Accelerating Smith-Waterman Local Sequence Alignment on GPU Cluster
Department of Information Systems, Hanoi University of Science and Technology, Hanoi, Vietnam
Annual International Conference on Advances in Distributed and Parallel Computing (ADPC2010), Mandarin Orchard, Singapore, 2010
@article{nguyenaccelerating,
title={Accelerating Smith-Waterman Local Sequence Alignment on GPU Cluster},
author={Nguyen, T.T. and Nguyen, D.H. and Pham, P.H. and Ta, N.M. and Duong, T.N. and Le, H.D. and Center, H.P.C.},
booktitle={Annual International Conference on Advances in Distributed and Parallel Computing (ADPC2010), 2010},
year={2010}
}
With a high accuracy, the Smith-Waterman local sequence alignment algorithm requires a very large amount of memory and computation, making implementations on common computing systems become less practical. In this paper, we present swGPUCluster – an implementation of the SmithWaterman algorithm on a cluster equipped with NVIDIA GPU graphics cards (called a GPU cluster). Our test was performed on a cluster of two nodes, one node is equipped with a dual graphics card NVIDIA GeForce GTX 295, a Tesla C1060 card, and the remaining node is equipped with 2 dual graphics cards NVIDIA GeForce GTX 295. Results show that the performance has increased significantly compared with the previous best implementations such as SWPS3 or CUDASW++. The performance of swGPUCluster has increased along with the lengths of query sequences, from 37.328 GCUPS to 46.706 GCUPS. These results demonstrate the great computing power of graphics cards and their high applicability in solving bioinformatics problems.
June 27, 2011 by hgpu
