From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: 107079,9d9cef47f5d1976c
X-Google-Thread: 103376,ec0aac9465177ad0,start
X-Google-Attributes: gid107079,gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news2.google.com!postnews.google.com!c28g2000cwb.googlegroups.com!not-for-mail
From: "mrohan@ACM.ORG" <mrohan@ACM.ORG>
Newsgroups: sci.math.num-analysis,comp.lang.ada
Subject: Re: Multithreaded scientific programming
Date: 4 Oct 2006 14:47:02 -0700
Organization: http://groups.google.com
Message-ID: <1159998421.978196.161210@c28g2000cwb.googlegroups.com>
References: <1159978124.458442.80520@m7g2000cwm.googlegroups.com>
   <4523e925$1@nntp.unige.ch>
NNTP-Posting-Host: 148.87.1.171
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
X-Trace: posting.google.com 1159998427 11598 127.0.0.1 (4 Oct 2006 21:47:07
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Wed, 4 Oct 2006 21:47:07 +0000 (UTC)
In-Reply-To: <4523e925$1@nntp.unige.ch>
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux i686; en-US;
 rv:1.7.13) Gecko/20060418 Fedora/1.0.8-1.1.fc4
 Firefox/1.0.8,gzip(gfe),gzip(gfe)
X-HTTP-Via: 1.1 inet-nc02 (NetCache NetApp/5.6.2R1D19)
Complaints-To: groups-abuse@google.com
Injection-Info: c28g2000cwb.googlegroups.com; posting-host=148.87.1.171;
   posting-account=gAOdzgwAAACn-YRtA_sS8eamp4mn5Zlu
Xref: g2news2.google.com sci.math.num-analysis:6270 comp.lang.ada:6873
Date: 2006-10-04T14:47:02-07:00
List-Id: <comp.lang.ada>

DP wrote:
> Rob McDonald wrote:
> ...
> > To go along with this, what old algorithms have been sitting ignored in
> > the back of textbooks because they need a ton of processors to be
> > efficient?  Maybe it is time to dust them off...
> >
>
> Some attempts to address multi-threaded parallelism that might be suited
> to future 80 cores processors:
>
> Nested data parallelism : http://www.cs.cmu.edu/~scandal/cacm/node4.html
>
> Cilk : http://supertech.lcs.mit.edu/cilk/
>
>
> Dan

Hi,

Since the choice of language is open, I would recommend investigating
using Ada 2005.  Ada has had tasks (mutlithreading) since it's creation
in 1983 and polymorphic object oriented features since the 1995
revision.

Numerical codes tend to be large and Ada's features for programming in
the large should certainly help.  Also, due to it's strong typing
system and overall emphasis on correctness, once something is
implemented in Ada that works, it's generally the "correct" solution.
I've frequently encountered people making updates to numerical code (in
C/Fortran) and summarizing their work with "I'm not sure what I did,
but it works now and I'm not going to touch again".

I've included comp.lang.ada on this thread as I'm sure there are people
there who could give you more advise.

Take care,
Michael.