Dirk Abendroth

Effective bandwidth shaping: a framework for resource dimensioning-extended version

Effective bandwidths is a statistical descriptor in the context of stochastic models for statistical sharing of resources. It is generally known that effective bandwidths assigned to independent flows being multiplexed behave additively. Furthermore it has been shown by Wischik that under the many sources limiting regime, an effective bandwidth assigned to a flow passing through a switch is not changed, and inductively a flow has the same effective bandwidth through the entire network. If all flows feeding a network under consideration have well defined effective bandwidths and QoS demands, resource dimensioning can be done based on simple large deviation principles. We address the problem of specifying and limiting effective bandwidths of arriving flows by introducing a new shaping and policing algorithm based on an on-line effective bandwidth estimator. The presented algorithms functionality is verified by both synthetic and real data traces and the performance of algorithm variants is discussed. Finally, it is shown that the invariance of effective bandwidths (as proved for an infinite number of sources by Wischik) approximately holds already for a finite ensemble of competing connections, even in the presence of TCP traffic.

The interaction between TCP and traffic shapers - clever alternatives to the leaky bucket

The Differentiated Services (DiffServ) architecture has been defined for implementing scalable service differentiation in the Internet. DiffServ edge routers employ traffic conditioners that contain meter, marker and shaper functionalities. The way of traffic shaping in traffic conditioners is an open issue and we believe, using intelligent shaping algorithms instead of leaky bucket algorithms (LBA) could increase the network performance significantly. We present two new shaping algorithms for variable packet size networks that emit packet streams which behave in a certain sense better than Poisson streams and compare them to the leaky bucket algorithm (LBA). The introduced shaping algorithms are proven to perform better than LBA and furthermore provide a means to network providers to answer the questions of network dimensioning.

Effective bandwidth shaping: a framework for resource dimensioning

Effective bandwidths is a statistical descriptor in the context of stochastic models for statistical sharing of resources. It is generally known [C.-S Chang and J.A. Thomas, August 1995] that effective bandwidths assigned to independent flows being multiplexed behave additively. Furthermore it has been shown [D. Wischik, 1999] that under the many sources limiting regime, an effective bandwidth assigned to a flow passing through a switch is not changed, and inductively a flow has the same effective bandwidth through the entire network. If all flows feeding a network under consideration have well defined effective bandwidths and QoS demands, resource dimensioning can be done based on simple large deviation principles. We address the problem of specifying and limiting effective bandwidths of arriving flows by introducing a new shaping and policing algorithm based on an on-line effective bandwidth estimator. The presented algorithms functionality is verified by both synthetic and real data traces. Finally the performance of proposed algorithm variants is discussed.

Traffic Shaping in a Traffic Engineering Context

Modern communication networks carry a broad range of various applications such as data, voice, and video, and network providers are increasingly faced to the problems of traffic and resource management in order to meet the increasing quality of service (QoS) requirements. In particular, multi-media and realtime applications are sensitive with respect to QoS. In this paper we propose novel traffic engineering concepts for QoS that, in contrast to already existing solutions, need no intelligent functionality inside the core network and furthermore provide access to resource dimensioning in a comfortable way.

Service Separation and QoS in ATM Networks: The RCC+ Multiplexer

Providing service separation and quality of service guarantees requires the use of scheduling algorithms in the switches and network interfaces. We present the RCC+ Mux, a hardware efficient Rate Controlled Cell Multiplexer that has the capablity to provide explicit bandwidth reservation guarantees and thereby an excellent service separation. The RCC+ Multiplexer is simulated in the presence of self-similar traffic and compared to the FIFO queue, the Worst-case Fair Weighted Fair Queueing plus (WF2Q+) scheduler, and the already existing RCC Multiplexer. Moreover we point out the most favourable and critical characteristics of the RCC+ Multiplexer.