Optimizing Smith-Waterman algorithm on Graphics Processing Unit


Local Sequence alignment is an important task for bioinformatics. The most widely used algorithm is Smith-Waterman [1] has a quadratic time complexity which is time consuming especially in large biological database search. Many attempts were made to accelerate Smith-Waterman using parallel architecture. In this paper a parallel implementation of Smith Waterman algorithm will be presented. This implementation will run on Graphics Processing Unit using an open standard (OpenCL) by carefully exploiting features of the graphics processing unit, this implementation will be more than two times faster than the other parallel implementation on the Graphics Processing Unit.

2 Figures and Tables

Cite this paper

@article{Khalafallah2010OptimizingSA, title={Optimizing Smith-Waterman algorithm on Graphics Processing Unit}, author={Ayman Khalafallah and H. F. Elbabb and Omar Mahmoud and Ahmed Elshamy}, journal={2010 2nd International Conference on Computer Technology and Development}, year={2010}, pages={650-654} }