高速交换结构多播技术研究
|
摘要
随着互联网高速化和宽带化步伐的加速以及各种新型应用的兴起，要求作为网络节点的核心设备－路由交换设备能够适应信息容量急剧膨胀和网络业务多样化的需求。多播传输技术能够有效解决点到多点发送相同数据的问题，从而实现网络中点到多点的高速数据传送，有效利用网络资源，节约网络带宽并降低网络负载。目前核心路由器多采用交叉开关交换结构，并且已经产生了大量性能优良、简单易实现的基于交叉开关的调度算法来支持单播业务，但是在多播业务支持方面，现有的算法则存在加速比过大、调度机制过于复杂和实现成本过高等问题，可扩展性较差，在高速环境下难以提供良好的多播业务支持。本文针对这些问题，对基于交叉开关的高速交换结构及多播调度算法进行了重点研究，主要工作和贡献如下：
1．研究了基于交叉开关结构的多播队头阻塞缓解机制，在此基础上提出三种具有良好可扩展性的输入排队多播调度算法（MDRR、MMWR和MMFR）。现有的基于交叉开关的多播调度算法多采用输入排队FIFO结构，其交换性能由于存在多播队头阻塞问题而受到限制。为了减小多播队头阻塞的影响，其它一些算法在各个输入端口为多播业务维护多个队列并采用迭代式的极大匹配算法对多播业务进行调度，在高速的交换环境下，较大的算法仲裁时间和控制信息量使得这类算法实现困难。本文充分利用多播信元一对多输出的特点和队头信元扇出的差异性，提出三种非迭代式的两相（请求-许可）多播匹配算法。算法采用新的多播队头阻塞缓解机制，以很小的吞吐率性能损失换取了算法仲裁时间和控制信息量的大幅减小，更适合高速、大容量的交换环境。
2．在交叉开关多播调度算法的基础上，研究了基于交叉开关结构、支持单多播混合业务交换的集成调度算法。传统的单多播集成调度算法只是将现有的单播调度算法与多播调度算法进行了简单集成，并采用多次迭代的方式来提高交换性能。随着链路速率的不断提高，对调度算法执行时间的要求越来越严苛，使得多次迭代很难实现。本文提出一种非迭代式的两相（请求-许可）单多播集成调度算法。算法在每个时隙对单多播业务串行进行匹配，避免了现有算法单次迭代所产生的许可阻塞的影响，提高了算法的吞吐率性能，同时大大减小了算法的复杂度。
3．在负载均衡交换机的理论基础上，提出一种基于多播负载均衡、支持单多播混合业务交换的两级交换结构。算法对单多播业务分两级并行调度，流水操作。交换结构的第一级完成单播业务交换，同时对多播业务进行负载均衡；交换结构的第二级完成对经过负载均衡的多播业务的交换，同时采用集成的调度算法解决单多播业务对于交换结构输出端口的争用问题。基于多播负载均衡的两级交换结构对于多种业务模型，特别是非均匀的单多播混合业务模型具有较低的交换时延和较高的吞吐率，同时相比现有的单多播集成算法，其仲裁时间和控制信息量大幅减小，是一种适合高速交换环境应用的单多播集成交换结构。
4．三级Clos交换网络由于其具有良好的可扩展性而受到研究者的重视。目前对于Clos网络的多播调度算法以及单多播集成调度算法的研究非常少，现有Clos网络对多播业务的支持是通过在输入端将多播信元扇出分割成多个单播信元，然后采用单播调度算法完成多播信元的交换。这种实现方式一方面保留了现有Clos网络单播调度算法在高速环境下应用所具有的可扩展性差的缺点，另一方面由于属于同一多播信元的多个复制信元在交换结构中经历不同的排队和调度时延，影响了多播信元输出的同步性，而且由于算法对单多播业务没有进行隔离，突发的业务流量会影响正常业务流的交换性能，降低缓存的利用率。针对以上问题，同时考虑多播信元的扇出分布特点，我们提出了一种全分布式、简单易实现的多播正交分路调度算法以及单多播正交分路集成调度算法。算法以各个交叉开关交换单元为基本调度单位，采用按时隙轮转的正交路由分配方式为信元划分无冲突路径，本身不需要迭代，级间也不需要交互信息，且具有更低的仲裁时间和控制信息量，提高了算法在高速交换环境下的可扩展性。
With the accelerated pace of high-speed and broadband Internet, and the rise ofvarious new applications, it is required that the core equipments in network nodes-therouting and switching equipments can adapt to the rapid expansion of information andthe diversification of network applications. Multicast technology can effectively solvethe problem that information transmitted from one source to multiple destinations, so asto realize the high-speed multicast service transmission, effective use of networkresources, save network bandwidth and reduce the network load. As Crossbar switchnetwork is the mainstream switching fabric in current core routers, a large amount ofscheduling algorithms supporting unicast services based on the Crossbar switchingfabric have been proposed with good performance and simple implementation. Withmulticast services supporting, there are some problems with existing multicastscheduling algorithms such as high speed-up, high complexity and hard implementation,which lead to some scalability issues, and therefore cannot provide effective multicastswitching in high-speed environments. As a result, the key issues of realizablehigh-speed switching fabrics and multicast scheduling algorithms are investigated inthis thesis. The major contributions of this thesis are as follows.
1. We research on the multicast HoL (Head of Line) blocking alleviation schemefor input-queued Crossbar switching fabrics, and present three scalable multicastscheduling algorithms (MDRR、 MMWR and MMFR). Most existing multicastscheduling algorithms allocate one FIFO (first-in-first-out) queue for multicast traffic ateach input, and the performance achievable by switches is limited because of thewell-known multicast HoL blocking. Other algorithms maintain a small number ofFIFO queues for multicast traffic at each input to reduce the HoL blocking problem,however, multiple iterations, as well as a large amount of control message exchange arerequired to achieve high switching performance, which makes it difficult to implementin a high-speed environment. As a result, we propose three scalable non-iterativetwo-phase (request-grant) multicast scheduling schemes. The proposed schedulingsemploy new multicast HoL blocking alleviation schemes by fully using the one-to-manyproperty of multicast cells，as well as the diversity of fan-out sets with HoL cells, whichreduces the complexity of schedulings efficiently with very little loss of switchingperformance, and offers a reasonable choice for high-speed input-queuedswitches/routers.
2. Based on the study of multicast scheduling algorithm, we present a scalableintegrated scheduling algorithm that supporting unicast and multicast trafficsimultaneously for Crossbar switching fabrics. Most integrated three-phase(request-grant-accept) scheduling algorithms presented are, in fact, a combination ofearlier unicast and multicast algorithms unified in one integrated scheduler, and multipleiterations are needed to improve the switching performance. As link speed growsdramatically, high-speed switches have less and less time to perform scheduling, thusiterative scheduling schemes are difficult to implement. We propose a non-iterativetwo-phase (request-grant) integrated scheduling in this thesis to address the scalabilityproblems, in which the unicast scheduling and multicast scheduling are performedsequentially to increase the matching size by avoiding the grant-blocking in each timeslot, and the complexity of the scheduling can also be reduced efficiently by employingthe two-phase (request-grant) scheduling scheme.
3. Motivated by the load-balanced Birkhoff-von Neumann switches, we propose amulticast load-balancing two-stage switching architecture to support the mixed trafficswitching. The unicast scheduling and multicast scheduling are performed in parallelwith pipeline method at different stages. The first stage of the switching fabric performsswitching for unicast traffic and load-balancing for multicast traffic, while the secondstage performs switching for multicast traffic, as well as uses an integrated schedulingto solve the output contentions for unicast and multicast traffic. The proposed two-stageswitching fabric is more tolerant to unbalanced mixed traffic model, and reduces thescheduling complexity dramatically, which is more suitable for high-speed applications.
4. The study on the3-stage Clos network has been increasingly valued by theresearchers as its good scalability. However, little research has been done on supportingmulticast traffic efficiently in the3-stage Clos network. The existing implementationscheme with3-stage Clos networks supporting multicast traffic is that multicast cells aredivided into several unicast cells at each input according to the fan-out sets, and thenadopts unicast scheduling to schedule the copied cells. This implementation schememaintains the limitations of existing unicast schedulings at high-speed, at the same time,since the copied cells experience a separate queuing and scheduling process, it alsocauses the cells belonging to the same multicast cell to get out of sync at the outputs. Inaddition, as same queues are associated with both unicast and multicast traffic, one ofthe bursty traffic could influence the switching performance of the other normal traffic,and also reduces the memory utilization rate at each input. In order to solve theseproblems, we propose a distributed multicast orthogonal scheduling algorithm, as well as a distributed unicast and multicast integrated orthogonal scheduling algorithm in thisthesis. By taking each Crossbar module as the elemental scheduling unit, the proposedalgorithms adopt the rotated orthogonal routing at each timeslot to solve the outputcontentions arisen at each switching module of the second stage, in which iterations andcontrol message exchanged between stages are not required, and have lower timecomplexity and control message complexity, which improve the scalability inhigh-speed switching environments.
1．研究了基于交叉开关结构的多播队头阻塞缓解机制，在此基础上提出三种具有良好可扩展性的输入排队多播调度算法（MDRR、MMWR和MMFR）。现有的基于交叉开关的多播调度算法多采用输入排队FIFO结构，其交换性能由于存在多播队头阻塞问题而受到限制。为了减小多播队头阻塞的影响，其它一些算法在各个输入端口为多播业务维护多个队列并采用迭代式的极大匹配算法对多播业务进行调度，在高速的交换环境下，较大的算法仲裁时间和控制信息量使得这类算法实现困难。本文充分利用多播信元一对多输出的特点和队头信元扇出的差异性，提出三种非迭代式的两相（请求-许可）多播匹配算法。算法采用新的多播队头阻塞缓解机制，以很小的吞吐率性能损失换取了算法仲裁时间和控制信息量的大幅减小，更适合高速、大容量的交换环境。
2．在交叉开关多播调度算法的基础上，研究了基于交叉开关结构、支持单多播混合业务交换的集成调度算法。传统的单多播集成调度算法只是将现有的单播调度算法与多播调度算法进行了简单集成，并采用多次迭代的方式来提高交换性能。随着链路速率的不断提高，对调度算法执行时间的要求越来越严苛，使得多次迭代很难实现。本文提出一种非迭代式的两相（请求-许可）单多播集成调度算法。算法在每个时隙对单多播业务串行进行匹配，避免了现有算法单次迭代所产生的许可阻塞的影响，提高了算法的吞吐率性能，同时大大减小了算法的复杂度。
3．在负载均衡交换机的理论基础上，提出一种基于多播负载均衡、支持单多播混合业务交换的两级交换结构。算法对单多播业务分两级并行调度，流水操作。交换结构的第一级完成单播业务交换，同时对多播业务进行负载均衡；交换结构的第二级完成对经过负载均衡的多播业务的交换，同时采用集成的调度算法解决单多播业务对于交换结构输出端口的争用问题。基于多播负载均衡的两级交换结构对于多种业务模型，特别是非均匀的单多播混合业务模型具有较低的交换时延和较高的吞吐率，同时相比现有的单多播集成算法，其仲裁时间和控制信息量大幅减小，是一种适合高速交换环境应用的单多播集成交换结构。
4．三级Clos交换网络由于其具有良好的可扩展性而受到研究者的重视。目前对于Clos网络的多播调度算法以及单多播集成调度算法的研究非常少，现有Clos网络对多播业务的支持是通过在输入端将多播信元扇出分割成多个单播信元，然后采用单播调度算法完成多播信元的交换。这种实现方式一方面保留了现有Clos网络单播调度算法在高速环境下应用所具有的可扩展性差的缺点，另一方面由于属于同一多播信元的多个复制信元在交换结构中经历不同的排队和调度时延，影响了多播信元输出的同步性，而且由于算法对单多播业务没有进行隔离，突发的业务流量会影响正常业务流的交换性能，降低缓存的利用率。针对以上问题，同时考虑多播信元的扇出分布特点，我们提出了一种全分布式、简单易实现的多播正交分路调度算法以及单多播正交分路集成调度算法。算法以各个交叉开关交换单元为基本调度单位，采用按时隙轮转的正交路由分配方式为信元划分无冲突路径，本身不需要迭代，级间也不需要交互信息，且具有更低的仲裁时间和控制信息量，提高了算法在高速交换环境下的可扩展性。
With the accelerated pace of high-speed and broadband Internet, and the rise ofvarious new applications, it is required that the core equipments in network nodes-therouting and switching equipments can adapt to the rapid expansion of information andthe diversification of network applications. Multicast technology can effectively solvethe problem that information transmitted from one source to multiple destinations, so asto realize the high-speed multicast service transmission, effective use of networkresources, save network bandwidth and reduce the network load. As Crossbar switchnetwork is the mainstream switching fabric in current core routers, a large amount ofscheduling algorithms supporting unicast services based on the Crossbar switchingfabric have been proposed with good performance and simple implementation. Withmulticast services supporting, there are some problems with existing multicastscheduling algorithms such as high speed-up, high complexity and hard implementation,which lead to some scalability issues, and therefore cannot provide effective multicastswitching in high-speed environments. As a result, the key issues of realizablehigh-speed switching fabrics and multicast scheduling algorithms are investigated inthis thesis. The major contributions of this thesis are as follows.
1. We research on the multicast HoL (Head of Line) blocking alleviation schemefor input-queued Crossbar switching fabrics, and present three scalable multicastscheduling algorithms (MDRR、 MMWR and MMFR). Most existing multicastscheduling algorithms allocate one FIFO (first-in-first-out) queue for multicast traffic ateach input, and the performance achievable by switches is limited because of thewell-known multicast HoL blocking. Other algorithms maintain a small number ofFIFO queues for multicast traffic at each input to reduce the HoL blocking problem,however, multiple iterations, as well as a large amount of control message exchange arerequired to achieve high switching performance, which makes it difficult to implementin a high-speed environment. As a result, we propose three scalable non-iterativetwo-phase (request-grant) multicast scheduling schemes. The proposed schedulingsemploy new multicast HoL blocking alleviation schemes by fully using the one-to-manyproperty of multicast cells，as well as the diversity of fan-out sets with HoL cells, whichreduces the complexity of schedulings efficiently with very little loss of switchingperformance, and offers a reasonable choice for high-speed input-queuedswitches/routers.
2. Based on the study of multicast scheduling algorithm, we present a scalableintegrated scheduling algorithm that supporting unicast and multicast trafficsimultaneously for Crossbar switching fabrics. Most integrated three-phase(request-grant-accept) scheduling algorithms presented are, in fact, a combination ofearlier unicast and multicast algorithms unified in one integrated scheduler, and multipleiterations are needed to improve the switching performance. As link speed growsdramatically, high-speed switches have less and less time to perform scheduling, thusiterative scheduling schemes are difficult to implement. We propose a non-iterativetwo-phase (request-grant) integrated scheduling in this thesis to address the scalabilityproblems, in which the unicast scheduling and multicast scheduling are performedsequentially to increase the matching size by avoiding the grant-blocking in each timeslot, and the complexity of the scheduling can also be reduced efficiently by employingthe two-phase (request-grant) scheduling scheme.
3. Motivated by the load-balanced Birkhoff-von Neumann switches, we propose amulticast load-balancing two-stage switching architecture to support the mixed trafficswitching. The unicast scheduling and multicast scheduling are performed in parallelwith pipeline method at different stages. The first stage of the switching fabric performsswitching for unicast traffic and load-balancing for multicast traffic, while the secondstage performs switching for multicast traffic, as well as uses an integrated schedulingto solve the output contentions for unicast and multicast traffic. The proposed two-stageswitching fabric is more tolerant to unbalanced mixed traffic model, and reduces thescheduling complexity dramatically, which is more suitable for high-speed applications.
4. The study on the3-stage Clos network has been increasingly valued by theresearchers as its good scalability. However, little research has been done on supportingmulticast traffic efficiently in the3-stage Clos network. The existing implementationscheme with3-stage Clos networks supporting multicast traffic is that multicast cells aredivided into several unicast cells at each input according to the fan-out sets, and thenadopts unicast scheduling to schedule the copied cells. This implementation schememaintains the limitations of existing unicast schedulings at high-speed, at the same time,since the copied cells experience a separate queuing and scheduling process, it alsocauses the cells belonging to the same multicast cell to get out of sync at the outputs. Inaddition, as same queues are associated with both unicast and multicast traffic, one ofthe bursty traffic could influence the switching performance of the other normal traffic,and also reduces the memory utilization rate at each input. In order to solve theseproblems, we propose a distributed multicast orthogonal scheduling algorithm, as well as a distributed unicast and multicast integrated orthogonal scheduling algorithm in thisthesis. By taking each Crossbar module as the elemental scheduling unit, the proposedalgorithms adopt the rotated orthogonal routing at each timeslot to solve the outputcontentions arisen at each switching module of the second stage, in which iterations andcontrol message exchanged between stages are not required, and have lower timecomplexity and control message complexity, which improve the scalability inhigh-speed switching environments.
引文
[1]格兰研究.2010中国三网融合报告. http://www.sinodtv.net.
[2]中国互联网络信息中心.第31次中国互联网络发展状况统计报告.http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201301.
[3]汪卫国.2005中国通信产业十大关键词-IPTV. http://www.cttl.cn/zhthg/invest/2005/dp/t20060621_368299.htm.
[4] A. M. Odlyzko. Comments on the Larry Roberts and Caspian Networks study ofInternet traffic growth. The Cook Report on the Internet, Dec.2001,12-15.
[5] S. Deering, R. Hinden. Internet Protocol, Version6(IPv6) Specification.RFC2460[S]. IETF,1998.
[6]中华人民共和国通信行业标准.路由器设备技术规范-低端路由器.YD/T1096-2001.
[7]中华人民共和国通信行业标准.路由器设备技术规范-高端路由器.YD/T1097-2001.
[8] S.Keshav and R. Sharma. Issues and Trends in Router Design. IEEECommunications Magazine, May1998,36(5):144-151.
[9] H.J. Chao.Next Generation Routers. Proceedings of the IEEE, Sep.2002,90(9):1518-1558.
[10] PMC-Sierra Inc. A New Architecture for Switch and Router Design.White paper,Dec.1999.
[11]徐恪,熊勇强,吴建平.宽带IP路由器的体系结构分析.软件学报,2000,11(2):179-186.
[12]周卫华.高性能路由器的交换技术研究（博士论文）.2004,北京邮电大学.
[13] N.F. Mir. Performance Evaluation of Efficient Multipath Crossbars.in Proc.8thInternational Conf.Communication Systems.2002,1:384-388.
[14] P. Wijetunga. High-Performance Crossbar Design for System-On-Chip.in Proc.3rdIEEE International Workshop System-On-Chip for Real-Time Applications,2003,138-143.
[15] Juniper Networks Inc. T-series Routing Platforms System and Packet ForwardingArchitecture.White Paper,2002.
[16]黄恺,徐志伟.可扩展并行计算技术、结构与编程.北京:机械工业出版社,2000年5月.
[17] L R Goke and G J Lipovski. Banyan networks for partitioning multiprocessorsystems. The1st Symp on Computer Architecture, Gainesville, FL, USA,1973.ACM,1973:21-28.
[18] S Turner. Design of a broadcast packet switching network. IEEE Transactions onCommunications,1988,36(6):734-743.
[19] F A Tobagi and T C Kwok. The tandem banyan switching fabric:a simplehigh-performance fast packet switch. Proceedings of INFOCOM’91, BalHarbour,Florida,1991. IEEE,1991:1245-1253.
[20] C. Clos. A Study of Non-Blocking Switching Network. Bell System TechnicalJournal,1953,32:406-424.
[21] Pankaj Gupta. Scheduling in Input-Queued Switches:A Survey. http://tiny_tera.stanford. edu/~Pankaj/papers.html.
[22] B. Prabhakar, N. McKeown, R. Ahuja. Multicast Scheduling for Input-QueuedSwitches. IEEE Journal on Selected Areas in Communications, Jun.1997,15(5):855-866.
[23] N. McKeown and B. Prabhakar. Scheduling multicast cells in an input-queuedswitch. in Proc.1996IEEE INFOCOM,1:271-278.
[24] N. McKeown. A fast switched backplane for a gigabit switched router. BusinessComm. Rev.,1997,27(12):1-17.
[25] H. Kuwahara, N. Endo,M. Ogino, et al. A shared buffer memory switch for anATM exchange. Proceedings of IEEE INFOCOM,1989,1:118-122.
[26] G.Kesidis and N.McKeown. Output-buffer ATM packet switching forintegrated-services communication networks. in Proc. IEEE ICC’97,Montreal,Canada， Aug.1997,3:1684-1688.
[27] B.Prabhakar, N.McKeown. On the speedup Required for Combined Input andOutput Queued Switching. Stanford University Technical Report, Nov.1997.
[28] I. Stoica, Zhang H. Exact Emulation of an Output Queuing Switch by a CombinedInput Output Queuing Switch. in proc.6th IEEE/IFIP IWQoS’s98,May1998.
[29] Lin D, Jiang Y, and M. Hamdi. Selective Request Round-Robin Scheduling forVOQ Packet Switch Architecture. Proceedings of IEEE International Conferenceon Communications (ICC),2011:1-5.
[30] Chang C S,Chen W J,Huang H Y. On service guarantees for input-bufferedcrossbar switches: a capacity decomposition approach by Birkhoff and vonNeumann. IEEE IWQoS’99,1999,7(4):79-86.
[31] Tamir Y,Chi H C. Symmetric crossbar arbiters for VLSI communication switches.IEEE Transaction of Parallel Distribut System,1993,4(3):13-27.
[32] T E Anderson, S S Owicki, J B Saxe,et al. High speed switch scheduling for localarea networks. ACM Transactions on Computer Systems,1993,11(3):319-352.
[33] P. Giaccone, D. Shah, B. Prabhakar. An implementable parallel scheduler forinput-queued switches. IEEE Micro, Jan.-Feb.2002,22(1):19-25.
[34] H.Obara, T.Yasushi. An efficient contention resolution algorithm for input queuingATM cross-connect switches. International Journal of Digital and Analog CabledSystems, Oct.-Dec.1989,2(4):261-267.
[35] M. J. Karol, K. Y. Eng, H. Obara. Improving the performance of input-queuedATM packet switches. Proceedings of IEEE INFOCOM’92,1992,1:110-115.
[36] M.J. Karol，M.G. Hluchyj, and S.P. Morgan. Input Versus Output Queueing on aSpace-Division Packet Switch. IEEE Transactions on Communications, Dec.1987,35(12):1347-1356.
[37] S.Q.Li and M.J.Lee.A Study of Traffic Imbalances in a Fast Packet Switch.Proceeding of IEEE INFOCOM’89，Apr.1989,2:538-547.
[38] Y. Tamir and G. Frazier. High Performance Multi-Queue Buffers for VLSICommunications Switches. Proceedings of Computer Architecture, Jun.1988,343-354.
[39] M.Ali andH.Nguyen. A neural network implementation of an input access schemein a high-speed packet switch. Proceeding of IEEE INFOCOM’89, Apr.1989,2:1192-1196.
[40] GSR12000Tech Product Description. Performing Internet Routing and Switchingat Gigabit speeds. Cisco Systems, http://www.cisco.com/warp/public/733/12000/index.shtml.
[41] J.Hui and E.Arthurs. A broadband packet switch for integrated transport. IEEEJ.Select. Areas Commun, Oct.1987,5(8):1264-1273.
[42] Y. Li. The dual round robin matching switch with exhaustive service. in Proc.Workshop on High Performance Switching and Routing,2002,58-63.
[43] C. Lund，S. Phillips，and N. Reingold. Fair prioritized scheduling in an inputbuffered switch. IFIP-IEEE Conference on Broadband Communications,Montreal, Canada, Apr.1996,358-369.
[44] N.McKeown,M.Izzard,A.Mekkittikul,et al. Tiny Tera: a packet switch core. IEEEMicro, Jan.-Feb.1997,17(1):26-33.
[45] Cook W J,Pulleyblank W R.Cunningham.Comb inatorial Optimization.WileyJohn&Sons Inc,1997,11(2):264-269.
[46] M. Arlitt, C. Williamson. Internet web servers workload characterization andperformance implications. IEEE/ACM Transactions on networking, Oct.1997,5(5):631-645.
[47] R. Caceres, P. Danzig, S. Jamin. Characteristics of wide-area TCP/IP conversatins.Proceedings of the ACM SIGCOMM'91. Zurich:ACM Press,1991:47-65.
[48] Leland W,Taqqu M,Willinger W,et al.On the self-similar nature of Ethernet traffic.Proceedings of ACM SIGCOMM'93. New York:ACM Press,1993:48-56.
[49] P Abry, D Veitch. Wavelet analysis of long rang dependent traffic. IEEETransaction on Information Theory,Jan.1998,44(1):2-15.
[50] A.Gupta and N.Georganas. Analysis of Packet Switch with Input and OutputBuffers and Speed Constraints. Proceeding of IEEE INFOCOM, Apr.1991,2:694-700.
[51] S.Chuang, A.Goel, N.McKeown,et al. Matching output queueing with a combinedinput output queued switch. IEEE INFOCOM,1999,3:1169-1178.
[52] S. Chuang, A. Goel, N. McKeown et al. Matching output queueing with acombined input/output-queued switch. IEEE Journal of Selected Areas inCommunications, Jun.1999,17(6):1030-1039.
[53] A.Smiljanic，R. Fan and G.Ramamurthy. RRGS round-robin greedy scheduling forelectronic/optical terabit switches.in Proc. GLOBECOM'99,2:1244-1250.
[54] D.Cavendish, CORPS: A pipelined fair Packet scheduler for high speedinputqueued switches. in Proc. IEEE Conference on High Performance Switchingand Routing,2000,55-64.
[55] A.Motoki, S.Kamiya, R.Ikematsu, et al. Group pipeline scheduler for inputbuffer switch. in Proc. Joint4th IEEE International Conference on ATM(ICATM2001) and High Speed Intelligent Internet Symposium,2001,158-162.
[56] E.Oki, R.Rojas-Cessa, H.J.Chao. PMM:a Pipelined maximal-sized matchingscheduling approach for input buffered switches. in Proc.of GLOBECOM,2001,1:35-39.
[57] H. Ahmadi, W. E. Denzel.A Survey of Modern High-Performance SwitchingTechniques. IEEE Journal on Selected Areas in Communications, Sept.1989,7(7):1091-1103.
[58] R. Y. Awdeh, H. T. Mouftah. Survey of ATM Switch Architectures. ComputerNetworks and ISDN Systems, Nov.1995,27(12):1576-1613..
[59] R. Rojas-Cessa, E. Oki, Jing Z, et al.. CIXB-1: combined input-one-cell-crosspoint buffered switch. IEEE Workshop on High Performance Switching andRouting,2001,324-329.
[60] M. Nabeshima. Performance evaluation of a combined input-and crosspoint-queued switch. IEICE Transaction on Communication, Mar.2000,3:737-741.
[61] K. Yoshigoe, K. J. Christensen. A parallel-polled virtual output queued switch witha buffered crossbar. IEEE workshop on high performance switching and routing,2001,271-275.
[62] K. Yoshigoe, K. J. Christensen. An evolution to crossbar switches with virtualoutput queuing and buffered cross points. IEEE Network, Sept.-Oct.2003,17(5):48-56.
[63] C. Ming, N.D. Georganas.A fast algorithm for multi-channel/port trafficassignment. International Conference on Communications,1994,1:96-100.
[64] M. Hluchyj, M.Karol.Queueing in high-performance packet switching. IEEEJ.Selected Area Communications, Dec.1988,6(9):1587-1597.
[65] N.McKeown, A.Mekkittikul, V.Anantharam,et al. Achieving100%throughput inan input-queued switch.IEEE Trans.Commu.,Aug.1999,47(8):1260-1267.
[66] A.Mekkittikul, N.McKeown. A Fair Algorithm for Achieving100%Throughput inan Input-Queued Switch. in Proceedings of IC3N96,Oct.1996,Maryland.
[67] Hakyong Kim, Changhwan Oh, Kiseon Kim. Throughput analysis of bifurcatedinput queued packet switches with restricted contention. Electronics Letters,Aug.1998,34(17):1561-1562.
[68] N. McKeown. Scheduling Algorithms for Input-Queued Cell Swithes:[PhDThesis].University of California at Berkeley,1995.
[69] J.E.Hopcroft, R.M. Karp.An n5/2Algorithm for Maximum Matching in BipartiteGraphs. SIAM Journal on Computing,1973,1(2):225-231.
[70] R.E.Tarjan. Data Structures and Network Algorithms. Philadelphia:Society forIndustrial and Applied Mathematics,1983:112-234.
[71] R.O.LaMaire, D.N.Serpanos. Two dimensional round-robin schedulers for packetswitches with multiple input queues. IEEE/ACM Trans. Networking, Oct.1994,2(5):471-482.
[72] N.McKeown. The iSLIP Scheduling Algorithm for Input-queued Switches.IEEE/ACM Transactions on Networking,Apr.1999,7(2):188-201.
[73] H.J.Chao. Saturn: A terabit packet switch using dual round robin. IEEECommunications Magazine, Dec.2000,38(12):78-84.
[74] M. Ajmone Marsan, A. Bianco, P. Giaccone, et al. On the throughput ofinput-queued cell-based switches with multicast traffic. in Proc.2001IEEEINFOCOM,3:1664–1672.
[75] G. Shanthi and A. Shanmugam. Mathematical analysis of the input-queued packetswitch under multicast traffic.IEE Proc.-Commun., Dec.2005,152(6):845-849.
[76] J. Y. Hui, T. Renner. Queueing Analysis for Multicast Packet Switching. IEEETransactions on Communications, Feb.-Apr.1994,42(2-4):723-731.
[77] Sun S T, He S M, Zheng Y F, et al. Multicast scheduling in buffered crossbarswitches with multiple input queues. in Proc. IEEE HPSR’05, Hong Kong, China.USA: IEEE Communication Society,2005:73-77.
[78] L. Mhamdi. On the integration of unicast and multicast cell scheduling in bufferedcrossbar switches. IEEE Trans. Parallel Distrib. Syst., Jun.2009,20(6):818–830.
[79] M. Song and W. Zhu. Throughput analysis for multicast switches with multipleinput queues. IEEE Commun. Lett., Jul.2004,8(7):479–481.
[80] Min Song, Weiying Zhu. Delay Analysis of Multicast Switches with MultipleInput Queues. High Performance Switching and Routing,2004,10-14.
[81] W. Zhu and M. Song. Performance analysis of large multicast packet switcheswith multiple input queues and gathered traffic. Computer Commun., May2010,33(7):803–815.
[82] J.Y. Hui. Switching and Traffic Theory for Integrated Broadband Networks.Kluwer Academic Publishers, Boston, MA,1990.
[83] Yu H, Ruepp S, Berger M S. Enhanced first-in-first-out-based round-robinmulticast scheduling algorithm for input-queued switches. IET Commun.,2011,5(8):1163-1171.
[84] A. Bianco, A.Scicchitano. Multicast support in multi-chip centralized schedulersin Input Queued switches. Computer Networks,2009,53(7):1040-1049.
[85] Hao Y, Ruepp S, Berger M S. Multi-Level Round-Robin Multicast Schedulingwith Look-Ahead Mechanism. Proceedings of IEEE International Conference onCommunications (ICC),2011:1-5.
[86] Pan Deng, Yang Yuanyuan. FIFO-Based Multicast Scheduling Algorithm forVirtual Output Queued Packet Switches. IEEE Transactions on Computers,Oct.2005,54(10):1283-1296.
[87] S. Gupta, A. Aziz. Multicast scheduling for switches with multiple input-queues.in Proc.2002Symposium on High Performance Interconnects,28–33.
[88] A. Bianco, P. Giaccone, E. Leonardi, et al. On the number of input queues toefficiently support multicast traffic in input queued switches. in Proc.2003Workshop on High Performance Switching and Routing,111–116.
[89] Hao Yu. A Novel Round-Robin Based Multicast Scheduling Algorithm for100Gigabit Ethernet Switches. IEEE INFOCOM, Mar.2010,1-2.
[90]戴精科,彭来献,张邦宁.一种支持单播与组播混合业务的高速Crossbar调度算法.电子与信息学报,2009,31(10):2299-2304.
[91] Zhu W., Song M. Integration of unicast and multicast scheduling in input-queuedpacket switches. Computer Networks, Apr.2006,50(5):667-687.
[92] M. Andrews, S. Khanna, K.Kumaran. Integrated scheduling of unicast andmulticast traffic in an input-queued switch. in Proc. IEEE INFOCOM, New York,USA, Mar.1999,3:1144–1151.
[93] K. Chin.A new integrated unicast/multicast scheduler for inputqueued Switches. inProc.8th Australasian Symp. on Parallel and Distributed Computing, Brisbane,Australia,2010,13–20.
[94] C.S.Chang,D.S.Lee, C.M.Lien. Load balanced Birkhoff-von Neumann switches,part I:one-stage buffering. Computer Comm.,2002,25:611-622.
[95] M.G.Nadkarni.Basic Ergoidc Theory.Berlin:Birkhaiiser,1998,21-45.
[96] K.Petersen.Ergodic Theory.Cambridge:University Press,1983,26-51.
[97] F.Baccelli,P.Bremand. Elements of queuing theory. New York: Springer-Verlag,1994,36-58.
[98] C.S.Chang. Performance guarantees in communication networks. London:Springer-Verlag,2000,11-24.
[99] M.Schwartz. Broadband integrated networks. New Jersey:Prentice Hall,1996,41-53.
[100] Juniper Networks Inc. T-series Routing Platforms: system and PacketForwarding Architecture, White Paper,2002.
[101] B.Alleyne, I.Scherson. On Evil-Twin Networks and the Value of LimitedRandomized Routing. IEEE Transactions Parallel and Distributed Systems,2000,l(9).910-925.
[102] E. Oki, Z. Jing, R. Rojas-Cessa, et al. Concurrent round-robin dispatchingscheme in a Clos-network switch. in: IEEE ICC_2001, Jun.2001,1:107-111.
[103] E. Oki, Z. Jing, R. Rojas-Cessa, et al. Concurrent Round-Robin-BasedDispatching Schemes for Clos-Network Switches. IEEE/ACM Trans.Networking,2002,10(6):830-844.
[104] P.Konghong, M.Hamdi. Static Round-robin Dispatching Schemes forClos-network Switches. IEEE HPSR.2002,329-333.
[105]杨帆,邱智亮,徐展琦.输入缓存Clos网络中分布式正交分路路由算法.计算机工程与应用,2007,43(10):162-166.
[106] W. Wang, L. Yu, G. Zhu, et al. Coding-driven Scheduling for Frame-basedMulticast Switches. in Proc. International Symposium on Network Coding, Jul.2011,4.
[107] M. Kim, J. K. Sundararajan, M. Medard. Network Coding for Speedup inSwitches. in Proc. IEEE International Symposium on Information Theory, Jun.2007,1086-1090.
[108] M. Kim, J. K. Sundararajan, M. Medard. Network Coding in a Multicast Switch.IEEE Transactions on Information Theory, Jan.2011,57(1):436-460.
[109] S. Yang, X. Wang, B. Li. Haste: Practical Online Network Coding in a MulticastSwitch. in Proc. IEEE INFOCOM,2010,5.
[110]李挥,林良敏,黄佳庆.应用网络编码的分组交换调度算法.中兴通讯技术，Feb.2009,15(1):20-27.
[111]胡曦明.有效支持IPTV业务的扇出无关交换结构（博士论文）.2007,解放军信息工程大学.
[2]中国互联网络信息中心.第31次中国互联网络发展状况统计报告.http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201301.
[3]汪卫国.2005中国通信产业十大关键词-IPTV. http://www.cttl.cn/zhthg/invest/2005/dp/t20060621_368299.htm.
[4] A. M. Odlyzko. Comments on the Larry Roberts and Caspian Networks study ofInternet traffic growth. The Cook Report on the Internet, Dec.2001,12-15.
[5] S. Deering, R. Hinden. Internet Protocol, Version6(IPv6) Specification.RFC2460[S]. IETF,1998.
[6]中华人民共和国通信行业标准.路由器设备技术规范-低端路由器.YD/T1096-2001.
[7]中华人民共和国通信行业标准.路由器设备技术规范-高端路由器.YD/T1097-2001.
[8] S.Keshav and R. Sharma. Issues and Trends in Router Design. IEEECommunications Magazine, May1998,36(5):144-151.
[9] H.J. Chao.Next Generation Routers. Proceedings of the IEEE, Sep.2002,90(9):1518-1558.
[10] PMC-Sierra Inc. A New Architecture for Switch and Router Design.White paper,Dec.1999.
[11]徐恪,熊勇强,吴建平.宽带IP路由器的体系结构分析.软件学报,2000,11(2):179-186.
[12]周卫华.高性能路由器的交换技术研究（博士论文）.2004,北京邮电大学.
[13] N.F. Mir. Performance Evaluation of Efficient Multipath Crossbars.in Proc.8thInternational Conf.Communication Systems.2002,1:384-388.
[14] P. Wijetunga. High-Performance Crossbar Design for System-On-Chip.in Proc.3rdIEEE International Workshop System-On-Chip for Real-Time Applications,2003,138-143.
[15] Juniper Networks Inc. T-series Routing Platforms System and Packet ForwardingArchitecture.White Paper,2002.
[16]黄恺,徐志伟.可扩展并行计算技术、结构与编程.北京:机械工业出版社,2000年5月.
[17] L R Goke and G J Lipovski. Banyan networks for partitioning multiprocessorsystems. The1st Symp on Computer Architecture, Gainesville, FL, USA,1973.ACM,1973:21-28.
[18] S Turner. Design of a broadcast packet switching network. IEEE Transactions onCommunications,1988,36(6):734-743.
[19] F A Tobagi and T C Kwok. The tandem banyan switching fabric:a simplehigh-performance fast packet switch. Proceedings of INFOCOM’91, BalHarbour,Florida,1991. IEEE,1991:1245-1253.
[20] C. Clos. A Study of Non-Blocking Switching Network. Bell System TechnicalJournal,1953,32:406-424.
[21] Pankaj Gupta. Scheduling in Input-Queued Switches:A Survey. http://tiny_tera.stanford. edu/~Pankaj/papers.html.
[22] B. Prabhakar, N. McKeown, R. Ahuja. Multicast Scheduling for Input-QueuedSwitches. IEEE Journal on Selected Areas in Communications, Jun.1997,15(5):855-866.
[23] N. McKeown and B. Prabhakar. Scheduling multicast cells in an input-queuedswitch. in Proc.1996IEEE INFOCOM,1:271-278.
[24] N. McKeown. A fast switched backplane for a gigabit switched router. BusinessComm. Rev.,1997,27(12):1-17.
[25] H. Kuwahara, N. Endo,M. Ogino, et al. A shared buffer memory switch for anATM exchange. Proceedings of IEEE INFOCOM,1989,1:118-122.
[26] G.Kesidis and N.McKeown. Output-buffer ATM packet switching forintegrated-services communication networks. in Proc. IEEE ICC’97,Montreal,Canada， Aug.1997,3:1684-1688.
[27] B.Prabhakar, N.McKeown. On the speedup Required for Combined Input andOutput Queued Switching. Stanford University Technical Report, Nov.1997.
[28] I. Stoica, Zhang H. Exact Emulation of an Output Queuing Switch by a CombinedInput Output Queuing Switch. in proc.6th IEEE/IFIP IWQoS’s98,May1998.
[29] Lin D, Jiang Y, and M. Hamdi. Selective Request Round-Robin Scheduling forVOQ Packet Switch Architecture. Proceedings of IEEE International Conferenceon Communications (ICC),2011:1-5.
[30] Chang C S,Chen W J,Huang H Y. On service guarantees for input-bufferedcrossbar switches: a capacity decomposition approach by Birkhoff and vonNeumann. IEEE IWQoS’99,1999,7(4):79-86.
[31] Tamir Y,Chi H C. Symmetric crossbar arbiters for VLSI communication switches.IEEE Transaction of Parallel Distribut System,1993,4(3):13-27.
[32] T E Anderson, S S Owicki, J B Saxe,et al. High speed switch scheduling for localarea networks. ACM Transactions on Computer Systems,1993,11(3):319-352.
[33] P. Giaccone, D. Shah, B. Prabhakar. An implementable parallel scheduler forinput-queued switches. IEEE Micro, Jan.-Feb.2002,22(1):19-25.
[34] H.Obara, T.Yasushi. An efficient contention resolution algorithm for input queuingATM cross-connect switches. International Journal of Digital and Analog CabledSystems, Oct.-Dec.1989,2(4):261-267.
[35] M. J. Karol, K. Y. Eng, H. Obara. Improving the performance of input-queuedATM packet switches. Proceedings of IEEE INFOCOM’92,1992,1:110-115.
[36] M.J. Karol，M.G. Hluchyj, and S.P. Morgan. Input Versus Output Queueing on aSpace-Division Packet Switch. IEEE Transactions on Communications, Dec.1987,35(12):1347-1356.
[37] S.Q.Li and M.J.Lee.A Study of Traffic Imbalances in a Fast Packet Switch.Proceeding of IEEE INFOCOM’89，Apr.1989,2:538-547.
[38] Y. Tamir and G. Frazier. High Performance Multi-Queue Buffers for VLSICommunications Switches. Proceedings of Computer Architecture, Jun.1988,343-354.
[39] M.Ali andH.Nguyen. A neural network implementation of an input access schemein a high-speed packet switch. Proceeding of IEEE INFOCOM’89, Apr.1989,2:1192-1196.
[40] GSR12000Tech Product Description. Performing Internet Routing and Switchingat Gigabit speeds. Cisco Systems, http://www.cisco.com/warp/public/733/12000/index.shtml.
[41] J.Hui and E.Arthurs. A broadband packet switch for integrated transport. IEEEJ.Select. Areas Commun, Oct.1987,5(8):1264-1273.
[42] Y. Li. The dual round robin matching switch with exhaustive service. in Proc.Workshop on High Performance Switching and Routing,2002,58-63.
[43] C. Lund，S. Phillips，and N. Reingold. Fair prioritized scheduling in an inputbuffered switch. IFIP-IEEE Conference on Broadband Communications,Montreal, Canada, Apr.1996,358-369.
[44] N.McKeown,M.Izzard,A.Mekkittikul,et al. Tiny Tera: a packet switch core. IEEEMicro, Jan.-Feb.1997,17(1):26-33.
[45] Cook W J,Pulleyblank W R.Cunningham.Comb inatorial Optimization.WileyJohn&Sons Inc,1997,11(2):264-269.
[46] M. Arlitt, C. Williamson. Internet web servers workload characterization andperformance implications. IEEE/ACM Transactions on networking, Oct.1997,5(5):631-645.
[47] R. Caceres, P. Danzig, S. Jamin. Characteristics of wide-area TCP/IP conversatins.Proceedings of the ACM SIGCOMM'91. Zurich:ACM Press,1991:47-65.
[48] Leland W,Taqqu M,Willinger W,et al.On the self-similar nature of Ethernet traffic.Proceedings of ACM SIGCOMM'93. New York:ACM Press,1993:48-56.
[49] P Abry, D Veitch. Wavelet analysis of long rang dependent traffic. IEEETransaction on Information Theory,Jan.1998,44(1):2-15.
[50] A.Gupta and N.Georganas. Analysis of Packet Switch with Input and OutputBuffers and Speed Constraints. Proceeding of IEEE INFOCOM, Apr.1991,2:694-700.
[51] S.Chuang, A.Goel, N.McKeown,et al. Matching output queueing with a combinedinput output queued switch. IEEE INFOCOM,1999,3:1169-1178.
[52] S. Chuang, A. Goel, N. McKeown et al. Matching output queueing with acombined input/output-queued switch. IEEE Journal of Selected Areas inCommunications, Jun.1999,17(6):1030-1039.
[53] A.Smiljanic，R. Fan and G.Ramamurthy. RRGS round-robin greedy scheduling forelectronic/optical terabit switches.in Proc. GLOBECOM'99,2:1244-1250.
[54] D.Cavendish, CORPS: A pipelined fair Packet scheduler for high speedinputqueued switches. in Proc. IEEE Conference on High Performance Switchingand Routing,2000,55-64.
[55] A.Motoki, S.Kamiya, R.Ikematsu, et al. Group pipeline scheduler for inputbuffer switch. in Proc. Joint4th IEEE International Conference on ATM(ICATM2001) and High Speed Intelligent Internet Symposium,2001,158-162.
[56] E.Oki, R.Rojas-Cessa, H.J.Chao. PMM:a Pipelined maximal-sized matchingscheduling approach for input buffered switches. in Proc.of GLOBECOM,2001,1:35-39.
[57] H. Ahmadi, W. E. Denzel.A Survey of Modern High-Performance SwitchingTechniques. IEEE Journal on Selected Areas in Communications, Sept.1989,7(7):1091-1103.
[58] R. Y. Awdeh, H. T. Mouftah. Survey of ATM Switch Architectures. ComputerNetworks and ISDN Systems, Nov.1995,27(12):1576-1613..
[59] R. Rojas-Cessa, E. Oki, Jing Z, et al.. CIXB-1: combined input-one-cell-crosspoint buffered switch. IEEE Workshop on High Performance Switching andRouting,2001,324-329.
[60] M. Nabeshima. Performance evaluation of a combined input-and crosspoint-queued switch. IEICE Transaction on Communication, Mar.2000,3:737-741.
[61] K. Yoshigoe, K. J. Christensen. A parallel-polled virtual output queued switch witha buffered crossbar. IEEE workshop on high performance switching and routing,2001,271-275.
[62] K. Yoshigoe, K. J. Christensen. An evolution to crossbar switches with virtualoutput queuing and buffered cross points. IEEE Network, Sept.-Oct.2003,17(5):48-56.
[63] C. Ming, N.D. Georganas.A fast algorithm for multi-channel/port trafficassignment. International Conference on Communications,1994,1:96-100.
[64] M. Hluchyj, M.Karol.Queueing in high-performance packet switching. IEEEJ.Selected Area Communications, Dec.1988,6(9):1587-1597.
[65] N.McKeown, A.Mekkittikul, V.Anantharam,et al. Achieving100%throughput inan input-queued switch.IEEE Trans.Commu.,Aug.1999,47(8):1260-1267.
[66] A.Mekkittikul, N.McKeown. A Fair Algorithm for Achieving100%Throughput inan Input-Queued Switch. in Proceedings of IC3N96,Oct.1996,Maryland.
[67] Hakyong Kim, Changhwan Oh, Kiseon Kim. Throughput analysis of bifurcatedinput queued packet switches with restricted contention. Electronics Letters,Aug.1998,34(17):1561-1562.
[68] N. McKeown. Scheduling Algorithms for Input-Queued Cell Swithes:[PhDThesis].University of California at Berkeley,1995.
[69] J.E.Hopcroft, R.M. Karp.An n5/2Algorithm for Maximum Matching in BipartiteGraphs. SIAM Journal on Computing,1973,1(2):225-231.
[70] R.E.Tarjan. Data Structures and Network Algorithms. Philadelphia:Society forIndustrial and Applied Mathematics,1983:112-234.
[71] R.O.LaMaire, D.N.Serpanos. Two dimensional round-robin schedulers for packetswitches with multiple input queues. IEEE/ACM Trans. Networking, Oct.1994,2(5):471-482.
[72] N.McKeown. The iSLIP Scheduling Algorithm for Input-queued Switches.IEEE/ACM Transactions on Networking,Apr.1999,7(2):188-201.
[73] H.J.Chao. Saturn: A terabit packet switch using dual round robin. IEEECommunications Magazine, Dec.2000,38(12):78-84.
[74] M. Ajmone Marsan, A. Bianco, P. Giaccone, et al. On the throughput ofinput-queued cell-based switches with multicast traffic. in Proc.2001IEEEINFOCOM,3:1664–1672.
[75] G. Shanthi and A. Shanmugam. Mathematical analysis of the input-queued packetswitch under multicast traffic.IEE Proc.-Commun., Dec.2005,152(6):845-849.
[76] J. Y. Hui, T. Renner. Queueing Analysis for Multicast Packet Switching. IEEETransactions on Communications, Feb.-Apr.1994,42(2-4):723-731.
[77] Sun S T, He S M, Zheng Y F, et al. Multicast scheduling in buffered crossbarswitches with multiple input queues. in Proc. IEEE HPSR’05, Hong Kong, China.USA: IEEE Communication Society,2005:73-77.
[78] L. Mhamdi. On the integration of unicast and multicast cell scheduling in bufferedcrossbar switches. IEEE Trans. Parallel Distrib. Syst., Jun.2009,20(6):818–830.
[79] M. Song and W. Zhu. Throughput analysis for multicast switches with multipleinput queues. IEEE Commun. Lett., Jul.2004,8(7):479–481.
[80] Min Song, Weiying Zhu. Delay Analysis of Multicast Switches with MultipleInput Queues. High Performance Switching and Routing,2004,10-14.
[81] W. Zhu and M. Song. Performance analysis of large multicast packet switcheswith multiple input queues and gathered traffic. Computer Commun., May2010,33(7):803–815.
[82] J.Y. Hui. Switching and Traffic Theory for Integrated Broadband Networks.Kluwer Academic Publishers, Boston, MA,1990.
[83] Yu H, Ruepp S, Berger M S. Enhanced first-in-first-out-based round-robinmulticast scheduling algorithm for input-queued switches. IET Commun.,2011,5(8):1163-1171.
[84] A. Bianco, A.Scicchitano. Multicast support in multi-chip centralized schedulersin Input Queued switches. Computer Networks,2009,53(7):1040-1049.
[85] Hao Y, Ruepp S, Berger M S. Multi-Level Round-Robin Multicast Schedulingwith Look-Ahead Mechanism. Proceedings of IEEE International Conference onCommunications (ICC),2011:1-5.
[86] Pan Deng, Yang Yuanyuan. FIFO-Based Multicast Scheduling Algorithm forVirtual Output Queued Packet Switches. IEEE Transactions on Computers,Oct.2005,54(10):1283-1296.
[87] S. Gupta, A. Aziz. Multicast scheduling for switches with multiple input-queues.in Proc.2002Symposium on High Performance Interconnects,28–33.
[88] A. Bianco, P. Giaccone, E. Leonardi, et al. On the number of input queues toefficiently support multicast traffic in input queued switches. in Proc.2003Workshop on High Performance Switching and Routing,111–116.
[89] Hao Yu. A Novel Round-Robin Based Multicast Scheduling Algorithm for100Gigabit Ethernet Switches. IEEE INFOCOM, Mar.2010,1-2.
[90]戴精科,彭来献,张邦宁.一种支持单播与组播混合业务的高速Crossbar调度算法.电子与信息学报,2009,31(10):2299-2304.
[91] Zhu W., Song M. Integration of unicast and multicast scheduling in input-queuedpacket switches. Computer Networks, Apr.2006,50(5):667-687.
[92] M. Andrews, S. Khanna, K.Kumaran. Integrated scheduling of unicast andmulticast traffic in an input-queued switch. in Proc. IEEE INFOCOM, New York,USA, Mar.1999,3:1144–1151.
[93] K. Chin.A new integrated unicast/multicast scheduler for inputqueued Switches. inProc.8th Australasian Symp. on Parallel and Distributed Computing, Brisbane,Australia,2010,13–20.
[94] C.S.Chang,D.S.Lee, C.M.Lien. Load balanced Birkhoff-von Neumann switches,part I:one-stage buffering. Computer Comm.,2002,25:611-622.
[95] M.G.Nadkarni.Basic Ergoidc Theory.Berlin:Birkhaiiser,1998,21-45.
[96] K.Petersen.Ergodic Theory.Cambridge:University Press,1983,26-51.
[97] F.Baccelli,P.Bremand. Elements of queuing theory. New York: Springer-Verlag,1994,36-58.
[98] C.S.Chang. Performance guarantees in communication networks. London:Springer-Verlag,2000,11-24.
[99] M.Schwartz. Broadband integrated networks. New Jersey:Prentice Hall,1996,41-53.
[100] Juniper Networks Inc. T-series Routing Platforms: system and PacketForwarding Architecture, White Paper,2002.
[101] B.Alleyne, I.Scherson. On Evil-Twin Networks and the Value of LimitedRandomized Routing. IEEE Transactions Parallel and Distributed Systems,2000,l(9).910-925.
[102] E. Oki, Z. Jing, R. Rojas-Cessa, et al. Concurrent round-robin dispatchingscheme in a Clos-network switch. in: IEEE ICC_2001, Jun.2001,1:107-111.
[103] E. Oki, Z. Jing, R. Rojas-Cessa, et al. Concurrent Round-Robin-BasedDispatching Schemes for Clos-Network Switches. IEEE/ACM Trans.Networking,2002,10(6):830-844.
[104] P.Konghong, M.Hamdi. Static Round-robin Dispatching Schemes forClos-network Switches. IEEE HPSR.2002,329-333.
[105]杨帆,邱智亮,徐展琦.输入缓存Clos网络中分布式正交分路路由算法.计算机工程与应用,2007,43(10):162-166.
[106] W. Wang, L. Yu, G. Zhu, et al. Coding-driven Scheduling for Frame-basedMulticast Switches. in Proc. International Symposium on Network Coding, Jul.2011,4.
[107] M. Kim, J. K. Sundararajan, M. Medard. Network Coding for Speedup inSwitches. in Proc. IEEE International Symposium on Information Theory, Jun.2007,1086-1090.
[108] M. Kim, J. K. Sundararajan, M. Medard. Network Coding in a Multicast Switch.IEEE Transactions on Information Theory, Jan.2011,57(1):436-460.
[109] S. Yang, X. Wang, B. Li. Haste: Practical Online Network Coding in a MulticastSwitch. in Proc. IEEE INFOCOM,2010,5.
[110]李挥,林良敏,黄佳庆.应用网络编码的分组交换调度算法.中兴通讯技术，Feb.2009,15(1):20-27.
[111]胡曦明.有效支持IPTV业务的扇出无关交换结构（博士论文）.2007,解放军信息工程大学.