Distributed multipath routing with packet allocation based on the attractor renewal model

Information communication networks are rapidly growing recently. Because of increase of communication overhead, traditional routing protocols using global information of the network, such as topology of the whole network or traffic demands between most of pairs of routers, are facing difficulty in reliable routing. To alleviate this, distributed routing protocols relying only on local observables of the network attract much attention recently. The nonrequirement of global knowledge of the whole network largely reduces communication overhead of these protocols. However, the lack of knowledge can also be a significant drawback of them because they cannot promptly respond to traffic changes that occur on out of their local scopes. It means that network resources cannot be utilized sufficiently. To solve the problem, here, by extending an existing distributed routing protocol called ARAS, we propose a novel routing protocol that utilizes multiple paths in parallel. The protocol adaptively modulates packet allocation ratio to paths based only on local observables of the network. Multipath routing, however, easily give flapping of packet allocation due to competition among multiple routers. We study competition between routers and provide ways to suppress the flapping. We show validity of the proposal using a network simulation where prompt response of packet reallocation is required.

[1]  Kenji Leibnitz,et al.  Biologically Inspired Adaptive Multi-Path Routing in Overlay Networks , 2005 .

[2]  Raj Jain,et al.  Architectures for the future networks and the next generation Internet: A survey , 2011, Comput. Commun..

[3]  Zygmunt J. Haas,et al.  Multipath routing in the presence of frequent topological changes , 2001, IEEE Commun. Mag..

[4]  Marco Dorigo,et al.  AntNet: Distributed Stigmergetic Control for Communications Networks , 1998, J. Artif. Intell. Res..

[5]  K. Kaneko,et al.  Adaptive Response of a Gene Network to Environmental Changes by Fitness-Induced Attractor Selection , 2006, PloS one.

[7]  Douglas C. Sicker,et al.  A Survey on Biologically Inspired Algorithms for Computer Networking , 2013, IEEE Communications Surveys & Tutorials.

[8]  Yong Deng,et al.  A Bio-Inspired Algorithm for Route Selection in Wireless Sensor Networks , 2014, IEEE Communications Letters.

[9]  Ditixa Vyas,et al.  Survey of Distributed Multipath Routing Protocols for Traffic Management , 2013 .

[10]  John Moy,et al.  OSPF for IPv6 , 1999, RFC.

[11]  Zygmunt J. Haas,et al.  Analysis of multipath routing, part 2: mitigation of the effects of frequently changing network topologies , 2004, IEEE Transactions on Wireless Communications.

[12]  Ning Wang,et al.  An overview of routing optimization for internet traffic engineering , 2008, IEEE Communications Surveys & Tutorials.

[13]  Naoki Wakamiya,et al.  An Adaptive Routing Protocol with Balanced Stochastic Route Exploration and Stabilization Based on Short-Term Memory , 2016, IEICE Trans. Commun..

[14]  Naoki Wakamiya,et al.  Distributed Routing Protocol Based on Biologically-Inspired Attractor Selection with Active Stochastic Exploration and a Short-Term Memory , 2014, 2014 IEEE 11th Intl Conf on Ubiquitous Intelligence and Computing and 2014 IEEE 11th Intl Conf on Autonomic and Trusted Computing and 2014 IEEE 14th Intl Conf on Scalable Computing and Communications and Its Associated Workshops.

[15]  Kenji Leibnitz,et al.  Biologically inspired self-adaptive multi-path routing in overlay networks , 2006, Commun. ACM.

[16]  Scott A. Brandt,et al.  Modeling, Analysis and Simulation of Flash Crowds on the Internet , 2004 .

[17]  BERNARD M. WAXMAN,et al.  Routing of multipoint connections , 1988, IEEE J. Sel. Areas Commun..