6533b85bfe1ef96bd12ba991

RESEARCH PRODUCT

Comparison and analysis of the revenue-based adaptive queuing models

Lari KannistoAlexander SayenkoHämäläinen TimoJyrki Joutsensalo

subject

Queueing theoryMathematical optimizationIntegrated servicesComputer Networks and CommunicationsComputer scienceQuality of serviceReal-time computingRevenueTotal revenueWeighted fair queueingScheduling (computing)Shared resource

description

This paper presents several adaptive resource sharing models that use a revenue criterion to allocate bandwidth in an optimal way. The models ensure QoS requirements of data flows and, at the same time, maximize the total revenue by adjusting parameters of the underlying schedulers. Besides, the adaptive models eliminate the need to find the optimal static weight values because they are calculated dynamically. The simulation consists of several cases that analyse the models and the way they provide the required QoS guarantees. The simulation reveals that the installation of the adaptive model increases the total revenue and ensures the QoS requirements for all service classes. The paper also presents how the adaptive models can be integrated with the IntServ and DiffServ QoS frameworks.

yearjournalcountryeditionlanguage
2006-06-01Computer Networks
https://doi.org/10.1016/j.comnet.2005.09.002