(j3.2006) (SC22WG5.5240) Interesting article

[Bill Long]

On May 8, 2014, at 1:28 PM, Van Snyder <Van.Snyder at jpl.nasa.gov> wrote:

> On Thu, 2014-05-08 at 10:35 +0100, John Reid wrote:
>> Hi,
>> 
>> Mike Metcalf has drawn my attention to this article
>> 
>> http://arstechnica.com/science/2014/05/scientific-computings-future-can-any-coding-language-top-a-1950s-behemoth/
> 
> If Julia really wants to get past half as fast as Fortran, maybe the
> target ought to be Fortran instead of LLVM.


LLVM has multiple parts.  The code-genrator part seems to be OK, and is handy for targeting multiple hardware platforms. The optimizer part might be a different story, but can be discarded in favor of a commercial optimizer that calls the LLVM code generator.

The real question is whether the language design of Julia helps or hurts optimization.  And if it helps, is the gain a sufficiently compelling reason for people to switch. Is it compelling enough for droves of C++ and Python programmers to jump and use Julia instead?   I see no evidence of that.  But then predicting failure of any new language has, historically, been a pretty safe bet.

Cheers,
Bill


> 
>> 
>> Regards,
>> 
>> John.
>> _______________________________________________
>> J3 mailing list
>> J3 at mailman.j3-fortran.org
>> http://mailman.j3-fortran.org/mailman/listinfo/j3
> 
> 
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3

Bill Long                                                                       longb at cray.com
Fortran Technical Suport  &                                  voice:  651-605-9024
Bioinformatics Software Development                     fax:  651-605-9142
Cray Inc./ Cray Plaza, Suite 210/ 380 Jackson St./ St. Paul, MN 55101