Newsgroups: comp.parallel.pvm
Path: ukc!uknet!liv!liverpool.ac.uk!bruce
From: bruce@liverpool.ac.uk (Bruce Stephens)
Subject: Re: PVM for Transputers
In-Reply-To: iinfd@cc.uab.es's message of 26 Apr 94 12:37:56 +0200
Message-ID: <BRUCE.94Apr27121752@cmsr3.liverpool.ac.uk>
Sender: news@liverpool.ac.uk (News System)
Nntp-Posting-Host: cmsr3.scm.liv.ac.uk
Organization: Centre for Mathematical Software Research, Univ. Liverpool
References: <1994Apr26.123756.1@cc.uab.es>
Date: Wed, 27 Apr 1994 11:17:52 GMT
Lines: 19

>>>>> On 26 Apr 94 12:37:56 +0200, iinfd@cc.uab.es said:

> I am working on the development of a parallel programming environment
> for Transputers. I am interested on a implentation of PVM for 
> Trasnputer-based systems. Specially, if exists a PVM for the machine
> Supernode SN-1000 from Parsys.

As far as I'm aware: no.  PARMACS certainly exists; version 5.1 can be
got from unix.hensa.ac.uk.  PVM strikes me as technically feasible,
using the VCR approach used in PARMACS, but as far as I'm aware it
hasn't been done yet.

Please let us all know if I'm wrong: we've got an underused supernode
too, and it would be lovely to be able to use it to benchmark PVM
code.

--
Bruce                    Institute of Advanced Scientific Computation
bruce@liverpool.ac.uk    University of Liverpool

