home · mobile · calendar · colloquia · 2009-2010 · 

Colloquium - Burtscher

Towards a Science of Parallel Programming
University of Texas at Austin
3/18/2010
3:30pm-4:30pm

When parallel programming started in the 70s and 80s, it was mostly art: languages such as functional and logic programming languages were designed and appreciated mainly for their elegance and beauty. More recently, parallel programming has become engineering: conventional languages like FORTRAN and C++ have been extended with constructs such as OpenMP, and we now spend our time benchmarking and tweaking large programs nobody understands to obtain performance improvements of 5-10%. In spite of all this activity, we have few insights into how to write parallel programs to exploit the performance potential of multicore processors.

To break this impasse, we need a science of parallel programming. In this talk, I will introduce a concept called "amorphous data-parallelism" that provides a simple, unified picture of parallelism in a host of diverse applications ranging from mesh generation/refinement/partitioning to SAT solvers, maxflow algorithms, stencil computations, and event-driven simulation. Then I will present a natural classification that provides insight into the structure of parallelism and locality in these algorithms and into appropriate language and systems support for exploiting this parallelism.

Martin Burtscher received the combined BS/MS degree in computer science from the Swiss Federal Institute of Technology (ETH) Zurich in 1996 and the PhD degree in computer science from the University of Colorado Boulder in 2000. Since then, he has been an assistant professor in the School of Electrical and Computer Engineering at Cornell University and a Research Scientist in the Institute for Computational Engineering and Sciences at the University of Texas at Austin. His current research focuses on automatic parallelization of irregular programs for multicore and GPU architectures as well as on automatic performance assessment and optimization of HPC applications. He is an associate editor of the Journal of Instruction-Level Parallelism and a senior member of the IEEE, its Computer Society, and the ACM.

Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
webmaster@cs.colorado.edu
www.cs.colorado.edu
May 5, 2012 (14:13)
XHTML 1.0/CSS2
©2012