Worst-Case Delay Control in Multigroup Overlay Networks

This paper proposes a novel and simple adaptive control algorithm for the effective delay control and resource utilization of end host multicast (EMcast) when the traffic load becomes heavy in a multigroup network with real-time flows constrained by (sigma, rho) regulators. The control algorithm is implemented at the overlay networks and provides more regulations through a novel (sigma, rho, lambda) regulator at each group end host who suffers from heavy input traffic. To our knowledge, it is the first work to incorporate traffic regulators into the end host multicast to control heavy traffic output. Our further contributions include a theoretical analysis and a set of results. We prove the existence and calculate the value of the rate threshold rho* such that for a given set of K groups, when the average rate of traffic entering the group end hosts rho macr > rho*, the ratio of the worst-case multicast delay bound of the proposed (sigma, rho, lambda) regulator over the traditional (sigma, rho) regulator is O(1/Kⁿ) for any integer n. We also prove the efficiency of the novel algorithm and regulator in decreasing worst-case delays by conducting computer simulations.