高维胖树系统中确定性路由容错策略实现

doi:10.11772/j.issn.1001-9081.2017103024

计算机应用 ›› 2018, Vol. 38 ›› Issue (5): 1393-1398.DOI: 10.11772/j.issn.1001-9081.2017103024

高维胖树系统中确定性路由容错策略实现

徐佳庆¹, 万文², 蔡东京¹, 唐付桥¹, 何杰¹, 张磊¹

1. 国防科技大学计算机学院, 长沙 410073;
2. 中山大学国家超级计算广州中心, 广州 510006

收稿日期:2017-12-25 修回日期:2017-12-25 出版日期:2018-05-10 发布日期:2018-05-24
通讯作者: 徐佳庆
作者简介:徐佳庆(1982-),男,江苏如东人,助理研究员,博士,CCF会员,主要研究方向:高速互连、高性能计算;万文(1987-),男,湖南永州人,工程师,硕士,主要研究方向:高性能计算、计算机体系结构;蔡东京(1986-),男,湖北浠水人,工程师,硕士,主要研究方向:高速互连、高性能计算;唐付桥(1989-),男,湖南永州人,工程师,硕士,主要研究方向:高速互连、高性能计算;何杰(1988-),男,湖南常德人,工程师,硕士,主要研究方向:高速互连、高性能计算;张磊(1984-),男,湖南浏阳人,工程师,主要研究方向:高速互连、高性能计算。
基金资助:
国家重点研发计划项目（2016YFB0200203）；国家自然科学基金面上项目（61572509）。

Implementation of deterministic routing fault-tolerant strategies for K-Ary N-Bridge system

XU Jiaqing¹, WAN Wen², CAI Dongjing¹, TANG Fuqiao¹, HE Jie¹, ZHANG Lei¹

1. School of Computer, National University of Defense Technology, Changsha Hunan 410073, China;
2. National Supercomputer Center in Guangzhou, Sun Yat-sen University, Guangzhou Guangdong 510006, China

Received:2017-12-25 Revised:2017-12-25 Online:2018-05-10 Published:2018-05-24
Contact: 徐佳庆
Supported by:
This work is partially supported by The National Key Research and Development Program of China (2016YFB0200203), the National Natural Science Foundation of China (61572509).

摘要/Abstract

摘要： 由于采用高维胖树拓扑结构的高性能计算机系统中叶交换机故障将严重影响系统使用，为了提高系统的可用性和可维性，基于误路由的思想提出了一套适用于高维胖树拓扑的确定性路由容错策略。其基本思路是通过误路由绕过发生故障的叶交换机，跳转至同维中其他叶交换机后，再通过正常路由到达目的节点。该容错策略可在不影响系统使用的情况下，实现故障叶交换机的屏蔽，并在实际的高维胖树系统中进行了容错实验。实验结果表明，该容错策略取得了可快速屏蔽故障叶交换机的预期效果，可以有效地提高系统维护的效率。

关键词: 高维胖树拓扑, 互连故障, 路由容错策略, 高性能计算, 网络维护

Abstract: The leaf switch failure would seriously affect the use of high performance computer system with K-Ary N-Bridge topology. In order to improve the usability and maintainability of that topology, a routing fault-tolerant strategy based on misrouting algorithm was proposed. The basic idea was to bypass the failed leaf switch leveraging misrouting, jump to other leaf switches in the same dimension, and then reached the destination node through the normal route. The proposed fault-tolerant strategy could shield the failed leaf switch without affecting the system usage. A fault-tolerant experiment was carried out in a practical K-Ary N-Bridge topology. The result shows that this fault-tolerant strategy can quickly shield the failed leaf switch as expected and can effectively improve the efficiency of system maintenance.

Key words: K-Ary N-Bridge, interconnection fault, routing fault-tolerance strategy, high performance computing, network maintenance

中图分类号:

TP393.1

徐佳庆, 万文, 蔡东京, 唐付桥, 何杰, 张磊. 高维胖树系统中确定性路由容错策略实现[J]. 计算机应用, 2018, 38(5): 1393-1398.

XU Jiaqing, WAN Wen, CAI Dongjing, TANG Fuqiao, HE Jie, ZHANG Lei. Implementation of deterministic routing fault-tolerant strategies for K-Ary N-Bridge system[J]. Journal of Computer Applications, 2018, 38(5): 1393-1398.

参考文献

[1] 徐佳庆,万文,蔡东京,等.胖树系统中分布式动态路由容错方法的实现[C]//第二十一届计算机工程与工艺年会.长沙:湖南科学技术出版社, 2017:74-82.(XU J Q, WAN W, CAI D Q, et al. Implementtation of distributed dynamic fault-tolerant routing in fat tree system[C]//Proceedings of the 21th National Conference on Computer Engineering and Technology Forum. Changsha:Hunan Science & Technology Press, 2017:74-82.)
[2] 方明. 高阶互连网络中路由器交换结构及互连拓扑结构研究[D].长沙:中南大学, 2013.(FANG M. Research on router switching fabric and network topology for high radix interconnection network[D]. Changsha:Central South University, 2013.)
[3] 王焕然, 徐明伟. SNMP网络管理综述[J]. 小型微型计算机系统, 2004, 25(3):358-366.(WANG H R, XU M W. Survey on SNMP network management[J]. Journal of Chinese Computer Systems, 2004, 25(3):358-366.)
[4] 温建伟. InfiniBand子网管理技术的研究与实现[D].长沙:国防科学技术大学, 2009.(WEN J W. Research and realization of InfiniBand subnet management technology[D]. Changsha:National University of Defense Technology, 2009.)
[5] KAMIURA N, KODERA T, MATSUI N. Design of a fault-tolerant multistage interconnection network with parallel duplicated switches[C]//Proceedings of the 15th IEEE International Symposium on Defect and Fault-Tolerance in VLSI Systems. Piscataway, NJ:IEEE, 2000:143-151.
[6] KONSTANTINIDOU S. The selective extra stage butterfly[J]. IEEE Transactions on Very Large Scale Integration Systems, 1993, 1(2):167-171.
[7] CHALSANI S, RAGHAVENDRA C S, VARMA A. Fault-tolerant routing in MIN-based supercomputers[C]//Proceedings of the 1990 ACM/IEEE conference on Supercomputing. New York:ACM, 1990:244-253.
[8] SHARMA N K. Fault-tolerance of a MIN using hybrid redundancy[C]//Proceedings of the 27th Annual Simulation Symposium. Piscataway, NJ:IEEE, 1994:142-149.
[9] LIN X, CHUANG Y, HUANG T. A multiple LID routing scheme for fat-tree-based InfiniBand networks[C]//Proceedings of the 18th International Parallel and Distributed Processing Symposium. Piscataway, NJ:IEEE, 2004:1-10.
[10] SEM-JACOBSEN F O, SKEIE T, LYSNE O, et al. Dynamic fault tolerance with misrouting in fat trees[C]//Proceedings of the 2006 International Conference on Parallel Processing. Washington, DC:IEEE Computer Society, 2006:33-44.
[11] MYSORE R N,GOMEZ M E, LOPEZ P, et al. FT²EI:a dynamic fault-tolerant routing methodology for fat tree with exclusion interval[J]. IEEE Transactions on Parallel and Distributed System, 2009, 20(6):802-817.
[12] GREENBERG A, HAMILTON J, JAIN N. VL2:a scalable and flexible data center network[C]//Proceedings of the ACM SIGCOMM 2009 Conference on Data Communication. New York:ACM, 2009:51-62.
[13] 曹继军, 刘路, 王永庆.源路由胖树网络的端节点动态容错路由方法[J]. 计算机工程与科学, 2013,35(3):8-14.(CAO J J,LU L,WANG Y Q. End-point dynamic fault-tolerant approach in source-routing fat trees[J]. Computer Engineering & Science, 2013,35(3):8-14.)
[14] 胡农达, 王达伟, 孙凝晖. 胖树中的分布式动态容错路由[J]. 计算机学报, 2010, 33(10):1799-1808.(HU N D,WANG D W,SUN N H. Distributed dynamic fault-tolerant routing in fat tree[J]. Chinese Journal of Computers, 2010, 33(10):1799-1808.)

高维胖树系统中确定性路由容错策略实现

Implementation of deterministic routing fault-tolerant strategies for K-Ary N-Bridge system

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 12

编辑推荐

Metrics

[1]	龚鸣清, 叶煌, 张鉴, 卢兴敬, 陈伟. 基于ARMv8架构的面向机器翻译的单精度浮点通用矩阵乘法优化[J]. 计算机应用, 2019, 39(6): 1557-1562.
[2]	赵士操, 肖永浩, 段博文, 李于锋. HSWAP:适用于高性能计算环境的数值模拟工作流管理平台[J]. 计算机应用, 2019, 39(6): 1569-1576.
[3]	王鹏, 周岩. 面向高性能应用的MPI大数据处理[J]. 计算机应用, 2018, 38(12): 3496-3499.
[4]	高剑, 于康, 卿鹏, 尉红梅. 面向高性能计算的分布式故障定位框架[J]. 计算机应用, 2018, 38(1): 44-49.
[5]	吕宏武, 谷雷, 王慧强, 邹世辰, 冯光升. 分层检查点的近似最优周期计算模型[J]. 计算机应用, 2017, 37(1): 103-107.
[6]	熊壬浩, 刘羽. A^*算法的改进及并行化[J]. 计算机应用, 2015, 35(7): 1843-1848.
[7]	李春艳张学杰. 基于高性能计算的开源云平台性能评估[J]. 计算机应用, 2013, 33(12): 3580-3585.
[8]	姚信安胡世平宋飞. 高性能计算机总线变换器输出阻抗优化设计[J]. 计算机应用, 2012, 32(06): 1774-1777.
[9]	姚信安宋飞胡世平. 高性能计算机系统电源设计[J]. 计算机应用, 2012, 32(04): 1176-1179.
[10]	曹荣强曹宗雁迟学斌肖海力. 基于RMI的高性能计算网格二次开发模型[J]. 计算机应用, 2010, 30(9): 2526-2529.
[11]	潘卫陈燎原李永革张锦华潘莉夏凡. 基于集群技术构建聚变研究高性能计算系统[J]. 计算机应用, 2009, 29(08): 2132-2135.
[12]	张骏陈良育曾振柄. 一种混合高性能计算机代数环境模型[J]. 计算机应用, 2007, 27(11): 2834-2837.