Adaptive scheduling using the revenue-based weighted round robin
Alexander Sayenko, Timo Hämäläinen, J. Joutsensalo, Pertti Raatikainen · 2005
This paper presents an adaptive resource allocation model that is based on the WRR queuing discipline. The model ensures QoS requirements and tries to maximize a service provider's revenue by manipulating weights of the WRR scheduler. To adjust the weights, it is proposed to use the revenue criterion, which controls the allocation of free resources. The simulation considers a single node with the implemented model that serves several classes with different QoS requirements and traffic characteristics. It is shown that the total revenue can be increased due to the allocation of unused resources to more expensive service classes. Furthermore, the adaptive model eliminates the need to find the optimal static weight values as they are calculated dynamically.