Workshop on General Purpose Computing on Graphics Processors

August 7-8, 2004

Los Angeles, California

OVERVIEW

For years the performance and functionality of graphics processors (GPUs) has been increasing at a faster pace than Moore’s Law (see figure at right). Recently, the major graphics chip manufacturers have added support for floating-point computation and have released compilers for high-level languages.

These GPUs are not like the array processors of the past. First, the prices of these commodity parts are more than an order of magnitude lower—the price of the highest performance graphics cards is only about $350. Furthermore, these chips are in practically every personal computer (PC), game console and workstation sold today.

Heretofore, the primary application of GPUs has been fast rendering of anti-aliased, textured and shaded geometric primitives (e.g. polygons). Their main market has been mostly computer games and entertainment business.

The performance and functionality of today’s GPUs make them attractive as co-processors for general-purpose computations. Over the last few years, many new algorithms and applications that exploit the inherent parallelism and vector processing capabilities of GPUs have been proposed. These include:

  • Scientific computation including FFTs, linear algebra solvers, differential equation solvers, multi-grid solvers and applications to fluid dynamics, visual simulation, ice crystal growth, etc.
  • Geometric computations including Voronoi diagrams, distance computation, motion planning, collision detection, object reconstruction, visibility, etc.
  • Advanced rendering including ray-tracing, radiosity, photon mapping, sub-surface scattering, shadows, etc.
  • Database operations including aggregates, predicates, Boolean combinations, selection queries, etc. on large databases.

See www.gpgpu.org for more detailed examples and recent work in these areas.

Given the increasing power and usage of commodity GPUs, this workshop will explore general purpose computing using GPUs. Some of the issues include:

  • Do GPUs have the potential of being a useful co-processor for a wide variety of applications? What are their algorithmic and architectural niches and can these be broadened?
  • What are the major issues in terms of programmability, language and compiler support and software environments for GPUs?
  • What are some of the future technology trends that can lead to more widespread use of GPUs?

This workshop will bring together leading researchers and practitioners from academia, research labs and industry working in computer graphics, scientific computation, high performance computing, computer architecture and related areas. The program will consist of invited talks, panels and poster presentations.

ORGANIZERS

Anselmo Lastra (UNC Chapel Hill)
Ming C. Lin (UNC Chapel Hill)
Dinesh Manocha (UNC Chapel Hill)

STEERING COMMITTEE

Andrew Chien (UC San Diego)
William Dally (Stanford)
Bob Graybill (DARPA)
Andrew Gruber (ATI)
Pat Hanrahan (Stanford)
Phil Heermann (Sandia)
Ari Kaufmann (SUNY Stony Brooks)
David Kirk (NVIDIA)
Fred Kitson (HP Labs)
Michael Macedonia (PEO STRI)
Dominic Mallinson (Sony)
William Pulleyblank (IBM)
Dan Reed (Illinois & UNC Chapel Hill)
Peter Schršder (Caltech)
Mark Swinson (Army Research Office)

Tentative Workshop Program

Comments Off