Topic 10: Parallel Numerical Algorithms

Description

Robust and efficient parallel numerical algorithms and their implementation into software tools capable of harnessing the processing power of modern computing architectures are crucial components for large-scale computational science and engineering applications. The sessions of this topic will provide a forum for the presentation and discussion of new developments in the field of parallel and distributed numerical algorithms. All aspects of their design and implementation will be addressed, ranging from fundamental algorithmic concepts, to their efficient implementation on modern parallel architectures (including heterogeneous multi-core systems, multi-GPU based computers, clusters and the Grid), to software design and prototyping in scientific computing or simulation software environments, and to performance analysis.

Contributions dealing with the previous aspects are welcome.

Focus

Areas of interest include (but are not limited to):

  • dense and sparse linear algebra
  • partial differential equations
  • ordinary differential and differential algebraic equations
  • domain decomposition
  • optimization
  • nonlinear systems
  • integral equations
  • transforms (wavelets, FFTs, ...)


Topic Committee

Global chair

Daniela di Serafino, Second University of Naples & ICAR-CNR, Italy

Local chair

Luc Giraud, INRIA, France

Vice-chairs

Martin Berzins, University of Utah, USA
Martin Gander, University of Geneva, Switzerland