![[logo]](Pictures/apparc.gif)
Performance Critical Applications of Parallel Architectures (APPARC)
ESPRIT EC-DGIII Basic Research Project Number:
6634
Work Area: Parallel Computing and Architectures
APPARC consortium
- UNI-C, Copenhagen, Denmark
- ONERA, Chatillon, France
- KFA,
Jülich, Germany
- Leiden University,
Leiden, The Netherlands
- Queen's University,
Belfast, Northern Ireland
- University of Patras, Patras, Greece
- IRISA, Rennes, France
- University of
Manchester, Manchester, England
- Polytechnic University of
Catalonia, Barcelona, Spain
Coordinator
Harry Wijshoff
Leiden University
Niels Bohrweg 1
2333 CA Leiden, the Netherlands
Contact person
Fred Bakker
Dept. of Computer Science
Leiden University
Niels Bohrweg 1
2333 CA Leiden
The Netherlands
phone: +31 71 5277042
fax: +31 71 5276985
email: bakker@cs.leidenuniv.nl
Keywords
Performance-Critical Components
Sparse Applications and Algorithms
Memory Organisation
Timetable
From 24/07/1992 to 24/09/1995
Synopsis
High performance computing is a critical technology for future economic
growth. Technical progress in high performance computing has been rapid
only in a few application areas. Unfortunately, a large and important
class of applications is not yet amenable to high performance execution
due to a mismatch with currently available techniques at all levels of
the computational problem solving process. We refer to this class of
computations as performance-critical applications. It requires a concerted
interdisciplinary research effort to break this barrier. The APPARC project
addresses this urgent research need.
Aims
Performance-critical applications typically involve manipulation of large,
sparse discrete data objects. In our view, one of the major causes of low
performance for these applications is the hardware memory organisation in
present-day high performance computers. Progress towards general applicability
of high performance computing depends upon the removal of this memory
performance barrier for the class of sparse computations. Hence, sparse
computations and hardware memory architecture, form the main technical
themes of the APPARC project.
Approach and Methods
The development of computational solutions to performance--critical
applications embraces many different levels of abstraction, from
mathematical modelling of the application, through algorithm and
software development, to hardware implementation.
Hence, tackling the two project themes is not a simple matter:
the interaction between the applications and the memory
architecture affects all levels of the problem solving process.
Consequently, the APPARC project intends to study the two themes
across eight interrelated enabling areas, as follows:
Performance-Critical Applications, Parallel Algorithms,
Problem Solving Environments, Performance Modelling and Evaluation,
High Level Languages, Compiler Design, Operating Systems,
Hardware Architecture.
The deliverables produced by the APPARC consortium as of yet are listed
here.
Newsletter
Every three months the APPARC Newsletter is released.
The purpose of this newsletter is to inform the research and
industrial community at large about the progress being made
within the APPARC consortium.
Released issues:
Related publications
The related publications describe research results obtained by the
APPARC partners. More than 100 papers in journals and conference
proceedings have been published.
A list of these publications (with abstracts) is given
in the Newsletters. For more information concerning these APPARC related
publications the following persons serve as contact persons per partner:
Last modified on July 2, 1996 by Lex Wolters.