Topic of the Workshop

In recent years, programming high performance computing systems has shifted from single layer programming models to multiple layers of massively parallel systems. Widely used message-passing or thread programming models do not longer map efficiently to upcoming heterogeneous systems. Further, different brands of graphics cards (GPU) and accelerators, as well as large numbers of processor (CPU) cores, each with increasing vector computing capability require dedicated computational kernels. However, topics like code optimization and portable code development remain challenging.

This workshop will focus on aspects of heterogeneous computing systems. This includes GPU computing, accelerators offloading, vector instruction sets and the multi level hierarchy of distributed and shared memory resources found in large scale computing.

The first day of the workshop is dedicated to an introduction to programming aspects. A tutorial including a hands-on session will give the opportunity for less experienced participants to learn about GPU computing.
The next day of the workshop will discuss how physics applications, especially in relativity and quantum field theory, may benefit from the use of accelerators. The topics covered by this workshops are:

  • Programming techniques, library and language design for heterogeneous systems. Issues of portability and performance.
  • Parallel numerical algorithms and algorithms tailored or tuned for a specific system.
  • Optimization of computational kernels. Tools and techniques for high performance, energy efficiency, or effective code development.

Abstracts should be submitted by end of April 2014 through the registration page.

Organizers

The workshop is organized by

  • B. Brügmann (University Jena)

  • X. Cai (Simula and University Oslo)

  • G. Haase (University Graz)

  • G. Zumbusch (Chair, University Jena)