All recently proposed packet-scheduling algorithms for output-buffered switches that support quality-of-service (QoS) transmit packets in some priority order, e.g., according to deadlines, virtual finishing times, eligibility times, or other time stamps that are associated with a packet. Since maintaining a sorted priority queue introduces significant overhead, much emphasis on QoS scheduler design is put on methods to simplify the task of maintaining a priority queue. In this paper, we consider an approach that attempts to approximate a sorted priority queue at an output-buffered switch. The goal is to trade off less accurate sorting for lower computational overhead. Specifically, this paper presents a scheduler that approximates the sorted queue of an earliest-deadline-first (EDF) scheduler. The approximate scheduler is implemented using a set of prioritized first-in/first-out (FIFO) queues that are periodically relabeled. The scheduler can be efficiently implemented with a fixed number of pointer manipulations, thus enabling an implementation in hardware. Necessary and sufficient conditions for the worst-case delays of the scheduler with approximate sorting are presented. Numerical examples, including traces based on MPEG video, demonstrate that in realistic scenarios, scheduling with approximate sorting is a viable option.

A background traffic model is fundamental to packet-level network simulation since the background traffic impacts packet drop rates, queueing delays, end-to-end delay variation, and also determines available network bandwidth. In this paper, we present a statistical characterization of wide-area IP traffic based on 90-minute traces taken from a week-long trace of packets exchanged between a large campus network a state wide educational network, and a large Internet service provider. The results of this analysis can be used to provide a basis for modelling background load in simulations of wide-area packet-switched networks such as the Internet, contribute to understanding the fractal behavior of wide-area network utilization, and provide a benchmark to evaluate the accuracy of existing traffic models. The key findings of our study include the following: (1) both the aggregate packet stream and its component substreams exhibit significant long-range dependencies in agreement with other traffic studies. (2) the empirical probability distributions of packet arrivals are log-normally distributed. (3) packet sizes exhibit only short-term correlations and (4) the packet size distribution and correlation structure are independent from both network utilization and time of day.

A packet scheduler in a quality-of-service (QoS) network should be sophisticated enough to support stringent QoS constraints at high loads, but it must also have a simple implementation so that packets can be processed at the speed of the transmission link. The earliest-deadline-first (EDF) scheduler is the optimal scheduler for bounded-delay services in the sense that it provides the tightest delay guarantees of any scheduler, but an implementation of EDF requires the sorting of packets, a complex operation that is not practical for high-speed networks. In this study we present the design, implementation and analysis of the novel rotating-priority-queues/sup +/ (RPQ/sup +/) scheduler that is near-optimal in the sense that it can approximate EDF with arbitrary precision. The RPQ/sup +/ scheduler uses a set of prioritized FIFO queues whose priorities are rearranged (rotated) periodically to increase the priority of waiting packets. We derive admission control tests for RPQ/sup +/ and show that it has the following desirable properties: its implementation requires operations independent of the number of queued packets, it can provide worst-case delay guarantees, and its efficiency is "between" that of EDF and static-priority (SP) schedulers. We use numerical examples, including examples based on MPEG video, to show that in realistic scenarios RPQ/sup +/ can closely approximate EDF even for infrequent queue rotations.

One of the most important traffic types in future packet-switched networks is high-bandwidth, variable-bit-rate (VBR) video. Since video is a delay-sensitive media, the network must allocate resources to maintain quality-of-service (QoS) guarantees on throughput, delay, and delay jitter to video connections. A key component of resource allocation is the traffic characterization of video sources that determines the resources required to support video connections. In this study, we propose a method for characterizing VBR video traffic with a fixed number of leaky buckets in networks with a deterministic service. We explore tradeoffs of network utilization in two directions: (1) the number of leaky buckets used for traffic characterisation, and (2) the amount of information from a video sequence used to produce the characterization. We evaluate our method with a set of 30-minute long MPEG-compressed video traces.

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.

Compressed digital video is one of the most important traffic types in future integrated services networks. However, a network service that supports delay-sensitive video imposes many problems since compressed video sources are variable bit rate (VBR) with a high degree of burstiness. In this paper, we consider a network service that can provide deterministic guarantees on the minimum throughput and the maximum delay of VBR video traffic. A common belief is that due to the burstiness of VBR traffic, such a service will not be efficient and will necessarily result in low network utilization. We investigate the fundamental limits and trade-offs in providing deterministic performance guarantees to video and use a set of 10 to 30 min. long MPEG-compressed video traces for evaluation. Contrary to conventional wisdom, we are able to show that, in many cases, a deterministic service can be provided to video traffic while maintaining a reasonable level of network utilization. We first consider an ideal network environment that employs the most accurate deterministic, time-invariant video traffic characterizations, the optimal earliest-deadline-first packet schedulers, and exact admission control conditions. The utilization achievable in this situation provides the fundamental limits of a deterministic service. We then investigate the utilization limits in a network environment that takes into account practical constraints, such as the need for simple and efficient policing mechanisms, packet scheduling algorithms, and admission control tests.

A novel packet multiplexing technique, called rotating-priority-queues (RPQ), is presented which exploits the tradeoff between high efficiency, i.e., the ability to support many connections with delay bounds, and low complexity. The operations required by the RPQ multiplexer are similar to those of the simple, but inefficient, static-priority (SP) multiplexer. The overhead of RPQ, as compared to SP, consists of a periodic rearrangement (rotation) of the priority queues. It is shown that queue rotations can be implemented by updating a set of pointers. The efficiency of RPQ can be made arbitrarily close to the highly efficient, yet complex, earliest-deadline-first (EDF) multiplexer. Exact expressions for the worst case delays in an RPQ multiplexer are presented and compared to expressions for an EDF multiplexer.

Compressed digital video is one of the most important traffic types in future integrated services networks. However, a network service that supports delay-sensitive video imposes many problems since compressed video sources are variable bit rate (VBR) with a high degree of burstiness. In this paper, we consider a network service that can provide deterministic guarantees on the minimum throughput and the maximum delay of VBR video traffic. A common belief is that due to the burstiness of VBR traffic, such a service will not be efficient and will necessarily result in low network utilization. We investigate the fundamental limits and tradeoffs in providing deterministic performance guarantees to video and use a set of 10 to 90 minute long MPEG-compressed video traces for evaluation. Contrary to conventional wisdom, we are able to show that, in many cases, a deterministic service can be provided to video traffic while maintaining a reasonable level of network utilization. We first consider an ideal network environment that employs the most accurate deterministic, time-invariant video traffic characterizations, Earliest-Deadline-First packet schedulers, and exact admission control conditions. The utilization achievable in this situation provides the fundamental limits of a deterministic service. We then investigate the utilization limits in a network environment that takes into account practical constraints, such as the need for fast policing mechanisms, simple packet scheduling algorithms, and efficient admission control tests.

Presents a new strategy for performance-driven global routing. This strategy focuses on the creation of spanning trees whose properties are under the control of the designer. The authors use it to construct a spanning tree with simultaneous, provable performance guarantees on total length, single-source shortest path length, and bottleneck path length. For rectilinear problems on n terminals in the plane, such a tree can be constructed in O(n log n) time.>