Newsgroups: comp.parallel.pvm
From: stellner@Informatik.TU-Muenchen.DE (Georg Stellner)
Subject: Re: Patch #8 for PVM version 3.3 is available on Netlib
Organization: Technische Universitaet Muenchen, Germany
Date: 3 Aug 1995 09:05:23 GMT
Message-ID: <3vq3gj$l2k@sunsystem5.informatik.tu-muenchen.de>


In article <DCoMyu.BIx@cs.dal.ca>, edemaine@ug.cs.dal.ca (Erik Demaine) writes:
|> Al Geist (geist@msr.epm.ornl.gov) wrote:
|> :    This is correct. IBM's version is called MPI-F. This is what we built
|> :    the PVM 3.3.8 SP-2 port on top of. In our tests PVM 3.3.8 is faster
|> :    than PVMe, but your mileage may vary.
[...]
|> Thanks Al.  An alternative is to use MPICH (as I am doing now).
|> 
|> If PVM is implemented on MPI, I guess it can now work on all MPI machines.
|> Is this correct?
Another interesting question would be, how the implementation of PVM on
top of MPI handles dynamic process creation with pvm_spawn. The MPI 
standard definition doesn't include process creation. Or does it mean,
PVM ontop of MPI doesn't offer dynamic processes?

Cheers, Georg

-- 
Georg Stellner                   stellner@informatik.tu-muenchen.de+------------
Institut fuer Informatik, SAB    phone:  +49-89-2105-2689          | (())_-_(())
Technische Universitaet Muenchen fax:    +49-89-2105-8232          |  | (* *) |
D-80290 Muenchen                 room:   S 1211                    | {  \_@_/  }
http://wwwbode.informatik.tu-muenchen.de/~stellner                 |   `-----' 

