We survey recent advances in theories and models for Internet Quality of Service (QoS). We start with the theory of network calculus, which lays the foundation for support of deterministic performance guarantees in networks, and illustrate its applications to integrated services, differentiated services, and streaming media playback delays. We also present mechanisms and architecture for scalable support of guaranteed services in the Internet, based on the concept of a stateless core. Methods for scalable control operations are also briefly discussed. We then turn our attention to statistical performance guarantees, and describe several new probabilistic results that can be used for a statistical dimensioning of differentiated services. Lastly, we review recent proposals and results in supporting performance guarantees in a best effort context. These include models for elastic throughput guarantees based on TCP performance modeling, techniques for some quality of service differentiation without access control, and methods that allow an application to control the performance it receives, in the absence of network support.

this article, we assume that an arriving packet increments only one counter. If we instead considered the case where each packet arrival updates C counters, the line rate on the interface would be CR

Abstract not found

We consider the transmission of variable bit rate (VBR) video over a network offering a guaranteed service such as ATM VBR or the guaranteed service of the IETF. The guaranteed service requires that the flow accepted by the network has to be conforming with a traffic envelope #; in return, it receives a service guarantee expressed by a network service curve #. Functions # and # are derived from the parameters used for setting up the reservation, for example, from the T-SPEC and R-SPEC fields used with the Resource Reservation Protocol (RSVP). In order to satisfy the traffic envelope constraint, the output of the encoder is fed to a smoother, possibly with some look-ahead. The resulting stream is transported by the network; at the destination, the decoder waits for an initial playback delay and reads the stream from the receive buffer. We consider the problem of whether there exists one optimal strategy at the smoother which minimizes the playback delay and the receive buffer size, given the traffic envelope # and the service curve #. We show that there does exist such an optimal smoothing, and give an explicit representation for it. We also obtain a simple expression for the smallest playback delay and playback buffer size which can be achieved over all possible smoothing and playback strategies. We show that the computation of optimal smoothing and minimum playback delay do not depend on the past. We show that separate delay equalization is optimal in the CBR case, but not otherwise. We also apply the theory to the analysis of which T-SPEC should be requested by a source-destination pair, given some playback delay and buffer constraint, and given the path characteristics advertised in RSVP PATH messages.

A shaper is a system that stores incoming bits in a buffer and delivers them as early as possible, while forcing the output to be constrained with a given arrival curve. A shaper is time invariant if the traffic constraint is defined by a fixed arrival curve# it is time varying if the condition on the output is given by a time varying traffic contract. This occurs, for example, with renegotiable variable bit rate (RVBR) services. We focus on the class of time varying shapers called time varying leaky bucket shapers# such shapers are defined by a fixed numbers of leaky buckets, whose parameters (rate and bucket size) are changed at specific transition moments. We assume that the bucket levels are kept unchanged at those transition moments ("no reset" assumption). Our main finding is an input-output characterisation for this class of time varying shapers. Then we apply it to the tradeoff in optimising the RVBR service, assuming that a perfect prediction of future traffic can be made. We provide two algorithms that solve the problem of finding, at any renegotiation, the parameters for a RVBR service, respectively when the knowledge of the input traffic is limited to the next interval (local optimisation problem) and when we dispose of the complete input traffic description (global optimisation problem). We compare, by means of simulation, the two resulting algorithms to study the validity of the local approach. We illustrate the impact of the "no-reset" assumption by analyzing on some examples the losses that occur when the source chooses the opposite approach, namely, the "reset" approach. Furthermore wesimulate the RVBR service versus the renegotiable constantbitrate (RCBR) service and illustrate that the RVBR approach can provide substantial benefits. Finally,we discuss...

This papers explores, primarily by means of analysis, the differences that can exist between individual and aggregate loss guarantees in an environment where guarantees are only provided at an aggregate level. The foc us is on understanding which traffic parameters are responsible for inducing possible deviations and to what extent. In addition, we seek to evaluate the level of additional resources, e.g., bandwidth or buffer, required to ensure that all individual loss measures remain below their desired target. The paper's contributions are in developing analytical models that enable the evaluation of individual loss probabilities in settings where only aggregate losses are controlled, and in identifying traffic parameters that play a dominant role in causing differences between individual and aggregate losses. The latter allows the construction of guidelines identifying what kind of traffic can be safely multiplexed into a common service class.

Packet switches (e.g., IP routers, ATM switches and Ethernet switches) maintain statistics for a variety of reasons: performance monitoring, network management, security, network tracing, and traffic engineering. The statistics are usually collected by counters which might, for example, count the number of arrivals of a specific type of packet, or count particular events, such as when a packet is dropped. The arrival of a packet may lead to several different statistics counters being updated. The number of statistics counters and the rate at which they are updated is often limited by memory technology. A small number of counters may be held in on-chip registers or in (on- or off-chip) SRAM. But often, the number of counters is very large, and hence they need to be stored in off-chip DRAM. However, the large random access times of DRAMs make it difficult to support high bandwidth links. The time taken to read, update and write a single counter would be too large, and worse still multiple counters may need to be updated for each arriving packet. In this paper we consider a specific architecture for storing and updating statistics counters. Smaller sized counters are maintained in fast (potentially on-chip) SRAM, while a large, slower DRAM maintains the full-sized counters. The problem is to ensure that the counter values are always correctly maintained at line-rate. We describe and analyze an optimal counter management algorithm (LCF-CMA), which minimizes the size of the SRAM required while ensuring correct line-rate operation of a large number of counters.

We propose Asymmetric Best-Effort, a novel service to provide a "throughput versus delay jitter" differentiated service for IP packets. With this service, every best effort packet is marked as either Green or Blue. Green packets, typically sent by real-time applications such as interactive audio, receive more losses during bouts of congestion than Blue ones. In return, they receive less delay jitter. Both Green and Blue services are best-effort. The incentive to choose one or other is based on the nature of one's traffic and on traffic conditions. If applications are TCP-friendly, those sending Blue packets will receive more throughput but also more delay jitter, than they would if they sent Green packets for a given network state and path.

The emergence of the Internet as the de facto communication infrastructure means that it is asked to carry an ever broadening range of application traffic with different requirements. This in turn has stressed its original, one-class, best-effort model, and has been one of the main drivers behind the many efforts aimed at introducing QoS. Those efforts have, however, experienced only limited success because the added complexity QoS introduces is at odds with the scaling requirements of the Internet. This has motivated a number of proposals that have strived to balance the need for service differentiation with the requirement for low complexity that the Internet scale mandates. This paper shares similar goals and proposes a simple scheme, BoundedRandomDrop (BRD), that supports multiple service classes. BRD focuses on loss differentiation, as although both losses and delay are important performance parameters, the steadily rising speed of Internet links, including access links, is progressively limiting the potential impact of delay differentiation. BRD offers strong loss differentiation capabilities with minimal implementation and deployment cost. BRD does not require the specification of a traffic profile, nor does it rely on call admission to control the amount of traffic in different classes. It guarantees each class losses that, when feasible, are no worse than a specified bound, but enforces differentiation only when required to meet those bounds. In addition, BRD is implemented using a single FIFO queue and a simple random dropping mechanism. The performance of BRD is investigated for a broad range of traffic mixes and shown to consistently achieve its goal of effective service differentiation. I.

Various real-time services, like webcasting, audio/videoconferencing and telemedicine, are being deployed over the Internet. This requires the network to provide the guarantee of the service being provided to the receiver. The needs of the application is specified in terms of the Quality of Service (QoS) metrics like desired bandwidth, response time, etc. End-to-end QoS can be provided most efficiently when each layer of the protocol stack translates the application provided requirement into layer specific requirement and satisfies the same. Network layer has a critical role to play in the QoS provision process. It provides the desired QoS by considering the QoS metrics in the path selection process. The focus of this paper is on the QoS routing algorithms and protocols for unicast and multicast in the IPv4 based Internet that constrains or optimizes an individual or combination of metrics. The desired features of a router supporting QoS have been discussed in detail. QoS routing leads to an increase