软件定义数据中心网络混合路由机制

doi:10.11959/j.issn.1000-436x.2016071

摘要/Abstract

摘要：

针对数据中心网络流量大小分布不均匀、传输性能需求不相同的特征，提出了面向传统树型数据中心网络结构的软件定义混合路由机制SHR（software-defined hybrid routing）。SHR通过统计计算将数据流分为大流和小流，为满足其不同的传输性能需求，对大流采用自适应路由算法，对小流采用流量无视路由算法。SHR 在OpenFlow机制的基础上，将部分控制权从控制器下放至交换机，以减轻网络的额外负载。在Fat-Tree网络拓扑结构上建立流量模型进行性能分析与仿真实验，结果表明，与传统的等价多路径转发 ECMP 算法相比，SHR 能够提高网络吞吐量，降低数据流丢弃率和分组端到端时延，同时减轻网络的额外负载。

关键词: 云计算, 数据中心网络, 软件定义网络, 路由算法, 开放流协议

Abstract:

In the current data center networks, the flow size distribution is not uniform and the transmission performance requirements of elephant flows and mice flows are different. To address this issue, a software-defined hybrid routing (SHR) scheme was proposed. SHR differentiate data flows by statistical calculat nt flows utilize the adap-tive routing algorithm while the mice flows use the oblivious routing algorithm. SHR extends the OpenFlow scheme by offloading some basic functions such as flow statistical detection and mice flow forwarding to switches to reduce the switch-controller interaction overhead. Performance evaluations of SHR were carried out using the fat-tree network to-pology. Results show that SHR can effectively increase network throughput and reduce the flow dropping rate as well as packet delay compared with the traditional ECMP algorithm.

Key words: cloud computing, data center network, software defined network, routing algorithm, OpenFlow

蔡岳平,王昌平. 软件定义数据中心网络混合路由机制[J]. 通信学报, 2016, 37(4): 44-52.

Yue-ping CAI,Chang-ping WANG. Software defined data center network with hybrid routing[J]. Journal on Communications, 2016, 37(4): 44-52.

图/表 6

图1

图2

图3

图4

图5

图6

参考文献 26

[1]	CHEN Y Y , JAIN S , ADHIKARI V K , et al. A first look at inter-data center traffic characteristics via Yahoo! datasets[C]// IEEE INFOCOM'11. Shanghai, China, c2011:1620-1628.
[2]	AL-FARES M , LOUKISSAS A , VAHDAT A . A scalable, commodity data center network architecture[C]// SIGCOMM '08. Seattle, WA, USA, c2008:63-74.
[3]	ANDREW R , KIM W H , PRAVEEN Y . Mahout low-overhead DC traffic management using end-host-based elephant detection[C]// Proc of IEEE INFOCOM'11. Shanghai, China, c2011:1629-1637.
[4]	GUO C , LU G , LI D , et al. Bcube: a high performance, server-centric network architecture for modular data centers[C]// SIGCOMM'09. Barcelona, Spain, c2009:63-74.
[5]	GREENBERG A , HAMILTON J R , JAIN N , et al. VL2: a scalable and flexible data center network[C]// SIGCOMM'09. Barcelona, Spain, c2009:51-62.
[6]	GUO C , WU H , TAN K , et al. Dcell: a scalable and fault-tolerant network structure for data centers[C]// SIGCOMM'08. Seattle, WA, USA, c2008:75-86.
[7]	LI D , GUO C , WU H , et al. FiConn: using backup port for server interconnection in data centers[C]// SIGCOMM'09. Barcelona, Spain, c2009:2276-2285.
[8]	MYSORE R N , PAMBORIS A , FARRINGTON N , et al. PortLand: a scalable fault-tolerant layer 2 data center network fabric[C]// SIGCOMM'09. Barcelona, Spain, c2009.
[9]	ABU-LIIBDEH H , COSTA P , ROWSTRON A , et al. Symbiotic routing in future data centers[C]// SIGCOMM'10. New Delhi, India, c2010:51-62.
[10]	WU H , LU G , LI D , et al. MDCube: a high performance network structure for modular data center interconnection[C]// ACM CoNext'09. Rome, Italy, c2009:25-36.
[11]	HOPPS C . Analysis of an equal-cost multi-path algorithm[S]. RFC 2992, IETF, 2000.
[12]	BENSON T , ANAND A , AKELLA A , et al. Understanding data center traffic characteristics[C]// Proc of SIGCOMM'09. Barcelona, Spain, c2009:92-99.
[13]	Cisco Systems. Cisco data center infrastructure 2.5 design guide[EB/OL]. , 2013.
[14]	LI D , XU M , ZHAO H , et al. Building mega data center from heterogene-ous containers[C]// ICNP'11. Vancouver, BC Canada, c2011:256-265.
[15]	AL-FARES M , RADHAKRISHNAN S , RAGHAVAN B , et al. Hedera:dynamic flow scheduling for data center networks[C]// Usenix NSDI'10. California, USA, c2010:19.
[16]	WILSON C , BALLANI H , KARAGIANNIS T , et al. Better never than late: meeting deadlines in datacenter networks[C]// SIGCOMM'11. Toronto, Ontario, Canada, c2011:50-61.
[17]	HONG C Y , CAESAR M , GODFREY P B . Finishing flows quickly with preemptive scheduling[J]. ACM Computer Communication Review, 2012,42(4):127-138.
[18]	ZATS D , DAS T , MOHAN P , et al. DeTail: reducing the flow completion time tail in datacenter networks[J]. ACM Sigcomm Computer Communication Review, 2012,42(4):139-150.
[19]	SUN Y , CHEN M , LIU B , et al. FAR: a fault-avoidance routing method for data center networks with regular topology[C]// The 9th ACM/IEEE Symposium on Architectures for Networking and Communications Systems IEEE Press. San Jose, CA, USA, c2013:181-189.
[20]	CAO J , XIA R , YANG P , et al. Per-packet load-balanced, low-latency routing for clos-based data center networks[C]// The 9th ACM Conference on Emerging Networking experiments and Tech logies ACM. Santa Barbara, CA, USA, c2013:49-60.
[21]	ZAHAVI E , KESLASSY I , KOLODNY A . Distributed adaptive routing convergence to non-blocking DCN routing assignments[J]. Selected Areas in Communications IEEE Journal, 2014,32(1):88-101.
[22]	ANDREW R , JEFFREY C , et al. DevoFlow: scaling flow manage-ment for high-performance networks[C]// SIGCOMM'11. Toronto, Ontario, Canada, c2011:254-265.
[23]	BENSON T , ANAND A , AKELLA A , et al. MicroTE: fine grained traffic engineering for data centers[C]// The 7th Conference on Emerging Networking Experiments and Technologies ACM. Tokyo, Japan, c2011:1-12.
[24]	BENSON T , AKELLA A , MALTZ D A . Network traffic characteristics of data centers in the wild[C]// IMC'10. Melbourne, Australia, c2010:267-280.
[25]	KANDULA S , SENGUPTA S , GREENBERG A , et al. The nature of data center traffic: measurements and analysis[C]// IMC'09. Chicago, Illinois, USA, c2009:202-208.
[26]	樊平毅，冯重熙．现代通信理论基础（中册）[M]．北京：清华大学出版社， 2007: 100-132. FAN P Y , FENG C X . Fundamentals of modern communications[M]. Beijing: Tsinghua University Press, 2007: 100-132.