Client Migration in a Continuous Data Network

Two major issues for continuous data delivery architectures on the Internet are scalabil-ity and data continuity. Scalability is addressed by providing multiple redundant stream sources. One solution to the data continuity problem is adaptive client migration. Adaptive client migration is the ability of a client to switch from one redundant stream source to another when it experiences discontinuities in its current stream. The switching mechanism allows the service providers to adapt to changing server loads and varying jitter due to network congestion. This thesis presents an adaptive client migration protocol that allows a client to migrate from one data stream to another without introducing discontinuities.