Smith-Waterman Protein Search with OpenCL on an FPGA

Abstract

The well-known Smith-Waterman (SW) algorithm is a high-sensitivity method for local alignments. Unfortunately, SW is expensive in terms of both execution time and memory usage, which makes it impractical in many scenarios. Previous research has shown that massively parallel architectures such as GPUs and FPGAs are able to mitigate the computational problems and achieve impressive speedups. In this paper we explore SW acceleration on an FPGA with OpenCL. We efficiently exploit data and thread-level parallelism on an Altera Stratix V FPGA, obtaining up to 39 GCUPS with less than 25 watt of power consumption.