The dramatically increased bandwidths and processing capabilities of future high-speed networks make possible many distributed real-time applications, such as sensor-based applications and multimedia services. Since these applications will have traffic characteristics and performance requirements that differ dramatically from those of current data-oriented applications, new communication network architectures and protocols will be required. In this paper we discuss the performance requirements and traffic characteristics of various real-time applications, survey recent developments in the areas of network architecture and protocols for supporting real-time services, and develop frameworks in which these, and future, research efforts can be considered.

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 conside...

Measurement-based traffic characterization has come to acquire a great deal of importance in high-speed networks. In this paper, we segregate the traffic behaviour into the macrodynamics and microdynamics depending on the time scales at which the process is observed. We examine the validity of the Markovian assumption which is commonly made for modeling the macrodynamics of correlated traffic in network analysis. A fundamental issue in traffic modeling is whether the Markovian assumption has any significance on the queueing solutions. Our study compares the queueing solutions obtained using traffic models with very different underlying structure viz. Markovian vs non-Markovian which are identical only in their second-order and steady-state statistics. Our study suggests that higher-order traffic statistics are generally unimportant to queueing solutions. In essence, for a certain class of stationary stochastic processes, the Markovian assumption can be made in traffic modeling to simpl...

In this paper we study source modeling, queueing analysis and bandwidth allocation issues of VBR MPEG-2 video traffic in ATM networks. We propose a mini-source based discrete-time Markov modulated deterministic process (D-MMDP) to model the macro-frame smoothed VBR MPEG-2 traffic. It is shown that such a modeling technique is sufficiently accurate and simplies the involved queueing analysis for relvent system performance measures. It also provides a simple and accurate solution to the bandwidth allocation problem for VBR MPEG-2 video traffic. Keywords: Source Model, MPEG-2, Traffic Smoothing, Queueing Analysis, Bandwidth Allocation. - 2 - 1. Introduction Digital video communication is expected to be one of the major services supported by future broadband integrated services digital networks (B-ISDNs). Depending on the specific video codec, video packets may be transmitted in constant bit rate (CBR) or in variable bit rate (VBR) Major tradeoffs between CBR video and VBR video are th...

The design and analysis of robust networking protocols that offer useful performance guarantees requires accurate traffic source models. In this paper we study the problem of characterizing and modeling the arrival process of compressed video. We extend earlier works in this area by including a factor that has previously been ignored, the effect of video capture rate on traffic characterization. Dynamic changes in available bandwidth due to the addition and/or removal of connections can trigger re-negotiation of bandwidth between the applications and the network. Such re-negotiation may result in applications changing the capture rate of video sequences, thus effecting the traffic generation process. We show that for several popular video coding schemes, the bit rate distribution at the output of the encoder changes with the capture and compression rate. Using a combination of distributions, and exploiting knowledge of the underlying compression algorithms we characterize variable bit rate (VBR) video by application type, compression algorithm, and frame rate. We conclude that no single distribution can describe all video traffic, and as an alternate suggest a three dimensional matrix in which each dimension represents a different video classification aspect. Each entry in this matrix is a distribution type that best fits the given combination of the aspects. We use this result to show how the problem of network capacity planning may be tackled.

this paper were based on the 10Base5 version of the IEEE 802.3 specification, they will have at least limited applicability to the new 100 Mbps CSMA/CD standards being proposed by the 802.3 committee. There would have been three main differences to the simulations if the new 100 Mbps standards had been used instead of the 10 Mbps 10Base5 standard. First, all of the latency measurements would have been 1/10th what they were. This will decrease the cost of packet starvations when compared to human reaction time and video frame rates. Second, the traffic loads selected will not have as much direct applicability to these new LANs as to 10Base5 LANs, and so might need to be modified. Finally, the hub-based topology of these new LANs may produce different collision effects and costs than the straight-line bus topology used. Despite these differences, we expect the PSE to be a significant problem for real time traffic in these new networks. As real time traffic is a primary driving force behind the new standards, we feel that this is of high importance. To solve these problems, we proposed the Fair Dual Distributed Queue algorithm. This algorithm provides two priority FCFS scheduling of packets with a scheduling resolution bounded by the maximum time it takes to send a single packet. It provides utilization equal or better than that of CSMA/CD in all but the highest loads, which are impractical for CSMA/CD to sustain because of the PSE. It has comparable average latency to CSMA/CD and eliminates the unfairness of CSMA/CD. FDDQ is stable for a constant number of senders and gracefully handles even extremely high offered loads. This is a very powerful tool, for with the use of adaptive stream protocols (such as TCP/IP), a bus that utilizes FDDQ can be used at very close to its f...