Improving the performance of End-to-End single rate Multicast Congestion Control

Abstract

End-to-end multicast congestion control (MCC) is a complex problem. There exist two categories of multicast congestion control. The first category is single rate MCC, where the source adjusts its rate according to the slowest receiver within the session. The second category is multi-rate MCC, where the source has to disseminate the data in multiple layers and each receiver could subscribe to the required layers according to their own capacities. Among the recently proposed multicast congestion control protocols is TFMCC, which is an equation based single rate protocol. In TFMCC, at each RTT, each receiver estimates its throughput using an equation that estimates the steady-state throughput of a TCP source. The receiver with the lowest such value is declared as the current limiting receiver (CLR). The source then adjusts its rate to that of the CLR. Our objections to TFMCC are with respect to the equation adopted as well as to the slow reaction of the sender to congestion conditions. In this paper we propose two modifications to TFMCC. First, each receiver estimates its throughput using an equation that models the steady state throughput of a multicast source controlled according the additive increase multiplicative decrease (AIMD) approach. The second modification consists of monitoring RTT variations at the source side in order to rapidly adjust the sending rate to network conditions. Simulation results show that these two modifications results in remarkable performance improvement with respect to throughput, time to react, and magnitude of oscillations.