Newsgroups: comp.parallel.mpi
From: romero@convex.com (Paco Romero)
Subject: Re: Memory Channel: Message Passing Latency 5.8 usecs, ~60 MB/sec @ 4KB messages
Organization: Hewlett Packard - Convex Division, Richardson, Tx USA
Date: 20 Aug 1996 02:30:04 -0500
Message-ID: <4vbphs$nqn@bach.convex.com>

In article <4vbdah$66m@s4leia1.kcc.co.jp>,  <tagata@kcc.co.jp> wrote:
>
> ... performance of HP MPI deleted ...
>
>I am interesting in a Scalable Parallel Computer. It's a nice speed. 
>I would like to know where I can find the Dongarra's ping-pong code.
>
>Thanks.
>
>-----------------
>TAGATA, Kazuo
>E-mail : tagata@kcc.co.jp
>

Hello Tagata-San,
   I am delighted to know that you want to learn more about HP's scalable
parallel systems.  Please check the SPP page at the HP technical computing 
site:
          http://www.hp.com/wsg/products/spp.html

which will give you an overview of the SPP1600 family.   Our group is putting
together an HP MPI web page that will be available as part of the HP Technical
Computing Web server later this month.  

Dongarra's ping-pong code is available from NETLIB (http://www.netlib.org/benchmark)
under the title "benchmark/comm.shar."  Please let me know if you want me to
email you a copy, it is very short.

Note that the ping-pong code is a synthetic benchmark and as such it might not
reflect the performance characteristics of your own message-passing application.

Best regards,

Paco Romero (romero@rsn.hp.com)
Explicit Parallel Models Group, Project Manager
Convex Division, Hewlett-Packard.

