Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!sun-barr!decwrl!decvax!ima!esegue!compilers-sender
From: deb@quimby.dartmouth.edu (Deb Banerjee)
Newsgroups: comp.compilers
Subject: Re: compiling for parallel machines
Message-ID: <1989Aug4.180349.3480@esegue.uucp>
Date: 4 Aug 89 18:03:49 GMT
Sender: compilers-sender@esegue.uucp
Reply-To: deb@quimby.dartmouth.edu (Deb Banerjee)
Organization: Compilers Central
Lines: 39
Approved: compilers@esegue.segue.bos.ma.us

Your suspicion about software rather than hardware being the challenge in
parallel programming systems may be correct.

I am working on high level programming languages & systems for SIMD machines
(I have the Connection Machine in mind).

References:

"Partitioning and Scheduling parallel programs for multiprocessors"
Vivek Sarkar, MIT press 1989.
--based on his Ph.D thesis (1987).


"Optimizing Supercompilers for supercomputers"
Michael Wolfe, MIT press 1989.
--deals with dependence analysis etc.


IEEE Software July 1989 - is a special issue on Parallel Programming
environments and languages.

The state of the art in parallel languages is rather poor. The languages are
too low level - puts more burden on the programmer and the excessive detail
the programmer specifies prevents program optimizations; portability is out
of the question.

There has been quite a bit of work on Parallel programming environments.
Doesn't look very elegant most of the time.

Could you send me any other info you get.  Hope you find this useful.

Thanks
--Deb
[From deb@quimby.dartmouth.edu (Deb Banerjee)]
-- 
--
Send compilers articles to compilers@ima.isc.com or, perhaps, Levine@YALE.EDU
{ decvax | harvard | yale | bbn }!ima.  Meta-mail to ima!compilers-request.
Please send responses to the author of the message, not the poster.