A practical and economical approach to resource allocation

Providing quality of service guarantees requires resource reservations. As congestion becomes the common scenario, it is important to reserve network resources efficiently. However, many proposed algorithms make assumptions that can not be validated in a realistic framework. We make practical assumptions to enable immediate implementation of the proposed algorithm. In order to make the best use of the existing network resources, simple pricing is used. We use the burstiness bound approximation (BBA) traffic model which is a derivative of the leaky bucket traffic model (leaky bucket parameters and the peak rate). The specified traffic parameters directly relate to the source's reserved buffer and bandwidth requirements. This paper describes a method in which the bandwidth and buffer resources to be reserved are determined based on the resource availabilities (prices) and the specified traffic parameters to yield efficient utilization of network resources. The simulation results are presented to show its effectiveness.