Digital Library

cab1

 
Title:      A SYSTOLIC ARRAY ARCHITECTURE FOR THE SMITH-WATERMAN ALGORITHM WITH HIGH PERFORMANCE CELL DESIGN
Author(s):      Laiq Hasan , Yahya M. Khawaja , Abdul Bais
ISBN:      978-972-8924-63-8
Editors:      Hans Weghorn and Ajith P. Abraham
Year:      2008
Edition:      Single
Keywords:      Bioinformatics, Sequence Alignment, Dynamic Programming, Smith-Waterman Algorithm, FPGAs, Systolic Arrays
Type:      Full Paper
First Page:      35
Last Page:      42
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      To infer homology and subsequently gene function, the Smith-Waterman (S-W) algorithm is used to find the optimal local alignment between two sequences. When searching sequence databases that may contain hundreds of millions of sequences, this algorithm becomes computationally expensive. In this paper, we present a systolic array architecture for the S-W algorithm with a new high performance cell design. The results demonstrate that the implementation of this architecture achieves a speedup of up to 652x, as compared to a software-only implementation, which is almost double the best case reported in the literature. The results also demonstrate that when mapped on the same FPGA platform, our design performs 1.47 to 5.75 times faster in terms of Cell Updates Per Second (CUPS), in comparison with other published systolic array designs, while utilizing 3.69 to 6.36 times less resources.
   

Social Media Links

Search

Login