An Experimental Investigation of Hyperbolic Routing with a Smart Forwarding Plane in NDN

V. Lehman, A. Gawande, R. Aldecoa, D. Krioukov, L. Wang, B. Zhang, and L. Zhang
IEEE/ACM International Symposium on Quality of Service
(IWQoS), 2016
November 1, 2016


Routing in NDN  networks must scale in terms of forwarding table size and routing protocol  overhead. Hyperbolic routing (HR) presents a potential solution to address  the routing scalability problem, because it does not use traditional  forwarding tables or exchange routing updates upon changes in network  topologies. Although HR has the drawbacks of producing suboptimal routes or  local minima for some destinations, these issues can be mitigated by NDN’s  intelligent data forwarding plane. However, HR’s viability still depends on  both the quality of the routes HR provides and the overhead incurred at the  forwarding plane due to HR’s sub-optimal behavior. We designed a new  forwarding strategy called Adaptive Smoothed RTT-based Forwarding (ASF) to  mitigate HR’s sub-optimal path selection. This paper describes our  experimental investigation into the packet delivery delay and overhead under  HR as compared with Named-Data Link State Routing (NLSR), which calculates  shortest paths. We run emulation experiments using various topologies with  different failure scenarios, probing intervals, and maximum number of next  hops for a name prefix. Our results show that HR’s delay stretch has a  median close to 1 and a 95th-percentile around or below 2, which does not  grow with the network size. HR’s message overhead in dynamic topologies is  nearly independent of the network size, while NLSR’s overhead grows  polynomially at least. These results suggest that HR offers a more scalable  routing