Accelerating String Matching on MIC Architecture for Motif Extraction

摘要

Identifying repeated factors that occur in a string of letters or common factors that occur in a set of strings represents an important task in computer science and biology. Such patterns are called motifs, and the process of identifying them is called motif extraction. In biology, motifs may correspond to functional elements in DNA, RNA, or protein molecules. In this article, we orchestrate MoTeX, a high-performance computing tool for MoTif extraction from large-scale datasets, on Many Integrated Core (MIC) architecture. MoTeX uses state-of-the-art algorithms for solving the fixed-length approximate string-matching problem. It comes in three flavors: a standard CPU version; an OpenMP version; and an MPI version. We compare the performance of our MIC implementation to the corresponding CPU version of MoTeX. Our MIC implementation accelerates the computations by a factor of ten compared to the CPU version. We also compare the performance of our MIC implementation to the corresponding OpenMP version of MoTeX running on modern Multicore architectures. Our MIC implementation accelerates the computations by a factor of two compared to the OpenMP version.