The IASTED International Conference on
Parallel and Distributed Computing and Networks
~PDCN 2008~
as part of the 26th IASTED International Multi-Conference on APPLIED INFORMATICS
February 12 – 14, 2008
Innsbruck, Austria
TUTORIAL SESSION
Programming Cell Processors and GPUs for High-Performance Simulations
Abstract
Cell processors and graphics processing units have evolved into very attractive hardware platforms for general-purpose computations due to their extremely high floating-point processing performance, huge memory bandwidth, and their comparatively low cost. The rapid evolution of Cells and GPUs in performance, architecture, and programmability can provide application potential beyond their primary purpose of graphics processing. This tutorial will explore programming the Cell processor and GPUs to achieve high performance for simulations using the IBM SDK and alternative software platforms such as CUDA or RapidMind. We will provide an overview of the Cell and GPU architecture, emphasizing hardware and software features exploited by the programming model used when creating or porting applications to the Cells/GPUs. Additionally, simulation programming examples detailing the use of both the IBM and alternative software development environment will be given. The speakers will provide hands-on tips and techniques as well as an overview of performance tools for achieving maximum Cell and GPU performance.
Objectives
The tutorial is lecture-style and open to all levels of attendees. It provides an overview of the basic concepts behind Cell and GPUs and will provide performance results for HPC simulations.
Biography of the Presenter
Olaf Schenk received his M.S. degree in Applied Mathematics and Computer Science from the University of Karlsruhe, Germany in 1996, and his Ph.D. degree in Technical Sciences from the Swiss Federal Institute of Technology (ETH), Zurich, Switzerland in 2000. Since 2001 he has been a Research Associate in the Computer Science Department of the University of Basel, Switzerland. His general research interests are in high-performance computing and applied informatics, with emphasis on algorithms in simulation and optimization. Schenk pays particular attention to nonlinear optimization, discrete mathematics, solving large systems of linear equations, and solving differential equations. He is chiefly interested in solving large-scale problems that involve information and communication technologies on high-performance computing architectures.
Matthias Christen received his M.S. degree in Mathematics from the University of Basel, Switzerland in 2006. Currently, he is a Ph.D. student in Computer Science at the University of Basel. His research interests are in computational science with emphasis on simulation and optimization, as well as high-performance computing.
