Provides the results of an economic study on the use of SONET digital cross-connect systems (DCS) to provide survivable transmission network architectures in local exchange networks. Three fundamental survivable transmission technologies are considered: (1) a SONET self-healing ring, (2) a SONET point-to-point fiber system with 1:1 automatic protection switching and diverse routing of protection facilities, and (3) a DCS mesh with automatic DCS restoration (rerouting) protection. These three technologies are used in various combinations to form six survivable network alternatives for evaluation. Two local exchange carrier (LEC) networks are used (a 15 node network and a 53 node network) and demand, network connectivity, and unit equipment cost sensitivities are evaluated on these alternatives. In addition, the survivability of each alternative in the event of a major node failure is calculated. The motivation for the study is to determine the viability of DCS-based survivable network architectures and, in particular, the viability of SONET DCS with integrated optical terminations. The study has two objectives: (1) given a specific survivable network technology, under what conditions is it economical to place a broadband DCS (B-DCS) in a central office as opposed add-drop multiplexers (ADM); and (2) which survivable technologies with B-DCS are economical, and under what conditions. The authors conclude that the most cost-effective networks consist of "hybrids" of SONET point-to-point, ring, and mesh technologies, and that the B-DCS is economically viable for interconnection between these technologies.>

Demonstrates that Ethernet LAN traffic is statistically self-similar, that none of the commonly used traffic models is able to capture this fractal-like behavior, that such behavior has serious implications for the design, control, and analysis of high-speed, cell-based networks, and that aggregating streams of such traffic typically intensifies the self-similarity ("burstiness") instead of smoothing it. These conclusions are supported by a rigorous statistical analysis of hundreds of millions of high quality Ethernet traffic measurements collected between 1989 and 1992, coupled with a discussion of the underlying mathematical and statistical properties of self-similarity and their relationship with actual network behavior. The authors also present traffic models based on self-similar stochastic processes that provide simple, accurate, and realistic descriptions of traffic scenarios expected during B-ISDN deployment.>

We demonstrate that Ethernet local area network (LAN) traffic is statistically self-similar, that none of the commonly used traffic models is able to capture this fractal behavior, and that such behavior has serious implications for the design, control, and analysis of high-speed, cell-based networks. Intuitively, the critical characteristic of this self-similar traffic is that there is no natural length of a "burst": at every time scale ranging from a few milliseconds to minutes and hours, similar-looking traffic bursts are evident; we find that aggregating streams of such traffic typically intensifies the self-similarity ("burstiness") instead of smoothing it.Our conclusions are supported by a rigorous statistical analysis of hundreds of millions of high quality Ethernet traffic measurements collected between 1989 and 1992, coupled with a discussion of the underlying mathematical and statistical properties of self-similarity and their relationship with actual network behavior. We also consider some implications for congestion control in high-bandwidth networks and present traffic models based on self-similar stochastic processes that are simple, accurate, and realistic for aggregate traffic.

In order to study the interactions between LANs and proposed interconnection networks, a high time-resolution hardware monitor for Ethernet LANs is presented that avoids the shortcomings of previous monitoring tools, such as traffic burst clipping and timestamp jitter. Using data recorded by a monitor for several hundred million Ethernet packets, an overview of the short-range time correlations in external LAN traffic is presented. The analysis shows that LAN traffic is extremely bursty across time domains spanning six orders of magnitude. This behavior is compared with simple formal traffic models. The results suggest that the pronounced short-term traffic correlations, together with the extensive time regime of traffic burstiness, strongly influence the patterns of loss and delay induced by LAN interconnection.>

The authors examine the phenomenon of congestion to better understand the congestion management techniques that will be needed in high-speed, cell-based networks. The first step of this study is to use high time-resolution local area network (LAN) traffic data to explore the nature of LAN traffic variability. Then the data are used for a trace-driven simulation of a connectionless service that provides LAN interconnection. The simulation allows one to characterize what congestion might look like in a high-speed, cell-based network. When realistic data are applied to simple models of LAN interconnection, it is observed that during periods, congestion persists and losses can be significant; congestion losses cannot be avoided by modest increases in buffer capacity; consequences of misengineering can be serious; and, fortunately, most congested periods are preceded by signs of impending danger.>

Dynamic load balancing in a system of loosely-coupled homogeneous processors may employ both judicious initial placement of processes and migration of existing processes to processors with fewer resident processes. In order to predict the possible benefits of these dynamic assignment techniques, we analyzed the behavior (CPU, disk, and memory use) of 9.5 million Unix* processes during normal use. The observed process behavior was then used to drive simulation studies of particular dynamic assignment heuristics.Let F(·) be the probability distribution of the amount of CPU time used by an arbitrary process. In the environment studied we found: (1-F(x)) n rx-c, 1.05

This paper presents a new family of undirected graphs that allows N processors to be connected in a network of diameter 3/2 log2 N + O(1), while only requiring that each processor be connected to three neighbors. The best trivalent graphs previously proposed require a diameter of 2 log2N + O(1).