What question did this study set out to answer?

Implement and evaluate the complete BWA-MEM read mapping pipeline on GPU architectures to resolve computational bottlenecks in high-throughput genomic data processing.

June 20, 2023Open Access

Accelerating BWA-MEM Read Mapping on GPUs

View Full Paper

Ask AI

Bookmark

Implement and evaluate the complete BWA-MEM read mapping pipeline on GPU architectures to resolve computational bottlenecks in high-throughput genomic data processing.
Ported and restructured core algorithmic stages of BWA-MEM—seeding, seed chaining, Smith-Waterman alignment, memory management, and I/O handling—for parallel GPU execution.
Benchmarked runtime and stage-wise performance on NVIDIA A40, A6000, and A100 GPUs against multi-threaded CPU baselines (BWA-MEM and BWA-MEM2 on a 64-thread CPU).
Achieved total wall-time speedups over CPU-based BWA-MEM of up to 5.8x on an NVIDIA A40, 6.1x on an NVIDIA A6000, and 6.8x on an NVIDIA A100.
Surpassed CPU-optimized BWA-MEM2 performance, reaching wall-time speedups of up to 3.2x (A40), 3.4x (A6000), and 3.8x (A100).
Demonstrated individual stage speedups on an A100 GPU of up to 4.0x for seeding and chaining, 2.5x for Smith-Waterman alignment, and 7.9x for SAM output generation.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Computational study demonstrates up to 6.8x faster genomic read mapping using GPUs versus 64-thread CPUs, highlighting a major efficiency boost for next-generation sequencing data analysis.

Cite This Study

Pham et al. (2023) studied this question.

synapsesocial.com/papers/6a12ed5306ed52b5c2c0c86b https://doi.org/10.1145/3577193.3593703

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates up to 6.8x faster genomic read mapping using GPUs versus 64-thread CPUs, highlighting a major efficiency boost for next-generation sequencing data analysis.

Accelerating BWA-MEM Read Mapping on GPUs

Key Points

Discussion

Implication

Cite This Study

Accelerating BWA-MEM Read Mapping on GPUs

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider