Admission control makes accept or deny decisions about incoming service requests in order to avoid overload. Existing media streaming server software includes only limited support for admission control by allowing for predefined static threshold values, mostly concerning the maximum utilization level of a single resource such as the network bandwidth. In media streaming applications, however, the bottleneck resource (CPU, network, etc.) might change over time depending on the current demand for different media types. This thesis proposes and experimentally evaluates a model and a respective controller for adaptive admission control for differentiated media streaming services in the presence of multiple scarce resources. Opportunity costs of a service request are determined at the moment of an incoming request and compared to the revenue generated by an accepted request in order to make an accept/deny decision. The model allows the early rejection of service requests to reserve capacity required to perform future service requests with higher revenues in order to maximize overall revenue.
«
Admission control makes accept or deny decisions about incoming service requests in order to avoid overload. Existing media streaming server software includes only limited support for admission control by allowing for predefined static threshold values, mostly concerning the maximum utilization level of a single resource such as the network bandwidth. In media streaming applications, however, the bottleneck resource (CPU, network, etc.) might change over time depending on the current demand for...
»