SC Conference - Activity Details

Harnessing Associative Computing for Sequence Alignment with Parallel Accelerators

Shannon I. Steinfadt  (Kent State University)
Doctoral Research Showcase Session
Thursday,  04:00PM - 04:15PM
Room 17A/17B
Sequence alignment is one of the most common operations used in computational molecular biology. It is used to better understand DNA's functionality. Fast heuristic alignment algorithms such as BLAST are not always adequate and the much more rigorous, albeit much slower Smith-Waterman algorithm is an invaluable tool. This research aims to provide a faster running, Smith-Waterman-like algorithm with increased information output. This associative algorithm is extensible to multiple SIMD-like platforms. By providing the functionality for associative algorithms to run on NVIDIA General Purpose Graphical Processing Units known as GPGPUs and the ClearSpeed CSX chipset, it also opens up an entire library of parallel associative computing algorithms, including bin-packing, string matching, and computational geometry that can be adapted to run on these modern run-time platforms.
   IEEE Computer Society  /  ACM     2 0   Y E A R S   -   U N L E A S H I N G   T H E   P O W E R   O F   H P C