Architecture for Guaranteed Delay Service in High Speed Networks

Title of Dissertation: Architecture for Guaranteed Delay Service in High Speed Networks Vinod Peris, Doctor of Philosophy, 1997 Dissertation directed by: Professor Armand Makowski Department of Electrical Engineering The increasing importance of network connections coupled with the lack of abundant link capacity suggests that the day when service guarantees are required by individual connections is not far o . In this dissertation we describe a networking architecture that can e ciently provide end-to-end delay guarantees on a per-connection basis. In order to provide any kind of service guarantee it is imperative for the source tra c to be accurately characterized at the ingress to the network. Furthermore, this characterization should be enforceable through the use of a tra c shaper (or similar device). We go one step further and assume an extensive use of tra c shapers at each of the network elements. Reshaping makes the tra c at each node more predictable and therefore simpli es the task of providing e cient delay guarantees to individual connections. The use of per-connection reshapers to regulate tra c at each hop in the network is referred to as a Rate Controlled Service (RCS) discipline in [52]. By exploiting some properties of tra c shapers we demonstrate how the per-hop reshaping does not increase the bound on the end-to-end delay experienced by a connection. In particular, we show that an appropriate choice of tra c shaper parameters enables the RCS discipline to provide better end-to-end delay guarantees than any other service discipline known today. The RCS discipline can provide e cient end-to-end delay guarantees to a connection; however, by de nition it is not work-conserving. This fact may increase the average delay that is observed by a connection even if there is no congestion in the network. We outline a mechanism by which an RCS discipline can be modi ed to be work-conserving without sacri cing the e cient end-to-end delay guarantees that can be provided to individual connections. Using the notion of service curves to bound the service process at each network element, we are able to provide an upper bound on the bu ers required to ensure zero loss at the network element. Finally, we examine how the RCS discipline can be used in the context of the Guaranteed Services speci cation that is currently in the process of being standardized by the Internet Engineering Task Force. ARCHITECTURE FOR GUARANTEED DELAY SERVICE IN HIGH SPEED NETWORKS