End-to-end quality of service specification and mapping: The third party approach

• Published in: Computer communications Vol. 33; no. 11; pp. 1354 - 1368
• Main Authors: Stojanovic, Mirjana D., Bostjancic Rakas, Slavica V., Acimovic-Raspopovic, Vladanka S.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier B.V 01.07.2010; Elsevier
• Subjects: Algorithms; Applied sciences; C (programming language); Computer science; control theory; systems; Computer simulation; Computer systems and distributed systems. User interface; Exact sciences and technology; Flexibility; Mapping; Mathematical models; Negotiations; Next generation network; Quality of service; Resource management; Service class selection; Service level agreement; Service negotiation; Software; Specifications; Service class selection; Service level agreement; Service negotiation; Next generation network; Quality of service; Performance evaluation; Adaptability; Equity; Modeling; Multiagent system; Bargaining; Resource management; Service quality
• ISSN: 0140-3664; 1873-703X
• DOI: 10.1016/j.comcom.2010.03.024

This article addresses an open issue of end-to-end service specification and mapping in next generation networks. A centralized approach has been considered, via the third party agent that manages negotiation process in a group of domains. We suggest a general structure of the service specification form, which contains technical parameters related with a particular service request. Relying on such general form, we further propose and evaluate an efficient algorithm for service class selection. The algorithm selects the most appropriate class for each domain by determining the closest conformance between the required and offered service level. We also present a numerical example that demonstrates operating of the algorithm. Performance evaluation has been performed on the prototype implementation. Results of the analysis have pointed out the following important advantages of our approach: fairness in resource consumption, flexibility in class selection according to specific management policy and adaptability to different quality of service models.