(j3.2006) Preparing for the Tokyo meeting

[Bill Long]

N.M. Maclaren wrote:

> While that is true, multi-user (shared) systems are NOT going away, and
> parallel applications are a right b*gg*r to schedule on such things.
>   

Really?  All of our systems are configured as multi-user (shared) 
systems, and run parallel applications almost exclusively.  Job 
scheduling for multi-processor systems (including vanilla clusters) has 
been a solved problem for years.


> Some administrators forbid them, and jump had on users caught running
> them.  

I can understand an administrator jumping on users for running serial 
codes on a cluster, on grounds that they are wasting resources since 
they can run codes like that on their desktop machines.   The problem of 
an administrator who bans parallel jobs from a parallel computer is 
probably handled best by the (un)employment office.  Certainly not an 
issue for WG5 or J3.



> But, in many more, they cause trouble to other users and/or run
> very badly (including taking many times as much CPU as they need).
>   

Badly written programs is a programmer training problem.  I would expect 
coarrays to actually help since they (in my experience) are easier to 
use correctly than MPI.

> So serial Fortran will not go away any time soon, and some people will
> positively want coarray-free compilers (or a fixable mode).
>
>   

I agree that having a compiler switch that allows the user to fix 
num_images at compile time (to 1, for example), or to disable the 
recognition of coarrays entirely, is a fine vendor feature.  That gives 
the user freedom to live in a serial-only world.  It is also explicitly 
outside the scope of the standard (1.3 Exclusions, page 1).  Having a 
separate compiler that cannot deal with coarrays at all seems like a bad 
idea on two grounds:  for the vendor there is added cost to test two 
different compilers, and for the user the need to keep two compilers 
around for the day when he discovers modern computers (or wants to use 
someone else's program that employs coarrays).


>
> Er, the vast majority of Fortran users have never USED or even SEEN a
> vector machine (and, no, I don't count SSE etc.)  You have, and I have,
> but the kiddies - including post-docs here :-) - I teach never have.
> Some have never even HEARD of them!
>
>   

Sorry, but SSE does count.  And if you write your codes to vectorize 
well, you get better performance even on commodity processors.   This is 
basic programmer training.  Where are these people (not) learning how to 
program?  This is not really an issue for the Fortran standard, but it 
sounds like computer science / programmer education generally is in a 
shocking state of disrepair!  (With some notable exceptions, like John 
Wallin, who are working to correct the problem.)


Cheers,
Bill


-- 
Bill Long                                   longb at cray.com
Fortran Technical Support    &              voice: 651-605-9024
Bioinformatics Software Development         fax:   651-605-9142
Cray Inc., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120