arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.00876cs.DCcs.CEcs.DS
收藏

基于GPU的快速泛基因组图布局

Rapid GPU-Based Pangenome Graph Layout

Cornell University(康奈尔大学);Technical University of Munich(慕尼黑工业大学);University of Tübingen(蒂宾根大学);UTHSC(田纳西健康科学大学)

Jiajie Li, Jan-Niklas Schmelzle, Yixiao Du, Simon Heumos, Andrea Guarracino, Giulia Guidi, Pjotr Prins, Erik Garrison, Zhiru Zhang

AI总结:

针对泛基因组图布局算法的高计算需求,通过缓存友好数据布局、合并随机状态和warp合并三种优化,在GPU上实现57.3倍加速且不损失布局质量。

摘要:

计算泛基因组学是一个新兴领域,它利用包含多个基因组的图结构研究遗传变异。可视化泛基因组图对于理解基因组多样性至关重要。然而,由于图布局过程的高计算需求,处理大型图可能具有挑战性。在这项工作中,我们对最先进的泛基因组图布局算法进行了彻底的性能表征,揭示了显著的数据级并行性,这使得GPU成为计算加速的有前景选择。然而,不规则的数据访问和算法的内存密集型特性带来了重大障碍。为了克服这些挑战,我们开发了一种解决方案,实现了三种关键优化:缓存友好的数据布局、合并随机状态和warp合并。此外,我们提出了一种用于可扩展评估泛基因组布局质量的定量指标。在24个人类全染色体泛基因组上的评估表明,我们的基于GPU的解决方案在不损失布局质量的情况下,相比最先进的多线程CPU基线实现了57.3倍的加速,将执行时间从数小时缩短到数分钟。

英文摘要:

Computational Pangenomics is an emerging field that studies genetic variation using a graph structure encompassing multiple genomes. Visualizing pangenome graphs is vital for understanding genome diversity. Yet, handling large graphs can be challenging due to the high computational demands of the graph layout process. In this work, we conduct a thorough performance characterization of a state-of-the-art pangenome graph layout algorithm, revealing significant data-level parallelism, which makes GPUs a promising option for compute acceleration. However, irregular data access and the algorithm's memory-bound nature present significant hurdles. To overcome these challenges, we develop a solution implementing three key optimizations: a cache-friendly data layout, coalesced random states, and warp merging. Additionally, we propose a quantitative metric for scalable evaluation of pangenome layout quality. Evaluated on 24 human whole-chromosome pangenomes, our GPU-based solution achieves a 57.3x speedup over the state-of-the-art multithreaded CPU baseline without layout quality loss, reducing execution time from hours to minutes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑