Abstract

Loss-load curves are a recently proposed feedback mechanism for rate-based congestion control in packet-switched networks. The novel feature of loss-load curves is that they explicitly provide a direct mathematical relationship between the offered load to the network and the level of packet loss within the network. With their feedback information, loss-load curves provide a means for sources to dynamically compete for and share the available bandwidth on a network. Switches within the network advertise their aggregate load information to the sources in the form of a loss-load curve. Sources at the edges of the network have the responsibility of choosing their own transmission rate based on their traffic requirements and on the loss-load tradeoff curves provided by the network. Sources can use loss-load curves to maximize raw throughput, to optimize response time for file transfer activities, or to control their end-to-end level of packet loss within the network. The focus of this paper is on the mathematical underpinnings of the loss-load model. The paper introduces the loss-load model, describes the mathematical foundations of loss-load curves, presents three key theoretical results for the loss-load model, and then discusses the advantages and disadvantages of the loss-load approach to dynamic bandwidth allocation. Formal proofs of the theoretical results are contained in the appendix of the paper.

Citations