Proportional differentiated services

Abstract

Internet applications and users have very diverse service expectations, making the current same-service-to-all model inadequate and limiting. In the relative differentiated services approach, the network traffic is grouped in a small number of service classes which are ordered based on their packet forwarding quality , in terms of per-hop metrics for the queueing delays and packet losses. The users and applications, in this context, can adaptively choose the class that best meets their quality and pricing constraints, based on the assurance that higher classes will be better, or at least no worse, than lower classes . In this work, we propose the proportional differentiation model as a way to refine and quantify this basic premise of relative differentiated services. The proportional differentiation model aims to provide the network operator with the ' tuning knobs ' for adjusting the quality spacing between classes, independent of the class loads ; this cannot be achieved with other relative differentiation models, such as strict prioritization or capacity differentiation. We apply the proportional model on queueing-delay differentiation only, leaving the problem of coupled delay and loss differentiation for future work. We discuss the dynamics of the proportional delay differentiation model and state the conditions under which it is feasible . Then, we identify and evaluate (using simulations) two packet schedulers that approximate the proportional differentiation model in heavy-load conditions, even in short timescales. Finally, we demonstrate that such per-hop and class-based mechanisms can provide consistent end-to-end differentiation to individual flows from different classes, independently of the network path and flow characteristics.

A simple and scalable handoff prioritization scheme

“…The Random-Move scheme consists of a 3x3 mobility cell scenario (with one base station in the center of each cell) where eight cells (2)(3)(4)(5)(6)(7)(8)(9) are in use (cf., Fig. 11).…”

Transform-Domain Analysis of Packet Delay in Network Nodes with QoS-Aware Scheduling

“…A simulation study of DiffServ with both delay and loss QoS classes is found in [44]. A controllable DiffServ mechanism is Proportional Differentiated Services [18], which aims to quantitatively differentiate the QoS among the classes relative to some predefined parameters. Many other scheduling mechanisms that limit the delay of a selected flow keep track of a 'deadline' for each queued packet, such as earliest deadline first (EDF) [43,38] and alternative best-effort [26] or use virtual clocks [52].…”

On tuning and complexity of an adaptive model predictive control scheduler

“…DiffServ has sacrificed accuracy to improve scalability of IntServ architecture. In many recent works the DiffServ accuracy is strengthened besides reduction in the complexity of the network especially at core routers (Dovrolis, 2000;Mahramian, Taheri, & Haeri, 2005;Stoika & Zhang, 1998;Striegel & Manimaran, 2002). They assume nothing about traffic shaping and minimize or avoid on line negotiations about traffic conditions between routers.…”

“…An idea to increase QoS accuracy while preventing complexity was proposed by Dovrolis (2000). He assumed proportional delay and loss guarantees for different traffic classes instead of an absolute delay or loss for each class.…”

“…This is done using efficient scheduling algorithms. The proportional average delay (PAD), waiting time priority (WTP) and hybrid proportional delay (HPD) scheduler are the proposed scheduling algorithms in Dovrolis (2000) that guarantee proportional delay, but they are not accurate enough for strong QoS.…”