A 135mW Fully Integrated Data Processor for Next-Generation Sequencing

📋 论文概要

该论文提出了一种全集成数据处理器，用于下一代测序（NGS）中的DNA映射加速。通过专用硬件设计实现Suffix Array排序和向后搜索，在135mW低功耗下完成高通量DNA数据分析，解决了传统软件处理时间过长的问题。

💡 主要创新点

• 首次提出全集成专用数据处理器，将NGS数据映射的两个关键步骤（SA排序和向后搜索）通过硬件加速实现，显著提升处理速度。
• 在仅135mW功耗下完成高性能DNA映射，实现了能效与吞吐量的良好平衡。

🏷 关键词

📄 原文摘要

National Chiao Tung University, Hsinchu, Taiwan 1 2 DNA sequencing is the process of determining the precise order of nucleotides (A, C, G, T) within a DNA molecule and is now indispensable for genetics and medical research. Next-generation sequencing (NGS), in which short DNA fragments can be sequenced in a massively parallel fashion, enables highthroughput sequencing [1]. However, the succeeding data analysis, also known as DNA mapping, is excessively time consuming. DNA mapping can be partitioned into Suffix Array (SA) Sorting and Backward Searching. Dedicated hardware designs have been proposed to enhance the speed, but only for the less complex Backward Searching [2]. Feasible hardware for the most complicated part, SA sorting, has never been explored. This work presents a fully integrated NGS data processor that realizes both SA sorting and Backward Searching. The distributed sort algorithm is utilized to reduce the sorting complexity. The throughput of SA

👥 作者与机构

Yi-Chung Wu1, Jui-Hung Hung2, Chia-Hsiang Yang1,2

National Taiwan University, Taipei, Taiwan