Future Internet Routing Design for Massive Failures and Attacks

Abstract

Given the high complexity and increasing traffic load of the Internet, geo-correlated challenges caused by large-scale disasters or malicious attacks pose a significant threat to dependable network communications. To understand its characteristics, we propose a critical-region identification mechanism and incorporate its result into a new graph resilience metric, compensated Total Geographical Graph Diversity. Our metric is capable of characterizing and differentiating resiliency levels for different physical topologies. We further analyze the mechanisms attackers could exploit to maximize the damage and demonstrate the effectiveness of a network restoration plan. Based on the geodiversity in topologies, we present the path geodiverse problem and two heuristics to solve it more efficiently compared to the optimal algorithm. We propose the flow geodiverse problem and two optimization formulations to study the tradeoff among cost, end-to-end delay, and path skew with multipath forwarding. We further integrate the solution to above models into our cross-layer resilient protocol stack, ResTP–GeoDivRP. Our protocol stack is prototyped and implemented in the network simulator ns-3 and emulated in our KanREN testbed. By providing multiple GeoPaths, our protocol stack provides better path restoration performance than Multipath TCP.