Ph.D. School in

Scientific GPU Computing

Copenhagen, 26 to 28 May 2010

Ph.D. School in Scientific GPU Computing

As part of the Ph.D. schools ITMAN and DCAMM, DTU Informatics hosts this summer school about utilizing massively parallel processors (GPUs) for desktop scientific computing. Speakers from NVIDIA will talk about latest hardware development, how to program GPUs using CUDA, and how to get the best performance. Speakers from academia will talk about numerics, parallel linear algebra, PDE solvers, and other scientific computing applications of the GPU. The following speakers have been invited to give the course:

Learning Objectives

A student who has met the objectives of the course will be able to:

  • Write CUDA programs.
  • Use CUDA numerics libraries (CUBLAS and CUFFT).
  • Parallelize dense and sparse linear algebra computations.
  • Discretize and parallelize PDE solvers.
  • Solve scientific problems using the GPU.
  • Estimate accuracy vs. speed-up of numeric algorithms running on GPUs.

A student who also completes the follow-up project will in addition be able to:

  • Identify parallelism in a scientific computing problem.
  • Arrange threads for parallel execution.
  • Reduce global memory traffic in device code.
  • Evaluate applicability of massively parallel processing to a specific scientific problem.

News update

Preliminary course program is now available (see Program tab). Breaks will be at the same time as the breaks in the Vision Days program.

NVIDIA is sponsoring a GTX480 graphics card(!) as poster prize for the Graphical Vision Day (which is part of this summer school). Guidelines for submitting a poster can be found here. Please send an email to the poster chairman if you would like to bring a poster to have a chance of winning the prize.

The DCAMM flyer for this event is now online.

This Ph.D. summer school will be collocated with the Vision Days 2010.

Organization

This event is organized by

Registration and practical matters

Ph.D. School in Scientific GPU Computing, Richard Petersens Plads, DTU - Building 321, DK-2800 Lyngby
dcamm@mat.dtu.dk