Single failure resiliency in greedy routing

Using greedy routing, network nodes forward packets towards neighbors which are closer to their destination. This approach makes greedy routers significantly more memory-efficient than traditional IP-routers using longest-prefix matching. Greedy embeddings map network nodes to coordinates, such that greedy routing always leads to the destination. Prior works showed that using a spanning tree of the network topology, greedy embeddings can be found in different metric spaces for any graph. However, a single link/node failure might affect the greedy embedding and causes the packets to reach a dead end. In order to cope with network failures, existing greedy methods require large resources and cause significant loss in the quality of the routing (stretch loss). We propose efficient recovery techniques which require very limited resources with minor effect on the stretch. As the proposed techniques are protection, the switch-over takes place very fast. Low overhead, simplicity and scalability of the methods make them suitable for large-scale networks. The proposed schemes are validated on large topologies with properties similar to the Internet. The performances of the schemes are compared with an existing alternative referred as gravity pressure routing.

[1]  Alon Itai,et al.  The Multi-Tree Approach to Reliability in Distributed Networks , 1988, Inf. Comput..

[2]  David E. Culler,et al.  Beacon vector routing: scalable point-to-point routing in wireless sensornets , 2005, NSDI.

[3]  Brad Karp,et al.  GPSR: greedy perimeter stateless routing for wireless networks , 2000, MobiCom '00.

[4]  Muriel Médard,et al.  Redundant trees for preplanned recovery in arbitrary vertex-redundant or edge-redundant graphs , 1999, TNET.

[5]  Albert,et al.  Emergence of scaling in random networks , 1999, Science.

[6]  Christos H. Papadimitriou,et al.  On a conjecture related to geometric routing , 2004, Theor. Comput. Sci..

[7]  Robert D. Kleinberg Geographic Routing Using Hyperbolic Space , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[8]  Marwan Krunz,et al.  Disjoint multipath routing using colored trees , 2007, Comput. Networks.

[9]  C. Antal,et al.  Distributed resilient architecture for Ethernet networks , 2005, DRCN 2005). Proceedings.5th International Workshop on Design of Reliable Communication Networks, 2005..

[10]  Dave Katz,et al.  Bidirectional Forwarding Detection (BFD) , 2010, RFC.

[11]  Pramod K. Varshney,et al.  A survey of void handling techniques for geographic routing in wireless networks , 2007, IEEE Communications Surveys & Tutorials.

[12]  Didier Colle,et al.  Link failure recovery technique for greedy routing in the hyperbolic plane , 2013, Comput. Commun..

[13]  Mark Crovella,et al.  Hyperbolic Embedding and Routing for Dynamic Graphs , 2009, IEEE INFOCOM 2009.