An Efficient GPUAccelerated Implementation of Genomic Short Read Mapping with BWAMEM

Abstract

Next Generation Sequencing techniques have resulted in an exponential growth in the generation of genetics data, the amount of which will soon rival, if not overtake, other Big Data fields, such as astronomy and streaming video services. To become useful, this data requires processing by a complex pipeline of algorithms, taking multiple days even on large clusters. The mapping stage of such genomics pipelines, which maps the short reads onto a reference genome, takes up a significant portion of execution time. BWA-MEM is the de-facto industry-standard for the mapping stage.Here, a GPU-accelerated implementation of BWA-MEM is proposed. The Seed Extension phase, one of the three main BWA-MEM algorithm phases that requires between 30%-50% of overall processing time, is offloaded onto the GPU. A thorough design space analysis is presented for an optimized mapping of this phase onto the GPU. The resulting systolic-array based implementation obtains a twofold overall application-level speedup, which is the maximum theoretically achievable speedup. Moreover, this speedup is sustained for systems with up to twenty-two logical cores. Based on the findings, a number of suggestions are made to improve GPU architecture, resulting in potentially greatly increased performance for bioinformatics-class algorithms.

GenStore: a high-performance in-storage processing system for genome sequence analysis

“…This is because ACC greatly reduces the computational bottleneck, which increases the relative effect of the storage subsystem on the end-to-end execution time. The ACC and Ideal-ISF+ACC results clearly show that data movement between the storage devices and the hardware accelerator, which has not been properly considered in prior read mapping accelerators [39,40,43,61,62,65,[70][71][72][73][74][75][76][77], can significantly bottleneck the potential benefits of the accelerator. Comparison to Other Near-Data Processing Systems.…”

SaLoBa: Maximizing Data Locality and Workload Balance for Fast Sequence Alignment on GPUs

“…As high-quality algorithms such as BWA-MEM [37] became a de facto standard, the usability of the GPU-aware alignment softwares were limited. Some approaches [13], [29]- [31] tackle this problem and design a seed extension kernel general enough to be used for BWA-MEM using intra-query performance. However, later approaches based on inter-query parallelism outperformed these kernels, which is the strategy adopted by the current state-of-the-art methods such as NVBIO [3] or GASAL2 [9].…”

Power-efficiency analysis of accelerated BWA-MEM implementations on heterogeneous computing platforms

“…To our knowledge the only application-level accelerated integrated implementations of BWA-MEM that exist are: an FPGA-accelerated implementation of the Seed Extension phase [15] achieving a 1.5x speedup, further improved in [16] for an overall 2.6x speedup; and a GPU implementation [9], further improved to achieve an up to 2x speedup [17]. The FPGA implementation used here builds on [15], and a comparison of the implementation here is made to the improved GPU implementation.…”

“…Then, the details of the FPGAaccelerated implementation on the Alpha Data card are given. Finally, details of the GPU implementation are briefly discussed (further details can be found in [17]). …”

“…Similar to the FPGA implementation, the GPU implementation, further described in [17], also accelerates the Seed Extension phase. However, in contrast to the FPGA implementation, which contains six Seed Extension modules, the available execution resources on the GPU depend on the actual GPU model being used.…”