Output characterization of constant bit rate traffic in FIFO networks

We provide an analytical proof that the departure rate of a CBR flow at an overloaded link with FIFO buffers is proportional to the flow?s share of the total offered load at the link. This property of FIFO scheduling was recently validated in [1] in a series of traffic measurement experiments. An extension of the analysis to a multi-node scenario shows that the output rate of a flow in a network with many overloaded FIFO switches approaches the pessimistic values given by blind multiplexing.     

Scaling properties of statistical end-to-end bounds in the network calculus

The stochastic network calculus is an evolving new methodology for backlog and delay analysis of networks that can account for statistical multiplexing gain. This paper advances the stochastic network calculus by deriving a network service curve, which expresses the service given to a flow by the network as a whole in terms of a probabilistic bound. The presented network service curve permits the calculation of statistical end-to-end delay and backlog bounds for broad classes of arrival and service distributions. The benefits of the derived service curve are illustrated for the exponentially bounded burstiness (EBB) traffic model. It is shown that end-to-end performance measures computed with a network service curve are bounded by /spl Oscr/(H log H), where H is the number of nodes traversed by a flow. Using currently available techniques, which compute end-to-end bounds by adding single node results, the corresponding performance measures are bounded by /spl Oscr/(H/sup 3/).     

A QoS architecture for quantitative service differentiation

For the past decade, a lot of Internet research has been devoted to providing different levels of service to applications. Initial proposals for service differentiation provided strong service guarantees, with strict per-flow bounds on delays, loss rates, and throughput, but required high overhead in terms of computational complexity and memory, both of which raise scalability concerns. Recently, the interest has shifted to class-based service architectures with low overhead. However, these newer service architectures only provide weak service guarantees, which do not always address the needs of applications. In this article we introduce a service architecture that supports strong per-class service guarantees, can be implemented with low computational complexity, and only requires maintenance of a little state information. A key mechanism of the proposed service architecture is that service rate allocation to classes is adaptive, and combined with buffer management. Furthermore, instead of using admission control or traffic policing, the proposed architecture exploits explicit congestion notification for the purpose of regulating the traffic entering the network.     

An Interactive Telelecture System with Hybrid ATM/IP Networking

Currently, there is much debate whether ATM (Asynchronous Transfer Mode) or IP (Internet Protocol) is the better internetworking technology for multiparty multimedia applications. Proponents of ATM argue that only a connection-oriented network can satisfy the stringent timeliness requirements of high-quality audio and video data. Proponents of the Internet emphasize the scalability and flexibility of connectionless networking. In this paper a multiparty multimedia telelecture system, called Distance Learning Controller (DLC), is presented that uses both ATM and IP, thus, attempting to exploit the advantages of both technologies, without suffering the drawbacks of either technology. The DLC system uses ATM for transmission of broadcast-quality video and CD-quality audio, and IP-over-ATM for low-bandwidth applications and conference control functions. This paper describes the hardware and software components of the DLC system.     

An adaptive multipath routing scheme for connectionless traffic in an ATM network

This paper addresses the problem of routing connectionless traffic through an ATM network. A solution is proposed based on a per-packet adaptive multipath routing scheme which is added to the routing algorithm implemented at the Inter-Working Units. A scheme is presented that distributes packets among multiple Virtual Paths (VPs) according to the utilization of the links on these VPs. The utilization of the VPs is determined by a periodic feedback mechanism. Simulation studies show the effectiveness of the proposed adaptive multipath routing scheme.The work by J. Sole-Pareta was supported in part by a CIRIT (Generalitat de Catalunya) grant (expedient number EE92/2-338), and in part by the National Science Foundation under Grant No. INT-94033646. The work by I. Akyildiz was supported in part by the National Science Foundation under Grant No. INT-94033646.     

Multimedia networks: issues and challenges

Video on demand, teleconferencing, digital libraries, and remote tutoring systems are just a few of the many distributed multimedia applications that have emerged in recent years. Widespread use of such applications requires networks that can offer transport services to both discrete media (such as text and digital images) and continuous media (such as audio and video). The discussion explores some of the issues involved in the design of such networks     

Exact admission control for networks with a bounded delay service

To support the requirements for the transmission of continuous media, such as audio and video, multiservice packet-switching networks must provide service guarantees to connections, including guarantees on throughput, network delays, and network delay variations. For the most demanding applications, the network must offer a service which provides deterministically bounded delay guarantees, referred to as “bounded delay service.” The admission control functions in a network with a bounded delay service require `schedulability conditions' that detect violations of delay guarantees in a network switch. Exact schedulability conditions are presented for three packet scheduling methods: earliest-deadline-first (EDF), static-priority (SP), and a novel scheduling method, referred to as rotating-priority-queues (RPQ). By characterizing the worst-case traffic with general subadditive functions, the presented schedulability conditions can be applied to a large class of traffic models. Examples, which include actual MPEG video traces, are presented to demonstrate the trade-offs involved in selecting a packet scheduling method for a bounded delay service     

Enhancing Class-Based Service Architectures With Adaptive Rate Allocation and Dropping Mechanisms

Class-based service differentiation can be realized without resource reservation, admission control and traffic policing. However, the resulting service guarantees are only relative, in the sense that guarantees given to a flow class at any time are expressed with reference to the service given to other flow classes. While it is, in principle, not feasible to provision for absolute guarantees (i.e., to assure lower bounds on service metrics at all times) without admission control and/or traffic policing, we will show in this paper that such a service can be reasonably well emulated using adaptive rate allocation and dropping mechanisms at the link schedulers of routers. We name the resulting type of guarantees best-effort bounds. We propose mechanisms for link schedulers of routers that achieve these and other guarantees by adjusting the drop rates and the service rate allocations of traffic classes to current load conditions. The mechanisms are rooted in control theory and employ adaptive feedback loops. We demonstrate that these mechanisms can realize many recently proposed approaches to class-based service differentiation. The effectiveness of the proposed mechanisms are evaluated in measurement experiments of a kernel-level implementation in FreeBSD PC-routers with multiple 100 Mbps Ethernet interfaces, complemented with simulations of larger scale networks.     

DQDB+-/: a fair and waste-free media access protocol fordual bus metropolitan networks

A media access protocol that achieves a fair distribution of the bandwidth in one round-trip delay is presented. The protocol is based on a unique solution to a fair and waste-free bandwidth allocation. This bandwidth allocation can be implemented in a distributed manner. A comparison of the new protocol with the DQDB (distributed queue dual bus) protocol shows considerable advantages regarding the transmission delay of messages and the time a station needs to obtain a fair portion of the available bandwidth. The advantages of the protocol become more apparent for large networks and high transmission speeds. In addition, the new protocol can perform nonuniform bandwidth allocations     

Marking algorithms for service differentiation of TCP traffic

Class-based service architectures for quality-of-service (QoS) differentiation typically provide loss, throughput, and delay differentiation. However, proposals for class-based service differentiation generally do not account for the needs of TCP traffic, which are characterized by a coupling of packet losses and achievable throughput. Ignoring this coupling may result in poor service differentiation at the microflow level. This paper shows how Explicit Congestion Notification (ECN) can be used to achieve service differentiation for TCP traffic classes at the microflow level. We present a traffic-marking algorithm for routers, which, if used in conjunction with ECN, regulates the transmission rate of TCP sources in such a way that packet drops due to buffer overflows are avoided. We demonstrate how the algorithm can be integrated in a service architecture with absolute and proportional QoS guarantees. Simulation results illustrate the effectiveness of the presented algorithms at avoiding packet losses and regulating traffic for meeting service guarantees, and provide a comparison with other algorithms proposed in the literature.