Atlas home || Conferences | Abstracts | about Atlas

International Conference on Mathematical Modeling and Scientific Computing
April 2-6, 2001
Middle East Technical University and Selcuk University
Ankara and Konya, Turkey

Organizers
F. Bornemann (Munich University of Tecnology, Germany), H. Bulgak (Selcuk University, Konya, Turkey), V. Ganzha (Munich University of Technology, Germany), B. Karasozen (METU, Ankara, Turkey), A. Sinan (Selcuk University, Konya, Turkey), C. Zenger (Munich University of Technology, Germany)

View Abstracts
Conference Homepage

Performance Benefits of Data-Driven Execution in Large Scientific Applications
by
Atilla Gürsoy
Department of Computer Science, Bilkent University, Ankara, Turkey

Software re-use via libraries or modules is a strategy that allows the cost of software to be amortized. Development of large parallel scientific applications can be facilitated if the programming environment supports the ability to develop modules that can be fitted together. Although it is possible to develop parallel reusable modules, it is more difficult to use parallel modules in comparison to sequential ones. Processor idling due to communication delays and load imbalances are among the major factors that affect the performance of parallel programs. Need to optimize performance often forces to sacrifice modularity. This talk focuses on the performance and modularity benefits of data-driven execution for large parallel scientific applications. Within this context, a core design of the parallel molecular dynamics simulation application that employs data-driven execution will be discussed. The design is based on separation of parallelism-control from the application-logic which further reduces the complexity of the parallel program that results in a modular efficient parallel application.

http://www.cs.bilkent.edu.tr/~agursoy

Date received: February 15, 2001

Copyright © 2001 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cagk-63.