[Beowulf] g77 limits...
Robert G. Brown
rgb at phy.duke.edu
Wed Feb 22 13:48:53 PST 2006
On Wed, 22 Feb 2006, Mark Hahn wrote:
>> archives because it is so time dependent an answer. In particular, can
>> an x64 box with a modern linux kernel and a modern g77 allocate 4-8 GB
> isn't "modern g77" a bit of an oxymoron?
> (given the existence of gfortran, I mean).
Hey, remember, as far as I >>personally<< am concerned fortran (of any
flavor) belongs carved like mystical hieroglyphs or runes on strange
obelisks and folded, spindled, and mutilated card decks -- without any
sort of co-translation that would enable it to be revived;-)
However, for the purposes of this project, it will at least
tentatively/initially be presumed that the software to be ported,
modernized, developed, will need to run across many platforms and OS's,
including on many old i386 boxes running old linux variants, possibly
back to RH 7.x or thereabouts, all over the world. AND, as noted, on
fairly modern x64 compute clusters, where at least some of the packages
would benefit from cone-head quantities of memory.
Now, one of the agendas I intend to pursue (if this gets funded) is
making it really really easy for the community in question to install a
uniform and modern linux distro (or even one of several choices). And
sure, it would be lovely to gently nudge the code development over to C
and/or C++ and away from Fortran -- without really wanting to start the
usual religious wars over what is or isn't a good language in spite of
my obelisk reference above;-)
Actually, from a purely practical point of view there is a lot of legacy
code and a medium amount of fairly recent (<10 year old) fortran code
out there in physics in general and this project in particular. We'll
doubtless rewrite some of it, where appropriate, but quite a bit of it
will almost certainly remain in fortran as that is what its developers
have used and will likely continue to use.
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
More information about the Beowulf