Higher bandwidths in computer networks have made application with real-time constraints, such as control, command, and interactive voice and video communication feasible. We describe two congestion control mechanisms that utilize properties of real-time applications. First, many real-time applications, such as voice and video, can tolerate some loss due to signal redundancy. We propose and analyze a congestion control algorithm that aims to discard packets if they stand little chance of reaching their destination in time as early on their path as possible. Dropping late and almost-late packets improves the likelihood that other packets will make their deadline. Secondly, in real-time systems with fixed deadlines, no improvement in performance is gained by arriving before the deadline. Thus, packets that are late and have many hops to travel are given priority over those with time to spare and close to their destination by introducing a hop-laxity priority measure. Simulation results show marked improvements in loss performance. The implementation of the algorithm within a router kernel for the DARTnet test network is described in detail. Because of its unforgiving real-time requirements, packet audio was used as one evaluation tool; thus, we developed an application for audio conferencing. Measurements with that tool show that traditional source models are seriously flawed. Real-time services are one example of traffic whose perceived quality of service depends not only on the loss rate but also on the correlation of losses. We investigate the correlation of losses due to buffer overflow and deadline violations in both continuous and discrete-time queueing systems. We show that loss correlation does not depend on value of the deadline for M/G/1 systems and is generally only weakly influenced by buffer sizes. Per-stream loss correlation in systems with periodic and Bernoulli on/off sources are evaluated analytically. Numerical examples indicate that loss correlation is of limited influence as long as each stream contributes less than about one-tenth of the overall network load.
Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:dissertations-8727 |
Date | 01 January 1993 |
Creators | Schulzrinne, Henning G |
Publisher | ScholarWorks@UMass Amherst |
Source Sets | University of Massachusetts, Amherst |
Language | English |
Detected Language | English |
Type | text |
Source | Doctoral Dissertations Available from Proquest |
Page generated in 0.0018 seconds