High performance protein sequence database scanning on the Cell Broadband Engine

Issue title: High Performance Computing with the Cell Broadband Engine

Article type: Research Article

Authors: Wirawan, Adrianto | Schmidt, Bertil | Zhang, Huiliang | Kwoh, Chee Keong

Affiliations: School of Computer Engineering, Nanyang Technological University, Singapore. E-mails: {adri0004, asbschmidt, hlzhang, asckkwoh}@ntu.edu.sg

Abstract: The enormous growth of biological sequence databases has caused bioinformatics to be rapidly moving towards a data-intensive, computational science. As a result, the computational power needed by bioinformatics applications is growing rapidly as well. The recent emergence of low cost parallel multicore accelerator technologies has made it possible to reduce execution times of many bioinformatics applications. In this paper, we demonstrate how the Cell Broadband Engine can be used as a computational platform to accelerate two approaches for protein sequence database scanning: exhaustive and heuristic. We present efficient parallelization techniques for two representative algorithms: the dynamic programming based Smith–Waterman algorithm and the popular BLASTP heuristic. Their implementation on a Playstation®3 leads to significant runtime savings compared to corresponding sequential implementations.

Keywords: Heterogeneous multi-cores, Cell BE, bioinformatics, dynamic programming, BLAST

DOI: 10.3233/SPR-2009-0274

Journal: Scientific Programming, vol. 17, no. 1-2, pp. 97-111, 2009