comp.lang.ada
 help / color / mirror / Atom feed
From: Ray Blaak <rAYblaaK@STRIPCAPStelus.net>
Subject: Re: PAR (Was: Embedded languages based on early Ada)
Date: Wed, 07 Mar 2007 20:17:40 GMT
Date: 2007-03-07T20:17:40+00:00	[thread overview]
Message-ID: <ur6s03j1c.fsf@STRIPCAPStelus.net> (raw)
In-Reply-To: 1l5727owshrjf$.uuylbc4ek430.dlg@40tude.net

"Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> writes:
> Why is it bad programming? Consider this:
> 
> declare
>    Sum : Numeric := 0.0;
> begin in parallel
>    Sum := Sum + Integrate (Series (1..N);
>    Sum := Sum + Integrate (Series (N+1..2*N);
>    Sum := Sum + Integrate (Series (2*N+1..3*N);
>    ...
> end;

This is also bad programming. The assignments are in parallel, overwriting
each other, and it is not clear in a given statement what initial value of Sum
is being worked with.

As shown, this example looks like it is after a sequential summation, and that
contradicts what the concurreny execution will do.

A better program would be:

 declare
    Partial1 : Numeric;
    Partial2 : Numeric;
    Partial3 : Numeric;
    Sum : Numeric;
 begin
   begin in parallel
     Partial1 := Integrate (Series (1..N);
     Partial2 := Integrate (Series (N+1..2*N);
     Partial2 := Integrate (Series (2*N+1..3*N);
   end;
   Sum := Partial1 + Partial2 + Partial3;
   ...
 end;

> > Ada already has synchronization constructs that need to be used to allow
> > dependent concurrent items to coordinate properly.
> 
> If so, then what would be the contribution of PAR? 

A syntactic convenience to tediously writing task bodies explicitly. Consider
the immediate example above written out explicitly.

> One important
> proposition is that there is no any use in PAR running absolutely
> independent code. So the question arise, what *exactly* PAR does with:
> 
> 1. ":=" x 2 times
> 2. "+"
> 3. "-"
> 4. "I" x 4 times

The short answer is nothing. PAR spawns some tasks and waits for them to
complete, nothing more or less.

Parallel programing is hard. Relying on implicit optimization is what I
recommend against, since in general it is a fiendishly complex problem as to
how to sort out the interdependencies.

Keep PAR simple, and leave it to the programmer to figure out the
synchronization.

-- 
Cheers,                                        The Rhythm is around me,
                                               The Rhythm has control.
Ray Blaak                                      The Rhythm is inside me,
rAYblaaK@STRIPCAPStelus.net                    The Rhythm has my soul.



  reply	other threads:[~2007-03-07 20:17 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-02-23  0:59 Preferred OS, processor family for running embedded Ada? Mike Silva
2007-02-23  4:41 ` Steve
2007-02-23 16:00   ` Mike Silva
2007-02-23  4:49 ` Jeffrey R. Carter
2007-02-23 13:13   ` Mike Silva
2007-02-23 13:56 ` Stephen Leake
2007-02-23 14:10   ` Mike Silva
2007-02-24 10:45     ` Stephen Leake
2007-02-24 12:27       ` Jeffrey Creem
2007-02-24 22:10         ` Dr. Adrian Wrigley
2007-02-25 13:10           ` roderick.chapman
2007-02-25 17:53             ` Jeffrey R. Carter
2007-02-25 15:08           ` Stephen Leake
2007-02-28 17:20             ` Embedded languages based on early Ada (from "Re: Preferred OS, processor family for running embedded Ada?") Colin Paul Gloster
2007-03-01  9:18               ` Jean-Pierre Rosen
2007-03-01 11:22               ` Dr. Adrian Wrigley
2007-03-01 11:47                 ` claude.simon
2007-03-01 13:57                 ` Dmitry A. Kazakov
2007-03-01 18:09                   ` Ray Blaak
2007-03-02 11:36                   ` Dr. Adrian Wrigley
2007-03-02 16:32                     ` Dmitry A. Kazakov
2007-03-03  0:00                       ` Dr. Adrian Wrigley
2007-03-03 11:00                         ` Dmitry A. Kazakov
2007-03-03 11:27                           ` Jonathan Bromley
2007-03-03 12:12                             ` Simon Farnsworth
2007-03-03 14:07                               ` Dr. Adrian Wrigley
2007-03-03 17:28                                 ` Pascal Obry
2007-03-03 18:11                                   ` Dmitry A. Kazakov
2007-03-03 18:31                                     ` Pascal Obry
2007-03-03 20:26                                       ` Dmitry A. Kazakov
2007-03-03 21:28                                   ` Dr. Adrian Wrigley
2007-03-03 22:00                                     ` Pascal Obry
2007-03-03 13:40                             ` Dr. Adrian Wrigley
2007-03-03 15:26                               ` Jonathan Bromley
2007-03-03 16:59                                 ` Dr. Adrian Wrigley
2007-03-05 15:36                               ` Colin Paul Gloster
2007-03-03  1:58                       ` Ray Blaak
2007-03-03  8:14                         ` Pascal Obry
2007-03-03 11:00                         ` Dmitry A. Kazakov
2007-03-03 21:13                           ` Ray Blaak
2007-03-05 19:01                             ` PAR (Was: Embedded languages based on early Ada) Jacob Sparre Andersen
2007-03-06  2:01                               ` Dr. Adrian Wrigley
2007-03-06  3:30                                 ` Randy Brukardt
2007-03-06  7:10                                   ` Ray Blaak
2007-03-06 18:05                                     ` Ray Blaak
2007-03-06  6:04                                 ` tmoran
2007-03-06  6:59                                 ` Ray Blaak
2007-03-06  7:07                                 ` Ray Blaak
2007-03-06  7:22                                 ` Martin Krischik
2007-03-06 13:18                                 ` Dr. Adrian Wrigley
2007-03-06 18:16                                   ` Ray Blaak
2007-03-06 23:49                                   ` Randy Brukardt
2007-03-07  8:59                                     ` Dmitry A. Kazakov
2007-03-07 18:26                                       ` Ray Blaak
2007-03-07 19:03                                         ` Dr. Adrian Wrigley
2007-03-07 19:55                                         ` Dmitry A. Kazakov
2007-03-07 20:17                                           ` Ray Blaak [this message]
2007-03-08 10:06                                             ` Dmitry A. Kazakov
2007-03-08 18:03                                               ` Ray Blaak
2007-03-07 20:18                                           ` Pascal Obry
2007-03-07 20:41                                             ` Dr. Adrian Wrigley
2007-03-08  5:45                                               ` Randy Brukardt
2007-03-08 10:06                                                 ` Dmitry A. Kazakov
2007-03-10  1:58                                                   ` Randy Brukardt
2007-03-10  9:11                                                     ` Dmitry A. Kazakov
2007-03-08 18:08                                                 ` Ray Blaak
2007-03-10  1:50                                                   ` Randy Brukardt
2007-03-05 15:23                   ` Embedded languages based on early Ada (from "Re: Preferred OS, processor family for running embedded Ada?") Colin Paul Gloster
2007-03-06  0:31                     ` Dr. Adrian Wrigley
2007-03-01 16:09                 ` Colin Paul Gloster
2007-03-01 13:23               ` Martin Thompson
2007-02-26 16:34           ` Preferred OS, processor family for running embedded Ada? Jean-Pierre Rosen
2007-02-26 21:18             ` Dr. Adrian Wrigley
2007-02-27 15:39               ` Jean-Pierre Rosen
2007-02-28 12:25                 ` Jerome Hugues
2007-02-24 19:11       ` Mike Silva
2007-02-24 13:59     ` Jacob Sparre Andersen
2007-03-01 19:32       ` Jacob Sparre Andersen
2007-03-01 20:22         ` Mike Silva
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox