Document Type

Conference Paper

Department

Computer Science and Engineering

Publication Date

2006-01-01

Filename

wucse-2006-21.pdf

Technical Report Number

WUCSE-2006-21

Abstract

The amount of biosequence data being produced each year is growing exponentially. Extracting useful information from this massive amount of data is becoming an increasingly difficult task. This thesis focuses on accelerating the most widely-used software tool for analyzing genomic data, BLAST. This thesis presents Mercury BLAST, a novel method for accelerating searches through massive DNA databases. Mercury BLAST takes a streaming approach to the BLAST computation by offloading the performance-critical sections onto reconfigurable hardware. This hardware is then used in combination with the processor of the host system to deliver BLAST results in a fraction of the time of the general-purpose processor alone. Mercury BLAST makes use of new algorithms combined with reconfigurable hardware to accelerate BLAST-like similarity search. An evaluation of this method for use in real BLAST-like searches is presented along with a characterization of the quality of results associated with using these new algorithms in specialized hardware. The primary focus of this thesis is the design of the ungapped extension stage of Mercury BLAST. The architecture of the ungapped extension stage is described along with the context of this stage within the Mercury BLAST system. The design is compact and performs over 20× faster than that of the standard software ungapped extension, yielding close to 50× speedup over the complete software BLAST application. The quality of Mercury BLAST results is essentially equivalent to the standard BLAST results.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7GQ6W4N

Share

COinS