Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Bio-sequence database scanning on a GPU

: Liu, W.G.; Schmidt, B.; Voss, G.; Schröder, A.; Müller-Wittig, W.K.

Großpietsch, K.-E.:
11th IEEE Workshop on Dependable Parallel, Distributed and Network-Centric Systems, DPDNS 2006 : April 25-29, 2006, Rhodes Island, Greece, held in conjunction with the 20th IEEE International Parallel & Distributed Processing Symposium
Los Alamitos, Calif.: IEEE, 2006
ISBN: 1-424-40054-6
8 S.
Workshop on Dependable Parallel, Distributed and Network-Centric Systems (DPDNS) <11, 2006, Rhodes>
International Parallel and Distributed Processing Symposium (IPDPS) <20, 2006, Rhodes>
Fraunhofer IGD ()
high performance computing; graphic hardware; dynamic programming; sequence alignment; general purpose computation on graphics processing unit (GPGPU)

Protein sequences with unknown functionality are often compared to a set of known sequences to detect functional similarities. Efficient dynamic programming algorithms exist for this problem, however current solutions still require significant scan times. These scan time requirements are likely to become even more severe due to the rapid growth in the size of these databases. In this paper, we present a new approach to bio-sequence database scanning using computer graphics hardware to gain high performance at low cost. To derive an efficient mapping onto this type of architecture, we have reformulated the Smith- Waterman dynamic programming algorithm in terms of computer graphics primitives. Our OpenGL implementation achieves a speedup of approximately sixteen on a high-end graphics card over available straightforward and optimized CPU Smith-Waterman implementations.