platforms like CUDA limits the program to
one vendor,
like Nvidia GPUs in this case. Having portability in mind one will
probably have to go for something like OpenCL in the end where you will
sacrifice
(a bit) of speed on the altar of portability.

> There are some utilities to convert Fortran 95 code to CUDA, obviously with a lot of manual work to
> specify what resources to allocate for particular code segments, unfortunately I have yet to find anything that can do this for Fortran 77.

I'm afraid that the chance of finding something suitable for F77 are
realistically below zero.
The F77 standard is more than 30 years old which renders it practically
a dinosaur, looking at the short turnover times seen in
the recent years. Furthermore, maintenance of the popular G77 compiler
was stopped many years ago, so I would not count
on anybody making the effort to provide support for modern GPUs with
respect to F77.

Cheers
Chris
Received on Sun Nov 28 2010 - 14:54:03 CET