Denmark: PostDoc and PhD Positions in Scientific GPU Computing, Technical University of Denmark

PostDoc and PhD Positions in Scientific GPU Computing, Technical University of Denmark, Denmark

One Post Doc position (2 years) and two PhD positions (3 years) are available in the area of Scientific GPU Computing, starting April 1, 2010. All three positions are with DTU Informatics (imm.dtu.dk) at the
Technical University of Denmark, and they are financed by a grant from the Danish Council for Independent Research.

The positions are part of the research project “Desktop Scientific Computing on Consumer Graphics Cards.” The aim of this project is to develop a framework for using GPUs and similar many-core hardware accelerators for desktop scientific computing, in collaboration with several international contacts and in such a way that the technology from an early stage of the project is useful in science and industry. The core activity of the project is the development, implementation, and testing of computational algorithms suited for GPU computing.

The Post Doc project involves the development of performance prediction models and a profiler environment, as well as computational modules for selected core computational tasks in scientific computing (in collaboration with the PhD projects). The Post Doc will also be responsible for collaboration and interaction with international research contacts, and for establishing and maintaining a national network between academia and industry.

The two PhD projects focus on computational methods for dynamical optimization and PDE solvers, respectively. The computational bottlenecks in these computations are the dense and sparse matrix
computations, and we aim to develop high-performance algorithms that overcome these bottlenecks using GPU-based computing. The algorithms have applications to other areas such as constrained optimization and computational tomography.

Candidates should have a background in CSE, applied mathematics, or scientific computing and, for the Post Doc position, a PhD degree (or equivalent). Preference will be given to candidates who can
document programming skills in high-performance and/or multi-core scientific computing, and in addition have a background in numerical methods for large-scale problems.

For more information about the positions and how to apply, follow the links on the university’s vacancy home page.

The Scientific Computing section is responsible for teaching in high-performance/scientific computing, and the research includes inverse problems, model-predictive control, high-order methods, appearance modelling, and HPC. We offer a cross-disciplinary research environment with activities in applied mathematics and scientific computing, with contacts to academia and industry.

Contact:
Professor Per Christian Hansen (pch[ at ]imm.dtu.dk)
DTU Informatics, Technical University of Denmark