From owner-sc22wg5+sc22wg5-dom8=www.open-std.org@open-std.org  Thu Jul  6 21:38:47 2017
Return-Path: <owner-sc22wg5+sc22wg5-dom8=www.open-std.org@open-std.org>
X-Original-To: sc22wg5-dom8
Delivered-To: sc22wg5-dom8@www.open-std.org
Received: by www.open-std.org (Postfix, from userid 521)
	id 682F5357250; Thu,  6 Jul 2017 21:38:47 +0200 (CEST)
Delivered-To: sc22wg5@open-std.org
Received: from mail-wr0-f173.google.com (mail-wr0-f173.google.com [209.85.128.173])
	(using TLSv1 with cipher AES128-SHA (128/128 bits))
	(No client certificate requested)
	by www.open-std.org (Postfix) with ESMTP id 2F56C357140
	for <sc22wg5@open-std.org>; Thu,  6 Jul 2017 21:38:47 +0200 (CEST)
Received: by mail-wr0-f173.google.com with SMTP id c11so16690608wrc.3
        for <sc22wg5@open-std.org>; Thu, 06 Jul 2017 12:38:47 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20161025;
        h=mime-version:in-reply-to:references:from:date:message-id:subject:to
         :cc;
        bh=bR8u2DDvcOZ/NvOQnB+hgEMeFi2/KbKAstmYLJbyDsM=;
        b=JYqttG+h3JTEoVL2cCVgq5xhkSnf4lAfe5GtKXp18pJKoNTTf7fc2FgmKRPG/CqY1e
         FRFBhYzaprQl0KxIyiyMtkpHuRxPn06uLhMznrT3y/Y+GZ+jn5+IQGhi0/dZ5+P241zR
         SArm/NatOiJJeX7KfUWW5xanKqgOql2GczPOWoMBD0rYQp7GZfNZX4NI11bFGA6okihP
         CWpbs1m7wWBHsMPHfq0rO8sArNGjZgVH+B5BuQpyT1yqGb2tg3z/1ZEmwla/kFhLYZJT
         C1tMsmcBabt0PoPq9vbj8iUu69lJ8ios6Rl/OpF2Iwh7oxON6ES/gABlAgkGERrMraMw
         ecHA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=1e100.net; s=20161025;
        h=x-gm-message-state:mime-version:in-reply-to:references:from:date
         :message-id:subject:to:cc;
        bh=bR8u2DDvcOZ/NvOQnB+hgEMeFi2/KbKAstmYLJbyDsM=;
        b=GWNRHM4tnvjRzQmP1VyF/ujeRm6ybVkJi9TTlS9JQ4ChXNi1qcsQEuIFS+tS+raXl1
         T8Ea2+TZZ4MjRLTEV798l98J1YgBt9554PmHH199MMIX2F10XOfgc5PJVzpnzn1vYSe9
         GIRO5OikDUaYtDnFTIWw6sF0cR/N2vgttI+0QkyFO24BLLD3C+GVGWytsZErFJRu5rOF
         SxM+aCS52DooAUGilNyUuYvViSqerQQqhSOegPXKq54aNixIB6NFMvC1jgU7jBhotSZl
         8kvWRnjuCaR/TPN1oMCGlu422GNqpNlpdUcKYvtPxk3jCKAKiICBCdb+Gq3+muXmIkUf
         v9Jg==
X-Gm-Message-State: AIVw111ro105VtmZOCZhvrwnZpJ5Ag8G/I7+L/TOQumU8IrZS/9dTaZH
	MSls/P4clce3vS15jL2nXZeZMAc8kQ==
X-Received: by 10.80.143.163 with SMTP id y32mr480084edy.103.1499369926671;
 Thu, 06 Jul 2017 12:38:46 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.80.224.131 with HTTP; Thu, 6 Jul 2017 12:38:06 -0700 (PDT)
In-Reply-To: <20170706190037.9CF963581F6@www.open-std.org>
References: <20170705131003.C2A753587D1@www.open-std.org> <677196EB-62B3-448D-8AD9-6D0E36BAFD32@cray.com>
 <888EAD5C-B10E-4E55-9F63-35F1BBE2F342@nasa.gov> <CA+KjR2bpJiEB+PRnxzrvX6VBoiM=ifOp-rOQdY4fMFatKhiQLQ@mail.gmail.com>
 <20170705194923.55D94358633@www.open-std.org> <CA+KjR2azkyDuo+osGkRiooJE6Guzb2F5PjkcDmiP8u=M+0=2+Q@mail.gmail.com>
 <74223362-5937-4576-81B7-86D2A75460D3@nasa.gov> <31050494-69BF-4BCD-991D-069E70443B1D@cray.com>
 <20170706190037.9CF963581F6@www.open-std.org>
From: Keith Bierman <khbkhb@gmail.com>
Date: Thu, 6 Jul 2017 13:38:06 -0600
Message-ID: <CAHSokPyY=424qua96aZMqXrFVkq3Z833kOOYJ6V9RJry1w1uDQ@mail.gmail.com>
Subject: Re: (j3.2006) (SC22WG5.5907) 3 levels of parallelism?
To: "Van.Snyder" <Van.Snyder@jpl.nasa.gov>, 
	fortran standards email list for J3 <j3@mailman.j3-fortran.org>
Cc: sc22wg5 <sc22wg5@open-std.org>
Content-Type: multipart/alternative; boundary="f403045c825addc5bc0553ab4399"
Sender: owner-sc22wg5@open-std.org
Precedence: bulk

--f403045c825addc5bc0553ab4399
Content-Type: text/plain; charset="UTF-8"

In terms of Standard(s) compliant ways to doing things ... wouldn't mmap
(posix) along with C interop be the "right" way to go for new/revised stuff?

Still lower level than one would like, but it's hard to be all things to
all languages and go fast.

Keith Bierman
khbkhb@gmail.com
303 997 2749

On Thu, Jul 6, 2017 at 1:00 PM, Van Snyder <Van.Snyder@jpl.nasa.gov> wrote:

> On Thu, 2017-07-06 at 17:52 +0000, Bill Long wrote:
> > Intentionally. The SPMD model, used by both Fortran and MPI, has
> > proved to provide the best scaling and performance in real
> > distributed-memory applications.
>
> MPMD is also useful.  Interactive graphics are not Fortran's strong
> suit.  We use PVM (for historical reasons) to communicate between our
> separately-launched Fortran code instances, and between Fortran and IDL
> for interactive graphics.
>
> I don't see a way to use coarrays for MPMD, but using PVM (and MPI) are
> kind of like programming in assembler.
>
>
> _______________________________________________
> J3 mailing list
> J3@mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3
>

--f403045c825addc5bc0553ab4399
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_default" style=3D"font-family:georgia,=
serif;color:#6633ff">In terms of Standard(s) compliant ways to doing things=
 ... wouldn&#39;t mmap (posix) along with C interop be the &quot;right&quot=
; way to go for new/revised stuff?</div><div class=3D"gmail_default" style=
=3D"font-family:georgia,serif;color:#6633ff"><br></div><div class=3D"gmail_=
default" style=3D"font-family:georgia,serif;color:#6633ff">Still lower leve=
l than one would like, but it&#39;s hard to be all things to all languages =
and go fast.</div></div><div class=3D"gmail_extra"><br clear=3D"all"><div><=
div class=3D"gmail_signature" data-smartmail=3D"gmail_signature"><div dir=
=3D"ltr"><div>Keith Bierman<br><a href=3D"mailto:khbkhb@gmail.com" target=
=3D"_blank">khbkhb@gmail.com</a><br>303 997 2749</div></div></div></div>
<br><div class=3D"gmail_quote">On Thu, Jul 6, 2017 at 1:00 PM, Van Snyder <=
span dir=3D"ltr">&lt;<a href=3D"mailto:Van.Snyder@jpl.nasa.gov" target=3D"_=
blank">Van.Snyder@jpl.nasa.gov</a>&gt;</span> wrote:<br><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex">On Thu, 2017-07-06 at 17:52 +0000, Bill Long wrote:<br>
&gt; Intentionally. The SPMD model, used by both Fortran and MPI, has<br>
&gt; proved to provide the best scaling and performance in real<br>
&gt; distributed-memory applications.<br>
<br>
MPMD is also useful.=C2=A0 Interactive graphics are not Fortran&#39;s stron=
g<br>
suit.=C2=A0 We use PVM (for historical reasons) to communicate between our<=
br>
separately-launched Fortran code instances, and between Fortran and IDL<br>
for interactive graphics.<br>
<br>
I don&#39;t see a way to use coarrays for MPMD, but using PVM (and MPI) are=
<br>
kind of like programming in assembler.<br>
<br>
<br>
______________________________<wbr>_________________<br>
J3 mailing list<br>
<a href=3D"mailto:J3@mailman.j3-fortran.org">J3@mailman.j3-fortran.org</a><=
br>
<a href=3D"http://mailman.j3-fortran.org/mailman/listinfo/j3" rel=3D"norefe=
rrer" target=3D"_blank">http://mailman.j3-fortran.org/<wbr>mailman/listinfo=
/j3</a><br>
</blockquote></div><br></div>

--f403045c825addc5bc0553ab4399--
