Ultra Efficient Acceleration for De Novo Genome Assembly via Near-Memory Computing

De novo assembly of genomes for which there is no reference, is essential for novel species discovery and metagenomics. In this work, we accelerate two key performance bottlenecks of DBG-based assembly, graph construction and graph traversal, with a near-data processing (NDP) architecture based on 3...

Full description

Saved in:
Bibliographic Details
Published in2021 30th International Conference on Parallel Architectures and Compilation Techniques (PACT) pp. 199 - 212
Main Authors Zhou, Minxuan, Wu, Lingxi, Li, Muzhou, Moshiri, Niema, Skadron, Kevin, Rosing, Tajana
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2021
Subjects
Accelerator Design
Bandwidth
Bioinformatics
Genomics
Message passing
Parallel processing
Precision medicine
Processing in memory
Vaccines
Online AccessGet full text
DOI10.1109/PACT52795.2021.00022

Cover

More Information
Summary:De novo assembly of genomes for which there is no reference, is essential for novel species discovery and metagenomics. In this work, we accelerate two key performance bottlenecks of DBG-based assembly, graph construction and graph traversal, with a near-data processing (NDP) architecture based on 3D-stacking. The proposed framework distributes key operations across NDP cores to exploit a high degree of parallelism and high memory bandwidth. We propose several optimizations based on domain-specific properties to improve the performance of our design. We integrate the proposed techniques into an existing DBG assembly tool, and our simulation-based evaluation shows that the proposed NDP implementation can improve the performance of graph construction by 33× and traversal by 16× compared to the state-of-the-art.
DOI:10.1109/PACT52795.2021.00022