摘要
伴随着互联网的快速发展和普及,在服务质量(比如抖动、时延、可靠性等)方面有较高要求的多媒体应用不断涌现,如VoIP (Voice over IP)、视频会议、流媒体音频/视频、VOD (Video On Demand)点播平台等。但是,目前互联网的网络层所提供的数据传输服务都是基于“尽力而为(Best Effort)"机制,其容错性较差,QoS机制不完善,无法保证上层应用的服务质量需求。
因此,重叠网应运而生。它是由分布在现有网络之上的各个自治域中的重叠网中继点以及中继点之间的逻辑链路组成的,区别于底层实际网络的虚拟网络。重叠网具有灵活、简单、方便对新业务进行扩展等特点,并且不需要对现有网络结构进行大规模修改,就可以为上层应用提供QoS支持,满足其对服务质量的要求。
重叠网路由(Overlay Routing)是通过重叠网络进行的路由模式是重叠网络的关键部分。本文重点研究了基于服务质量的重叠网路由技术,包括:
1)重叠网中继节点的选择与部署:重叠网节点的选择决定了重叠网的拓扑结构,而重叠网的拓扑特性又能够影响重叠网路由的性能。所以,对于重叠网节点部署问题的研究,在提高网络性能,增强网络弹性等方面具有重要意义;
2)重叠网路由扩展性的优化:伴随着重叠网应用的越来越普遍,重叠网的规模以及使用范围也在逐渐扩大。所以,为了减小不断增加的路由表对于系统工作效率的直接影响,对于重叠网路由扩展性的研究非常重要;
3)重叠网智能节点的设计与实现:重叠网由重叠网中继节点以及中继节点之间的逻辑链路组成。重叠网中可以提供对媒体数据的转发、对网络状态的监控以及对网络资源的调配等功能的中继节点,我们可以称之为智能节点。智能节点是重叠网络为上层应用提供服务的基础。因此,对于重叠网智能节点的研究同样具有重要的现实意义。
传统IP路由选择技术的实现过程可以称之为基于目标的逐跳单播路由选择技术(Hop-by-Hop Destination-based Unicast Routing)。随着网络技术的发展与互联网的快速增长,传统的路由技术显露出了它的局限性:
1)逐跳转发策略是基于IP目标查找的;
2)IP目标查找完全是基于目标单播地址的;
3)目的地相同的所有分组都沿相同的路径传输;
4)所有的路由器都必须包含完整的路由选择信息,以便可靠地转发分组。
与传统的互联网路由相比,重叠网路由具有更大的灵活性。重叠网中的中继节点可以被看作是“智能网关”,它不仅可以根据“源地址、目的地址”,也可以根据报文的内容,并且结合其它需求进行下一跳节点的选择。因此,对于重叠网路由的研究已经成为业界的热门课题,越来越受到人们的关注。
本文一方面深入研究了重叠网路由技术以及相关问题(分为三个方面:面向网络层的重叠网路由:面向传输层的重叠网路由;面向应用层的重叠网路由),提出了基于复杂网络特性和基于路径差异程度的两种重叠网节点部署算法,以及用于优化路由扩展性的基于蚁群算法的邻居节点选择算法;另一方面将IPv6协议中的流标签作为切入点,把重叠网路由技术与下一代互联网QoS技术相结合,利用流标签的标识功能对重叠网智能节点进行设计,并提供相应的QOS算法。本文主要贡献和创新点包括如下几个方面:
1)本文基于复杂网络特性中节点的核数概念,提出了一种基于核数的重叠网节点部署算法-NCB (Node-Coreness Based),它能够在满足路径丰富程度的基础上,增强路由的稳定性以及可靠性,在网络拓扑发生变化的时候,保证数据和服务的有效性以及服务质量不会明显降低。此外,本算法还具有依据节点位置信息进行聚类的功能,它可以在满足节点部署基本要求的基础上,实现重叠网节点的均匀分布,从而避免中继点过于紧凑的情况。算法分为三个阶段:核数计算阶段;节点聚类阶段;中继点生成阶段。这种方法需要静态的全局拓扑信息,适用于比较稳定的拓扑环境,比如在大范围网络内规划部署覆盖节点。
2)本文提出了一种基于路径差异化系数的重叠网节点部署算法-PDB (Path Diversity based),算法中定义了节点的路径差异化系数,并将路径差异化系数的相关性定义为两个节点引起路由差异的近似性,将相关性超过一定阀值的节点聚为一类,在选择部署重叠网中继节点的位置时,只要在每个聚类中随机选择一个点即可,并且通过调整阈值,可以改变聚类的数量,也就改变了可以选择的中继节点的数量。该算法通过合理的选择重叠网中继节点,尽量增大重叠网路由与底层IP路由之间的差异,从而能够提高重叠网的故障绕避能力。这种方法同样需要静态的全局拓扑信息,适用于比较稳定的拓扑环境,比如在大范围网络内规划部署覆盖节点。
3)本章提出了一种基于蚁群算法的重叠网邻居节点选择算法来处理邻居节点集合的维护问题:通过蚂蚁爬行的过程来建立邻居节点集合,利用集合质量评价函数来判断结果集合对于网络覆盖度的高低,对于网络可靠性的影响强弱,并控制每个节点上的邻居集合大小,最终输出结果。本算法的核心思想是在不影响网络可靠性的基础上,对路由扩展性进行优化,从而减小由于不断增大的网络规模对系统工作效率产生的负面影响。
4)为了克服传统IP路由的种种局限性,优化网络性能表现,保障端到端通信的QoS需求,本文提出了一种面向QoS的重叠网路由服务网络-SQRON(Sensitive QoS Routing Overlay Network)。该方案利用重叠网智能节点作为载体,为上层的各种应用提供QoS支持。该网络提供性能测量、QoS等级映射、QOS路由以及网络性能监控等功能。此外,本文还对下一代互联网协议IPv6中的流标签字段进行了详细定义,对其格式进行了重新设计,增加了对QoS等级映射的支持等相应功能,并利用流标签能够标识数据流的特性,对重叠网路由服务网络-SQRON中的QoS路由协议进行了实现。
With the rapid development and comprehensive deployment of Internet, more and more multimedia applications have been appearing, which has special request for QoS (jitter, delay, reliability and etc), such as VoIP, video meeting, streaming audio\video, VOD and so on. However, the present Internet can only offer the transmission service based on "Best Effort" mechanism, it can not satisfy the QoS requests of the upper applications, since it has lots of disadvantages, such as less fault-tolerance, imperfect QoS mechanism and so on.
Therefore, the overlay network emerges. Different from the physical substrate, the overlay is virtual network that is composed of overlay relay nodes, distributing in the current network, and logic links among nodes. The overlay is capable to offer QoS support for upper applications and satisfy their QoS request, without massive modification on the current network structure due to its flexibility, simplicity and scalability.
The overlay routing, which is the key component of overlay network, is a routing pattern based on overlay network. This paper focuses on the overlay network routing technology on the basis of QoS, including:
1) Selection and deployment of overlay relay nodes:the overlay topology depends on the selection of overlay relay nodes, and the features of topology have influence on the overlay routing performance. Hence, the node deployment problem is important;
2) The optimization of overlay routing performance:with the overlay applications widely used, the scale and range of overlay network is expanding, and its influence keeps going strong. Consequently, the related research on overlay routing performance becomes more and more important;
3) Design and realization of overlay broker:the overlay is composed of relay nodes and logic links among them. These nodes, called overlay brokers that are foundation of services offered for upper applications, can provide many functions, such as data transmission, network state monitoring, network resource allocation and so on. Therefore, the corresponding research of overlay broker is also significant.
The realization of traditional IP routing is called as Hop-by-Hop Destination-based Unicast Routing. Along with development of technology and growth of Internet, the scheme reveals its limitations.
1) The hop-by-hop look-up strategy is based on IP destination;
2) The destination look-up is totally depending on unicast address;
3) All the packets have the same transmission path to the same destination;
4) All the routers must store the complete routing selection information for reliable packet transmission.
Compared to the conventional IP routing, the overlay routing has better flexibility. The overlay nodes, as "smart proxy", can choose the next hop not only by the "source, destination", but also the data content, even combining with other requests. Hence, the research of overlay routing has become a hot topic, attracting more and more attentions.
On the one hand, this paper focuses on the overlay routing and related problem, according to the Internet laying scheme, and proposes the node deployment algorithms based on the complex network topology characteristic and the neighbor node selection algorithm based on ACO (Ant Colony Optimization). On the other hand, taking advantage of the flow label in IPv6 main header, combining the overlay routing with the NGI QoS technology, this paper presents the design of overlay smart node relying on the identification feature of flow label, and provides the corresponding QoS routing algorithm. The main contributions and innovations are as follow:
1) This paper proposes an overlay node deployment algorithm, NCB (Node-Coreness Based), according to node-coreness, which can enhance the routing stability and reliability on the basis of the path diversity satisfaction. When network topology changing, the scheme is able to guarantee the data and service availability and prevent the severe QoS variation. Besides that, this algorithm has clustering function based on nodes location, it can achieve the uniform node distribution, avoiding the excessively compact node deployment. The algorithm is divided into three parts:coreness calculation phase; clustering phase; relay node generation phase. This placement method is suitable for stable network environment, such as the node deployment in big scale network, since it needs static overall topology information.
2) This paper proposes another overlay node deployment algorithm, PDB (Path-Diversity-Based), according to path diversity coefficient. All the potential overlay nodes are clustered based on the correlation of path diversity coefficient, which means the proximity of path difference caused by two different nodes, and the overlay node is selected randomly from each cluster. In this way, by means of reasonable overlay node selection, expanding the difference between overlay routing and IP routing, and overlay routing themselves can improve the overlay routing performance and fault detouring ability. This placement method is suitable for stable network environment, such as the node deployment in big scale network, since it needs static overall topology information.
3) This paper proposes an overlay neighbor selection algorithm based on ACO to handle neighborhood set maintenance:the neighborhood set is formed in the progress of ant crawling, which is determined according to the quality evaluation function in order to control the size of set and the influence, caused by neighborhood set, on the network cover extent and overlay routing reliability. The main idea is to reduce the negative influence, caused by increasing network scale, to system working efficiency, for the sake of improving the routing scalability, without network reliability dropping.
4) In order to overcome the limitations of traditional IP routing, improve the network performance and satisfy the E2E QoS request, this paper proposes an overlay network, called SQRON (Sensitive QoS Routing Overlay Network). It offers QoS support for upper applications by means of overlay brokers, which provide network performance measurement, QoS class mapping, QoS routing and network performance monitoring. Besides that, taking advantage of the flow label in IPv6 main header, combining the overlay routing with the NGI QoS technology, this paper presents the design of overlay smart node relying on the identification feature of flow label, and provides the corresponding QoS routing algorithm.
引文
[1]Gary Miller, "Overlay routing network". MIT 18.996. MIT technical report,2002.
[2]Y.Rekhter, T.J.Watson, T.Li. "A Border Gateway Protocol 4 (BGP-4)" IETF RFC 1771,1995.4, available at http://www.ietf.org/rfc/rfc1771.txt.
[3]D.G.Andersen, H.Balakrishnan, M.F.Kaashoek, and R.Morris, "Resilient Overlay Networks", in Proc. SOSP,2001, pp.131-145.
[4]Ion Stoica Lakshminarayanan Subramanian, Hari Balakrishnan, Randy Katz, "OverQoS: Offering QoS using Overlays", Proceedings of the First Workshop on Hot Topics in Networks 2002 (HotNets-1'02),2002.10,33(1), pp.11-16.
[5]Z.Duan, Z.Zhang, and Y. T. Hou, "Bandwidth provisioning for service overlay networks", SPIE ITCOM Scalability and Traffic Control in IP Networks (Ⅱ)'02, Boston, MA,2002, pp.139-150.
[6]M. Ripeanu, "Peer-to-Peer Architecture Case Study:Gnutella Network", in International Conference on Peer-to-Peer Computing (P2P2001),2001, pp.99-100.
[7]Y. Chu, S.G. Rao, and H. Zhang, "A case for end system multicast", in Proc. SIGMETRICS, 2000,pp.1-12.
[8]N. M. Malouch, Z.Liu, D.Rubenstein, and S. Sahu, "A graph theoretic approach to bounding delay in proxy-assisted end-system multicast", presented at the IWQoS'02, Miami Beach, FL, May 2002, pp.106-115.
[9]B. Krishnamurthy, C.E. Wills, and Y. Zhang, "On the use and performance of content distribution networks", in Proc. Internet Measurement Workshop,2001, pp.169-182.
[10]Y. Chawathe, S. McCanne, and E.A. Brewer, "RMX:Reliable Multicast for Heterogeneous Networks", in Proc. INFOCOM,2000, pp.795-804.
[11]P. Karwaczynski and J. Kwiatkowski, "Analysis of Overlay Network Impact on Dependability", in Proc. HICSS,2005, pp.290-291.
[12]Thomas Fuhrmann, "On the Topology of Overlay-Networks", ICON2003. The 11th IEEE International Conference,2003.9, pp.271-276.
[13]A.D. Keromytis, V. Misra, and D. Rubenstein, "SOS:secure overlay services", in Proc. SIGCOMM,2002,pp.61-72.
[14]John Jannotti, "Network layer support for overlay networks", In Proc. IEEE OPENARCH (New York),2002.6, pp.3-13.
[15]R. Pi and S. Junde, "Multi-Path Transmission based on Overlay Network", in Proc. AINA (2), 2004, pp.330-334.
[16]N.S.V. Rao, "Overlay networks of in situ instruments for probabilistic guarantees on message delays in wide-area networks", presented at IEEE Journal on Selected Areas in Communications,2004, pp.79-90.
[17]李海梅,“覆盖网络虚拟链路技术研究”,浙江大学硕士学位论文,2006.5
[18]Y. Amir and C. Danilov, "Reliable Communication in Overlay Networks", in Proc. DSN, 2003, pp.511-520.
[19]S. Shi and J.S. Turner, "Routing in Overlay Multicast Networks", in Proc. INFOCOM,2002.
[20]薛峰,赵问道,陈惠芳,“基于最大网络收益的DNS内容路由算法”浙江大学学报2004,10,pp.1270-1274.
[21]Sherlia Y.Shi, Jonathan S. Turner, "Multicast Routing and Bandwidth Dimensioning in Overlay Networks", IEEE Journal on Selected Area in Communication,22(8),2002.10, pp.1444-1455.
[1]Savage S, Anderson T, Aggarwal A, Becker D, Cardwell N, Collins A, Hoffman E, Snell J, Vahdat A, Voelker G, Zahorjan J. Detour:A case for informed internet routing and transport. IEEE Micro,1999, pp.50-59.
[2]Savage S, Collins A, Hoffman E, Snell J, Anderson TE. The end-to-end effects of Internet path selection. ACM SIGCOMM Computer Communication Review,1999, pp.289-299.
[3]Andersen DG, Balakrishnan H, Kaashoek MF, Morris R. Resilient overlay networks. In: Marzullo K, Satyanarayanan M, eds. Proc. of the 18th ACM Symp. on Operating Systems Principles (SOSP 2001). New York:ACM,2001, pp.131-145.
[4]Rewaskar S, Kaur J. Testing the scalability of overlay routing infrastructures. In:Chadi B, Ian P, eds. Proc. of the 5th Int'l Workshop on Passive and Active Network Measurement (PAM 2004). Berlin:Springer-Verlag,2004, pp.33-42.
[5]Gummadi KP, Madhyastha HV, Gribble SD, Levy HM, Wetherall D. Improving the reliability of Internet paths with one-hop source routing. In:Brewer E, Chen P, eds. Proc. of the 6th Symp, on Operating System Design and Implementation (USENIX OSDI 2004). Berkeley: USENIX Association,2004, pp.183-197.
[6]Fei T, Tao S, Gao LX, Guerin R. How to select a good alternate path in large peer-to-peer systems? In:Domingo-Pascual J, ed. Proc. of the IEEE INFOCOM 2006. Piscataway:IEEE, 2006, pp.1-13.
[7]X. Gu, K. Nahrstedt, R.N. Chang, and C. Ward, "QoS-Assured Service Composition in Managed Service Overlay Networks", in Proc. ICDCS,2003, pp.194-197.
[8]J.Jin, K.Nahrstedt. "Source-Based QoS Service Routing in Distributed Service Networks", in Proc. of IEEE International Conference on Communication 2004 (ICC2004), Paris, France, 2004, pp.2036-2041.
[9]J.Jin, K.Nahrstedt, "QoS Service Routing in One-to-One and One-to-Many Scenarios in Next Generation Service-Oriented Networks", in Proc. of The 23rd IEEE International Performance Computing and Communications Conference (IPCCC2004), Phoenix, AZ,2004, pp.503-510.
[10]J.Jin, K.Nahrstedt, "QoS Service for Supporting multimedia application", IEEE. Select Areas Communication,2002, pp.1228-1234.
[11]Z. Li and P. Mohapatra, "QRON:QoS-aware routing in overlay networks", presented at IEEE Journal on Selected Areas in Communications,2004, pp.29-40.
[12]S.Chen, K.Nahrstedt. "Distributed Quality-of-Service Routing in Ad-Hoc Networks", IEEE Journal on Special Areas in Communications,1999, pp.1-18.
[13]Nakao A, Peterson L, Bavier A. A routing underlay for overlay networks. In:Proc. of the ACM SIGCOMM 2003. New York:ACM,2003, pp.11-18.
[14]Kokku R, Bohra A, Ganguly S, Venkataramani A. A multipath background network architecture. In:Proc. of the IEEE INFOCOM 2007. Piscataway:IEEE,2007, pp.1352-1360.
[15]Venkatramani A, Kokku R, Dahlin M. TCP-Nice:A mechanism for background transfers. In: Proc. of the 5th Symp. on Operating System Design and Implementation (OSDI 2002). Berkeley:USENIX Association,2002, pp.329-343.
[16]Zhang L, Paxson V. Stream control transmission protocol. RFC2960,2000. http://www.ietf.org/rfc/rfc2960.txt
[17]Iyengar JR, Amer PD, Stewart R. Concurrent multipath transfer using SCTP multihoming over independent end-to-end paths. IEEE/ACM Trans. on Networking,2006, pp.951-964.
[18]Hsieh HY, Sivakumar R. pTCP. An end-to-end transport layer protocol for striped connections. In:Proc. of the 10th IEEE Int'l Conf. on Network Protocols (ICNP 2002). Piscataway:IEEE,2002, pp.24-33.
[19]Magalhaes L, Kravets R. Transport level mechanisms for bandwidth aggregation on mobile hosts. In:Proc. of the 9th Int'l Conf. on Network Protocols (ICNP 2001). Piscataway:IEEE, 2001,pp.165-171.
[20]Amir Y, Danilov C. Reliable communication in overlay networks. In:Proc. of the 2003 Int'l Dependable Systems and Networks (DSN 2003). Piscataway:IEEE,2003, pp.511-520
[21]Pucha H, Hu YC. Overlay TCP:Ending end-to-end transport for high throughput. In:Proc. of the ACM SIGCOMM 2005 (Poster). New York:ACM,2005, pp.22-26.
[22]Kwon GI, Byers JW. ROMA:Reliable overlay multicast with loosely coupled TCP connections. In:Proc. of the IEEE INFOCOM 2004. Piscataway:IEEE,2004, pp.385-395.
[23]Baccelli F, Chaintreau A, Liu Z, Riabov A. The one-to-many TCP overlay:A scalable and reliable multicast architecture. In:Makki K, Knightly E, eds. Proc. of the IEEE INFOCOM 2005. Piscataway:IEEE,2005, pp.1629-1640.
[24]Li Y, Zhang Y, Qiu L, Lam S. SmartTunnel:Achieving reliability in the Internet. In:Proc. of the IEEE INFOCOM 2007. Piscataway:IEEE,2007, pp.830-838.
[25]Hosseini M, Ahmed DT, Shinnohammadi S, Georganas ND. A survey of application-layer multicast protocols. IEEE Communications Surveys and Tutorials,2007, pp.58-74.
[26]Wu JG. A survey of routing algorithms and protocols in overlay multicast. Computer Science, 2007,pp.7-12.
[27]Christin N, Chuang J. A cost-based analysis of overlay routing geometries. In:Makki K, Knightly E, eds. Proc. of the IEEE INFOCOM 2005. Piscataway:IEEE,2005, pp.2566-2577.
[28]Li Z, Mohapatra P. The impact of topology on overlay routing service. In:Proc. of the IEEE INFOCOM 2004. Piscataway:IEEE,2004, pp.408-418.
[29]Ratnasamy S, Handley M, Karp RM, Shenker S. Topologically-aware overlay construction and server selection. In:Proc. of the IEEE INFOCOM 2002. Piscataway:IEEE,2002, pp.1190-1199.
[30]Jiang WJ, Chiu DM, Liu JCS. On the interaction of multiple overlay routing. Performance Evaluation,2005, pp.229-246.
[31]Qiu LL, Yang YR, Zhang Y, Shenker S. On selfish routing in Internet-like environments. In: Feldmann A, Zitterbart M, Crowcroft J, Wetherall D, eds. Proc. of the ACM SIGCOMM 2003. New York:ACM,2003, pp.151-162.
[32]Liu Y, Zhang HG, Gong WB, Towsley DF. On the interaction between overlay routing and underlay routing. In:Makki K, Knightly E, eds. Proc. of the IEEE INFOCOM 2005. Piscataway:IEEE,2005, pp.2543-2553.
[33]Seetharaman S, Ammar MH. On the interaction between dynamic routing in native and overlay layers. In:Proc. of the IEEE INFOCOM 2006. Piscataway:IEEE,2006, pp.1-12.
[34]Shi E, Stoica I, Andersen D, Perrig A. OverDoSe:A generic DDoS protection service using an overlay network. Technical Report, CMU-CS-06-114, Pittsburgh:Carnegie Mellon University,2006.
[35]Keromytis AD, Misra V, Rubenstein D. SOS:Secure overlay services. In:Proc. of the ACM SIGCOMM 2002. New York:ACM,2002, pp.61-72.
[36]Zhu YP, Rexford J, Bavier A, Feamster N. UFO:A resilient layered routing architecture. ACM SIGCOMM Computer Communication Review,2008, pp.59-62.
[37]Xu W, Rexford J. MIRO:Multi-Path interdomain Routing. In:Rizzo L, Anderson TE, McKeown N, eds. Proc. of the ACM SIGCOMM 2006. New York:ACM,2006, pp.171-182.
[38]Seetharaman S, Hilt V, Hofmann M, Ammar MH. Preemptive strategies to improve routing performance of native and overlay layers. In:Proc. of the IEEE FNFOCOM 2007. Piscataway: IEEE,2007.463-471.
[39]Kushman N, Kandula S, Katabi D, Maggs B. R-BGP:Staying connected in a connected world. In:Proc. of the 4th Symp. on Networked Systems Design and Implementation (NSDI 2007). Berkeley:USENIX Association,2007, pp.341-354.
[1]Savage S, Anderson T, Aggarwal A, Becker D, Cardwell N, Collins A, Hoffman E, Snell J, Vahdat A, Voelker G, Zahorjan J. Detour:A case for informed internet routing and transport. IEEE Micro,1999, pp.50-59.
[2]Andersen DG, Balakrishnan H, Kaashoek MF, Morris R. Resilient overlay networks. In: Marzullo K, Satyanarayanan M, eds. Proc. of the 18th ACM Symp. on Operating Systems Principles (SOSP 2001). New York:ACM,2001, pp.131-145.
[3]Mahadevan P, Krioukov D, Fomenkov M. The Internet AS-level topology:Three data sources and one definitive metric. Computer Communication Review,2006, pp.17-26.
[4]Alvarez-Hamelin J I, Dall' Asta L, Barrat A, et al. K-core decomposition:a tool for the visualization of large scale networks [EB/OL]. [2009-09-03]. http:// xavier.informatics.indiana.edu/lanet-vi/k-cores-AHDBV3.pdf.
[5]Marques JP, Written; Wu YF, Trans. Pattern Recognition Concepts, Methods and Applications. 2nd ed., Beijing:Tsinghua University Press,2002, pp.51-74.
[6]Huang Z. A fast clustering algorithm to cluster very large categorical data sets in data mining. In:Proc. of the SIGMOD Workshopon Research Issues on Data Mining and Knowledge Discovery. Tucson,1997, pp.146-151.
[7]Faloutsos C, Faloutsos M, Faloutsos P. On power-law relationships of the Internet topology. In: Proc. of the ACM SIGCOMM,1999, pp.251-262.
[8]Albert R, Jeong H, Barabasi AL. Error and attack tolerance in complex networks. Nature, 2000, pp.387-482.
[9]Newman M. Properties of highly clustered networks. Physical Review E,2003.
[10]Guo L, Xu XM. The Complex Network. Shanghai:Shanghai Scientific and Technological Education Publication,2006.
[11]Carlson JM, Doyle J. Complexity and robustness. Proc. of the National Academy of Sciences of the USA,2002, pp.2539-2545.
[12]Bollobas B, Riordan O. Robustness and vulnerability of scale-free random graphs. Internet Math,2003, pp.1-35.
[13]H. Pucha and Y. C. Hu, "Slot:Shortened Loop Overlay Transport," TRECE-05-12, Purdue University, Tech. Rep., July 2005.
[14]Pucha H, Hu YC. Overlay TCP:Ending end-to-end transport for high throughput. In:Proc. of the ACM SIGCOMM 2005 (Poster). New York:ACM,2005, pp.22-26.
[15]Kwon GI, Byers JW. ROMA:Reliable overlay multicast with loosely coupled TCP connections. In:Proc. of the IEEE INFOCOM 2004. Piscataway:IEEE,2004, pp.385-395.
[16]K. Calvert and E. Zegura, "Internetwork topology models (gt-itm)," http://www.cc.gatech.edu/fac/Ellen.Zegura/gt-itm.
[17]K. P. Gummadi, H. Madhyastha, S. D. Gribble, H. M. Levy, and D. J.Wetherall, "Improving the Reliability of Internet Paths with One-hop Source Routing," in Proc. of OSDI,2004, pp.183-198.
[18]Fei T, Tao S, Gao LX, Guerin R. How to select a good alternate path in large peer-to-peer systems? In:Domingo-Pascual J, ed. Proc. of the IEEE INFOCOM 2006. Piscataway:IEEE, 2006, pp.1-13.
[19]Zheng He, Yuheng Cao, Xiaohong Huang, and Yan Ma, "Overlay Node Deployment based on Node Coreness", Journal of China Universities of Posts and Telecommunications,2010.v17. pp:99-103.
[20]Zheng He, Yuheng Cao, Xiaohong Huang, Yan Ma, "Clustering based Overlay Node Placement Research",2010 The 3rd International Conference on Computational Intelligence and Industrial Application (PACIIA), Wuhan, China, December.
[1]Savage Stefan, Anderson Thomas, Aggarwal Amit et al. Detour:Informed Internet routing and transport. IEEE Micro,1999,19(1), pp.50-59.
[2]Dorigo, Maniezzo, Colomi. Ant system:optimization by a colony of cooperating agents[J]. IEEE Transactions on SMC,1996,26(1), pp.8-41.
[3]Dorigo, Gambardella. Ant colony system:a cooperative learning approach to the traveling salesman problem[R]. IEEE Transactions on Evolutionary Computing,1997,1(1), pp.53-56.
[4]韩冬梅,牛文清,杨荣.线性与非线性最优组合预测方法的比较研究[J].情报科学,2007,25(11),pp.1672-1678.
[5]Zheng He, Fan Wei, Xiaohong Huang, Yan Ma, "One-Hop Path Routing based on Heuristic-K method using ACO",2010 International Conference on Management Science and Information Engineering (ICMSIE), Zhengzhou, China, December.
[1]Savage Stefan, Anderson Thomas, Aggarwal Amit et al. Detour:Informed Internet routing and transport. IEEE Micro,1999,19(1), pp.50-59
[2]REKHTER, Y., AND LI, T. A Border Gateway Protocol 4 (BGP-4). Internet Engineering Task Force,1995. RFC 1771.
[3]CHANDRA, B., DAHLIN, M., GAO, L., AND NAYATE, A. End-to-end WAN Service Availability. In Proc.3rd USITS (San Francisco, CA,2001), pp.97-108.
[4]PAXSON, V. End-to-End Routing Behavior in the Internet. In Proc. ACM SIGCOMM'96 (Stanford, CA, Aug.1996), pp.25-38.
[5]PAXSON, V. End-to-End Internet Packet Dynamics. In Proc. ACM SIGCOMM (Cannes, France, Sept.1997), pp.139-152.
[6]J. Rajahalme, A. Conta, B. Carpenter, and S. Deering, "IPv6 Flow Label Specification," RFC3697, Mar.2004.
[7]M.Marchese, "QoS Over Heterogeneous Networks", John Wiley & Sons, Ltd,2007.
[8]IETF Working Group on Integrated Service. http://www.ietf.org/html/charters/ intserv-charter.html
[9]IETF Working Group on Differentiated Service. http://www.ietf.org/html/charters/ diffserv-charter.html
[10]http://www.ietf.org/
[11]http://www.3gpp.org/
[12]http://www.itu.int/ITU-T/
[13]http://www.etsi.org/
[14]ITU-T Recommendation Y.1541
[15]Y.Chu, S.G. Rao, and H. Zhang, "A case for end system multicast", in Proc. SIGMETRICS, 2000, pp.1-12.
[16]Zheng He, Xiaohong Huang, Yan Ma, "Overlay Routing Middleware for Multimedia Communications", Future Intelligent Information System, Lecture Notes in Electrical Engineering series 86, pp.543-550.
[17]He Zheng, Huang Xiaohong, Ma Yan, Sun Qiong, "The Optimization of QoS Path-Selection Using Flow Label based on Overlay Network",2009 International Conference on Communication Technology and Application (ICCTA), Beijing, China, Oct.2009, pp.241-245.
[18]Huang Xiaohong, He Zheng, Ma Yan, Sun Qiong, " Packet-switch mechanism based on QoS class mapping using flow label for overlay network", Journal of China Universities of Posts and Telecommunications,2010.v17. pp:17-23.