基于定制协处理器的基因重测序加速技术研究

汤文; 张春明; 谭光明; 张佩珩; 孙凝晖

doi:10.7544/issn1000-1239.2014.20130987

基于定制协处理器的基因重测序加速技术研究

¹(中国科学院计算技术研究所高性能计算机研究中心北京 100190)
²(中国科学院大学北京 100049) (tangwen@ncic.ac.cn)

基金项目: 国家“九七三”重点基础研究发展计划基金项目(2012CB316502)

详细信息

中图分类号: TP302
计量
- 文章访问数: 01059
- HTML全文浏览量: 0
- PDF下载量: 0575
出版历程
- 发布日期: 2014-08-31

A Customized Coprocessor Acceleration of Genome Re-Sequencing

¹(High Performance Computer Research Center, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190)
²(University of Chinese Academy of Sciences, Beijing 100049)

摘要

摘要: 自2008年1月高通量测序技术应用以来，测序的通量和成本都在不断下降.然而基因数据的爆发式增长速度已经超过了摩尔定律，对海量数据的计算处理能力成为制约基因测序应用推广的瓶颈.以基于Hash索引的重测序算法为目标，对计算和访存行为进行分析，从而提出了一个现场可编程门阵列(field programmable gate array, FPGA)作为协处理器的架构，并在Convey公司的HC-1ex平台上进行了设计与实现.其基本处理单元内部采用全流水的设计及FIFO隔离计算模块和访存模块，可以完整执行重测序算法的核心流程.通过将基本处理单元和访存端口的一对一绑定，在4块Xilinx Virtex-6 LX760上实现了64路并行处理流程,总平均读内存带宽可达22.59GBps.与8核Intel Xeon处理器相比，可以提升28.5倍的性能.
- 高通量测序技术 /
- 短序列比对 /
- Hash索引 /
- 现场可编程门阵列 /
- 异构体系结构
Abstract: Since January 2008 when the next-generation DNA sequencing platforms were developed, the sequencing throughput has been significantly improved. However, this technology has been challenged by the large amount of sequencing data which grows dramatically even over the Moore's Law. As an emerging data-intensive workload, the high-throughput re-sequencing tools like Hash-based programs shows different characteristics from traditional computational applications. Both low arithmetic intensity and irregular memory access pattern are major sources of inefficiency on commodity multi-core platforms. In this paper, we propose co-processor architecture for accelerating a short reads mapping algorithm. The complete mapping flow in one processing element (PE) is integrated to an exclusive memory port to improve the parallel performance. This proposed architecture is then implemented on a Convey HC-1ex reconfigurable computer. The design includes 64 parallel PEs on 4 Xilinx Virtex-6 LX760 that operate at 150MHz. Compared with an Intel Xeon 8-cores CPU, the speedup achieves 28.5 times, and the average memory read bandwidth achieves 22.59GBps. Therefore, this proposed design can potentially supply a solution to the large-amount data challenge and be applied in high throughput re-sequencing.
- high-throughput sequencing /
- short reads mapping /
- Hash-index /
- field programmable gate array (FPGA) /
- heterogeneous architecture

HTML全文

参考文献(0)

施引文献(13)

期刊类型引用(6)

1.	宋传鸣，王一琦，武惠娟，何熠辉，洪飏，王相海. 深度卷积网络的自然场景文本检测研究综述. 小型微型计算机系统. 2023(09): 1996-2008 . 百度学术
2.	朱建伟，李朝奎，黄云涛，王佳欣，钟森. 车载遥感高速公路广告影像的文本信息提取研究与应用. 遥感信息. 2022(02): 126-130 . 百度学术
3.	赵芳，贺怡. 基于人工电场优化的软件定义物联网路由算法. 计算机工程与设计. 2021(10): 2725-2732 . 百度学术
4.	李凯勇. 大区域图像局部破损点优化提取仿真. 计算机仿真. 2020(05): 439-442+457 . 百度学术
5.	李朝献. 基于自适应三维立体图像增强优化处理研究. 计算机仿真. 2020(12): 358-361 . 百度学术
6.	索岩，崔智勇. 场馆监控图像的DCT域视觉显著性检测仿真. 计算机仿真. 2020(12): 421-425 . 百度学术