Newsgroups: comp.parallel.mpi,comp.parallel.pvm
From: Yon Han CHONG <Y.H.Chong@cranfield.ac.uk>
Subject: Re: mpi vs. pvm for a multiprocessor computer
Organization: Cranfield University
Date: Tue, 03 Mar 1998 13:37:28 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=EUC-KR
Content-Transfer-Encoding: 7bit
Message-ID: <34FC0798.7571B376@cranfield.ac.uk>

Lee, Bo-sung wrote:
> 
> The performance between PVM and MPI is negligible in the
> same machine. But, if you want to use dynamic process
> management, then pvm is more powerful than mpi because
> mpi is more proper in SPMD program model.
> 
> If you spawn(create) child processes dynamically
> in your host(master) process, use pvm.
> 
[...]

Even though MPI standard 1 does not specify spawning, standard 2 does.
As far as I know there is no implementation available yet with full 
standard 2 specification but some already do have spawning implemented, 
for example LAM6.1.

  http://www.osc.edu/lam.html

-- 
----------------------------------------------------------------------
Yon Han Chong                     |
Flow Control and Prediction Group | Tel: + 44 1234 750 111 ext. 5420
Cranfield College of Aeronautics  | E-mail: Y.H.Chong@cranfield.ac.uk
Cranfield University              | Fax: + 44 1234 752 149
United Kingdom                    | Mobile: + 44 958 381 395
----------------------------------------------------------------------