[Cplex] Keywords, compliance, and a parallel UNCOL.
robert.geva at intel.com
Sat Jun 29 09:43:15 CEST 2013
The problem address by OpenACC is support for accelerators / coprocessors / etc, mostly with either non shared memory or heterogeneous execution units (or both).
This problem is not in the scope of what WG14 asked this study group to work on.
We need to solve the "simpler" problem of shared memory parallel programming.
From: cplex-bounces at open-std.org [mailto:cplex-bounces at open-std.org] On Behalf Of Matthew Markland
Sent: Friday, June 28, 2013 7:31 PM
To: Nelson, Clark; cplex at open-std.org
Subject: Re: [Cplex] Keywords, compliance, and a parallel UNCOL.
Clark et al:
> -----Original Message-----
> From: Nelson, Clark [mailto:clark.nelson at intel.com]
> Sent: Friday, June 28, 2013 6:02 PM
> To: Matthew Markland; cplex at open-std.org
> Subject: RE: Keywords, compliance, and a parallel UNCOL.
> > As a follow on to the above, is the focus here going to be simply
> > shared memory parallelism or will spawn be allowed to off-load work
> > to an accelerator? Data movement becomes critical if that is
> > allowed, and would probably add to the complexity of this base runtime.
> Another very good question. My impression is that there is no mature
> language support for distributed-memory systems; in other words, we're
> short of existing practice that's ready for standardization. (MPI is
> mature, of course, but it's library, not language.)
> But I sure wouldn't mind learning otherwise.
The only item I would have to offer up in that direction would be OpenACC which is not a mature standard, but does have the interest of some. I hesitate to mention it because I do not want to foreground disagreements between the OpenMP and OpenACC groups. My position in Cray does not favor one over the other; I have to implement both. :-)
Cplex mailing list
Cplex at open-std.org
More information about the Cplex