GeneTEK: Low-power, high-performance and scalable genome sequence matching in FPGAs

Abstract: The advent of next-generation sequencing (NGS) has revolutionized genomic research by enabling high-throughput data generation through parallel sequencing of a diverse range of organisms at significantly reduced costs. This breakthrough has unleashed a "Cambrian explosion" in genomic data volume and diversity. This volume of workloads places genomics among the top four big data challenges anticipated for this decade. In this context, pairwise sequence alignment represents a very time- and energy-consuming step in common bioinformatics pipelines. Speeding up this step requires the implementation of heuristic approaches, optimized algorithms, and/or hardware acceleration. Whereas state-of-the-art CPU and GPU implementations have demonstrated significant performance gains, recent field programmable gate array (FPGA) implementations have shown improved energy efficiency. However, the latter often suffer from limited scalability due to constraints on hardware resources when aligning longer sequences. In this work, we present a scalable and flexible FPGA-based accelerator template that implements Myers's algorithm using high-level synthesis and a worker-based architecture. GeneTEK, an instance of this accelerator template in a Xilinx Zynq UltraScale+ FPGA, outperforms state-of-the-art CPU and GPU implementations in both speed and energy efficiency, while overcoming scalability limitations of current FPGA approaches. Specifically, GeneTEK achieves at least a 19.4% increase in execution speed and up to 62x reduction in energy consumption compared to leading CPU and GPU solutions, while fitting comparison matrices up to 72% larger compared to previous FPGA solutions. These results reaffirm the potential of FPGAs as an energy-efficient platform for scalable genomic workloads.