Path: utzoo!attcan!uunet!olivea!decwrl!elroy.jpl.nasa.gov!swrinde!zaphod.mps.ohio-state.edu!samsung!emory!hubcap!ncr-sd!ncrwic!wsuiar!hllee
From: ncr-sd!ncrwic!wsuiar!hllee@ucsd.edu
Newsgroups: comp.parallel
Subject: I need some help please on OOPS in Numerical Linear Algebra
Message-ID: <12807@hubcap.clemson.edu>
Date: 28 Jan 91 13:27:44 GMT
Sender: fpst@hubcap.clemson.edu
Organization: Wichita State Univ., Wichita KS
Lines: 31
Approved: parallel@hubcap.clemson.edu
Apparently-To: sdcsvax.ucsd.edu!comp-parallel

Hello,

This may be a very basic question but I hope some of the readers of this 
section can help shed some light to my problem.

Currently, I am working on an OOP on a basic linear algebra software for a
homogeneaus, parallel system.  We do not have any parallel processing 
capability here on campus. Thus what I have to work with is just technical 
papers and other forms of literature. The language with I intend to base the
software on is a form of C++ which is developed by MCC for MIMD or SIMD.

Thus I would appreciate I some of you who have hands on experience on the 
CRAY II , the iWARP or any other supercomputers can share some of your 
experiences with me. 

I am really interested in some of the general information like

1.	Recommendation to some interesting techinical papers regarding	
	OOP linear algebra on a parallel system; which I may not have 
	read.
2.	Some of the benchmarking algorithms and programs used to evaluate
	the performance concentrated arrays on the system.
3.	Some names and e-mail addresses of those who may be involved with
	some project which are directly and indirectly related to mine ...


Regards ......
LEE

in%"hllee@wsuiar.wsu.ukans.edu"