Solving The Longest Overlap Region Problem for Noncoding DNA Sequences with GPU (1404.3448v2)

Abstract: Early hardware limitations of GPU (lack of synchronization primitives and limited memory caching mechanisms) can make GPU-based computation inefficient. Now Bio-technologies bring more chances to Bioinformatics and Biological Engineering. Our paper introduces a way to solve the longest overlap region of non-coding DNA sequences on using the Compute Unified Device Architecture (CUDA) platform Intel(R) Core(TM) i3- 3110m quad-core. Compared to standard CPU implementation, CUDA performance proves the method of the longest overlap region recognition of noncoding DNA is an efficient approach to high-performance bioinformatics applications. Studies show the fact that efficiency of GPU performance is more than 20 times speedup than that of CPU serial implementation. We believe our method gives a cost-efficient solution to the bioinformatics community for solving longest overlap region recognition problem and other related fields.