Revenue-maximizing server selection and admission control for IPTV content servers using available bandwidth estimates

We present a server selection and admission control algorithm for IPTV networks that uses available bandwidth estimation to assess bandwidth available on the path from an end-user point of attachment to one or more IPTV content servers and that employs a revenue maximising admission decision process that prioritizes requests for high revenue content item types over requests for lower revenue item types. The algorithm operates by estimating expected request arrival rates for different content item types based on past arrival rates and, based on these and available bandwidth estimates decides whether to accept a new request and, when accepting requests, which of the available content servers to use. Results of a simulation study show that the algorithm succeeds in 1) maintaining acceptable packet delays for accepted flows in the presence of fluctuating background traffic on network paths and 2) when available bandwidth is limited prioritizing requests for higher revenue content types.