Parallel mapping algorithms for a novel mapping & configuration software for the FACETS project

Authors:
Matthias Ehrlich;Karsten Wendt;René Schüffny
Affiliations:
Technische Universität Dresden, Institute of Circuits and Systems, Dresden, Germany;Technische Universität Dresden, Institute of Circuits and Systems, Dresden, Germany;Technische Universität Dresden, Institute of Circuits and Systems, Dresden, Germany
Venue:
CEA'08 Proceedings of the 2nd WSEAS International Conference on Computer Engineering and Applications
Year:
2008

Citing 3
Cited 0

Dynamic Instrumentation of Large-Scale MPI and OpenMP Applications

IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Brook for GPUs: stream computing on graphics hardware

ACM SIGGRAPH 2004 Papers
A software framework for tuning the dynamics of neuromorphic silicon towards biology

IWANN'07 Proceedings of the 9th international work conference on Artificial neural networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents results and work progress on development of parallel mapping algorithms for the FACETS project. We will give an introduction to the FACETS hardware system and its software framework accordingly. A survey of state-of-the-art and emerging parallelization paradigms as well as software performance measurement methods was added to reasonably apply these methods to now a days cluster architectures and rate the implemented softwares performance. The contribution also proposes a novel clustering principle, the Net Layout Organizer which is therefore described in detail.