2001 TUTORIAL 2
Monday, April 23rd,
8 AM - 12 Noon
Introduction to Effective Parallel Computing
Quentin F. Stout and Christiane Jablonowski
University of Michigan
WHO SHOULD ATTEND
New parallel computing users, as well as students, potential users, managers, and anyone who wants an overview of parallel programming.
75% Introductory | 25% Intermediate
This tutorial provides a comprehensive overview of parallel computing, emphasizing those aspects most relevant to the user. It discusses both hardware and software, with an emphasis on standards, portability, and systems that are now (or soon will be) commercially or freely available. Systems examined range from low-cost clusters to highly integrated supercomputers. The tutorial surveys basic concepts and terminology, and gives parallelization examples selected from engineering, scientific, and data intensive applications. These real-world examples are targeted at distributed memory systems, using
MPI, and shared memory systems using OpenMP. The tutorial shows basic parallelization approaches, step-by-step performance improvement, and discusses some of the software engineering aspects of the parallelization process. The tutorial also provides pointers to the
literature and web-based resources.
Quentin F. Stout is Professor of Electrical Engineering and Computer Science, and director of the Center for Parallel Computing, at the University of Michigan. In over 15 years of research in parallel computing, he has published over 100 articles, book chapters, and books on the topic and has utilized a range of systems for a wide spectrum of applications. He has won Best Presentation and Best Paper awards, and two of his doctoral students have won Best Thesis awards for their work in parallel computing. He and his students have had research and consulting funded by DARPA, NSF, NASA, NRL, Digital, IBM, ATT, Unisys, Xerox, Ford Motor, and Dow Research.
Christiane Jablonowski is a meteorologist currently pursuing her Ph.D. in Atmospheric Science and Scientific Computing at the University of Michigan. She has over 10 years of experience in large-scale applications on parallel and parallel vector machines. She has worked at Siemens/Fujitsu; the "Vector and Parallel Computing" section at Aachen University of Technology, Germany; and the European Center for Medium-Range Weather Forecasts in Reading, England. Her research has been focused on climate and weather prediction modeling, with strong emphasis on parallel computing aspects. At the German National Research Center for Information Technology she won the Best Thesis award.