An analytical framework to design a diffserv network supporting EF-, AF- and BE-PHBs

Authors:
Mario Barbera;Alfio Lombardo;Giovanni Schembra;Andrea Trecarichi
Affiliations:
D.I.I.T. – University of Catania, Catania, Italy;D.I.I.T. – University of Catania, Catania, Italy;D.I.I.T. – University of Catania, Catania, Italy;D.I.I.T. – University of Catania, Catania, Italy
Venue:
QoS-IP'05 Proceedings of the Third international conference on Quality of Service in Multiservice IP Networks
Year:
2005

Citing 5
Cited 0

Random early detection gateways for congestion avoidance

IEEE/ACM Transactions on Networking (TON)
Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED

Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Modeling TCP behavior in a differentiated services network

IEEE/ACM Transactions on Networking (TON)
Modeling random early detection in a differentiated services network

Computer Networks: The International Journal of Computer and Telecommunications Networking
A fluid-based model of time-limited TCP flows

Computer Networks: The International Journal of Computer and Telecommunications Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

Differentiated Services (DiffServ) architecture enables IP networks to offer different QoS levels to different users and applications. In this context, the definition of analytical tools to dimension network resources and buffer parameters has become a challenging problem for telecommunications research. The target of this paper is to define a new model of a DiffServ domain, loaded by TCP sources requiring different PHB's. In particular three different PHBs are considered: the AF-PHB, used for TCP data traffic with throughput requirements, the EF-PHB, chosen by all aggregates of TCP traffic with both throughput and delay requirements, and the BE-PHB, assigned to TCP traffic with no QoS requirements. The analytical framework is applied to a case study to evaluate how a different setting of the bandwidth reserved for EF traffic can impact the performance offered by the Diffserv network, not only to EF traffic but also to AF and BE traffic.