[J3] Performance Portability and Fortran: Making Fortran cool again

[Gary Klimowicz]

> On 1/17/19, 11:27 AM, "J3 on behalf of Bill Long via J3" <j3-bounces at mailman.j3-fortran.org on behalf of j3 at mailman.j3-fortran.org> wrote:
> 
>     Interesting.  Michael sometimes has strongly held opinions about things in computing. But often they are right, so people do pay attention.   Thanks for sharing. 
>     
>     Cheers,
>     Bill
    
    > > On Jan 17, 2019, at 10:21 AM, Tom Clune via J3 <j3 at mailman.j3-fortran.org> > wrote:
    > > 
    > > A colleague just shared a link that seems relevant to one of the subthreads of this discussion:
    > > 
    > > https://www.nextplatform.com/2019/01/16/burying-the-openmp-versus-openacc-hatchet/
    > > 
    > > Note - I think the author has stated things exactly backwards in several places, but I think the committee will generally like the conclusion at the end.
    > > 
    > > - Tom
    > > 
    > > 
 
    
Yes, Tom, thanks for sharing this. Michael sits two cube rows away from me, and I didn't know he was writing something on this.

One hot topic in the panel discussion this morning (that you all participated in, at a distance) was what do people do to manage the risk associated with adopting a particular performance portability strategy (directives, language features, programming models like Kokkos and Raja). Being a feature in the language standard provides high confidence, if it's implemented in more than one compiler. (As Michael says: No directives are better than directives. But directives seem to be a good way to experiment with these ideas.)

There was also a lot of discussion on the issue of data management. I understand Kokkos goes to great lengths to give the programmer power to specify the layout of the data for performance. I need to learn more about this issue.


-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------