TI A Methodology for the Design and Implementation of Efficient
Algorithms for Scalable Parallel Architectures 

LT 91-018 

YR 1991 

AU Decker, Karsten M.  

OR BERN 

AV ftp iam.unibe.ch:TechReports1991iam-91-018.ps.Z 

AB Parallel programs exploiting as much as possible of the
nominal performance of distributed memory parallel processor
systems not only require efficient implementations but in
particular careful algorithm design.  In this paper we propose a
design and implementation methodology for parallel algorithms
incorporating the concept of literate programming and report on
the experience gained in the development of a prototype library. 
The investigation is part of the SPADE project which aims at the
development of an integrated program and application development
environment for scientific applications on parallel architectures
with distributed memory.