Showing papers on "Network traffic simulation published in 1993"

On the self-similar nature of Ethernet traffic

Abstract: 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.

Application of sampling methodologies to network traffic characterization

Abstract: The relative performance of different data collection methods in the assessment of various traffic parameters is significant when the amount of data generated by a complete trace of a traffic interval is computationally overwhelming, and even capturing summary statistics for all traffic is impractical. This paper presents a study of the performance of various methods of sampling in answering questions related to wide area network traffic characterization. Using a packet trace from a network environment that aggregates traffic from a large number of sources, we simulate various sampling approaches, including time-driven and event-driven methods, with both random and deterministic selection patterns, at a variety of granularities. Using several metrics which indicate the similarity between two distributions, we then compare the sampled traces to the parent population. Our results revealed that the time-triggered techniques did not perform as well as the packet-triggered ones. Furthermore, the performance differences within each class (packet-based or time-based techniques) are small.

Method and apparatus for planning a cellular radio network by creating a model on a digital map adding properties and optimizing parameters, based on statistical simulation results

Abstract: A method and a system for planning a cellular radio network. The method includes creation of a model representing the cellular radio network and its radio environment on a digital map. The method further includes adding system properties affecting the traffic control process of the cellular radio network to the model representing the cellular radio network and its radio environment for route specific operational simulation of the cellular radio network. The simulation preferably utilizes subscriber mobility models and immobile and/or mobile individual subscribers generated on the digital map. The simulation events are stored separately for each street, location, network element and/or cause. The optimization of the parameters of the cellular network is performed on the basis of the stored information. Simulation of an operating cellular network on the basis of statistical data obtained from the network allows an adaptive control of the system parameters.

Method simulating data traffic on network in accordance with a client/server paradigm

Abstract: A method for modeling traffic on a network according to a Client/Server paradigm is disclosed. The method includes the steps of exctracting the model parameters from known or measured interactions between a client node running a particular application and a server node with no other activity on the network. The method is repeated for each nodal configuration and for each application until the traffic for each node-application combination has, been modeled with no other load on the network. Scripts are created from the Paradigm parameters and are used in conjunction with a traffic generator to create discreet events mimicking the traffic which would be generated from each node. The model will insure that the traffic generated by each traffic source will be adjusted as a function of the response of the network under the load created by all of the traffic loading the network.

Vehicle probes as real-time atms sources of dynamic o-d and travel time data

Abstract: This paper has two main objectives. First, the process of obtaining network traffic data from probe vehicles is examined such that statistical sampling equations can be developed that provide both theoretical and practical estimates of the reliability of these probe data. Second, using an example network, the traffic simulation model results are compared against the analytical results. The validation of simulation based results permits the simulation model to be used for more complex situations, for which statistical analyses are no longer practical, and to determine the overall impact of the various statistical approximations that had to be made within the analytical solutions.

Modeling and distributed simulation of a broadband-ISDN network

Abstract: A distributed approach to communication network simulation using a network of workstations configured as a loosely coupled parallel processor to model and simulate the broadband integrated services digital network (B-ISDN) is proposed. In a loosely coupled parallel processor system, a number of concurrently executable processors communicate asynchronously using explicit messages over high-speed links. Since this architecture is similar to that of B-ISDN networks, it constitutes a realistic testbed for their modeling and simulation. The authors describe an implementation of this approach on 50 Sun workstations at Brown University. Performance results, based on representative B-ISDN networks and realistic traffic models, indicate that the distributed approach is efficient and accurate. >

Parallel communications for ATM network control and management

Abstract: This paper describes an end-do-end parallel communications scheme for ATM network control and management. An information string is partitioned into m parts, which are then coded into k>m parts and sent out on k separate subchannels to the receiver As soon as m of the L part are received, the original information can be reconstructed. Two desirable effects are shown to be achievable: (1) reduction of the burstiness of the source traffic; and (2) controllability of the quality of service. Network control and management using the framework are discussed. >

Performance Study of an ATM Self‐Routing Multistage Switch with Bursty Traffic: Simulation and Analytic Approximation

Abstract: An N x N self-routing multistage switching network with bursty and uniform traffic, which is composed of S x 5 switching elements with output queueing and has K = logS N stages, is considered in this paper. The input traffic is modelled by a two-state (active/silent) Markov chain. Each arriving cell selects one of the N outlets as its destination independently and with equal probability I/N. The performance of the switching network is studied by means of a simulation method. An approximate traffic model of a four-state Markov chain is proposed to characterize the cell arrival processes to further stages and its related switching element performance is analyzed. Comparison with the simulation results shows that the proposed traffic model gives a good approximation, which implies that the performance of the switching element mainly depends on the first two moments of the active and silent periods and is not sensitive to the shapes of their distributions. The influence of bursty traffic on the performance of the switching network is discussed and illustrated by some numerical examples. The impact of traffic randomization on the performance of the switching network is also investigated.

An enroute ATC simulation experiment for sector capacity estimation

Abstract: A series of enroute ATC simulation experiments was conducted to estimate the airspace capacity of existing sectors in Japan and to evaluate a new route structure for accommodating the traffic increase in the future. A simulation method for capacity estimation and the results of the experiments focusing on the controller's workload limit are presented. From the results of regression analysis, the relationship between the airspace capacity and the controller's intervention in the traffic flow is discussed. About 40% increase of capacity is attained with an introduction of parallel routes and traffic flow management. >

A Simulation-based Framework for the Analysis of Traffic Networks Operating with Real-time Information

Abstract: The focus of this research project was to develop a new simulation framework for analyzing traffic networks with Advanced Traveler Information Systems (ATIS) and/or Advanced Traffic Management Systems ( ATMS). The report describes the development of DYNASMART (DYnamic Network Assignment-Simulation Model for Advanced Road Telematics). The report is presented as follows: Chapter 1 discusses the simulation approach of DYNASMART and explains the traffic control features incorporated in it during the research. Chapter 2 discuss the simulations performed with a trial network with ATMS controls. Chapter 3 concentrates on the ATIS simulations on the Anaheim network. Results from the simulations of traffic management for special-events traffic from the Anaheim stadium are presented in Chapter 4, followed by overall conclusions.

Pseudo Aircraft Systems - A multi-aircraft simulation system for air traffic control research

Abstract: Pseudo Aircraft Systems (PAS) is a computerized flight dynamics and piloting system designed to provide a high fidelity multi-aircraft real-time simulation environment to support Air Traffic Control research. PAS is composed of three major software components that run on a network of computer workstations. Functionality is distributed among these components to allow the system to execute fast enough to support real-time operation. PAS workstations are linked by an Ethernet Local Area Network, and standard UNIX socket protocol is used for data transfer. Each component of PAS is controlled and operated using a custom designed Graphical User Interface. Each of these is composed of multiple windows, and many of the windows and sub-windows are used in several of the components. Aircraft models and piloting logic are sophisticated and realistic and provide complex maneuvering and navigational capabilities. PAS will continually be enhanced with new features and improved capabilities to support ongoing and future Air Traffic Control system development.

SCOUT: an approach to automating diagnoses of faults in large scale networks

Abstract: There has been some success in automating the diagnosis of transmission problems in large telecommunications networks such as AT&T's digital communication network. This paper describes a new approach to automatic diagnosis that uses network topology and network behavior history to automate the troubleshooting of transmission problems in any large telecommunications network. We have developed and deployed a system based on this approach. This system is currently being used by technicians to troubleshoot problems in AT&T's network and it has been found to be very effective. >

User-friendly simulation model for traffic networks with ATIS/ATMS

Abstract: Implementation of ATIS (Advanced Traffic Information Systems) and/or ATMS (Advanced Traffic Management Systems) are proceeding in many cities around the world, but only very few frameworks are available to study the effectiveness of these. Such systems are complicated thanks to complex interactions among the dynamics of path-based traffic conditions in the network, information/guidance supply strategies and the driver response in the form of their route-choice decisions. In view of the analytical intractability of these component phenomena, simulation frameworks could well be the only reasonable solution. DYNASMART (Dynamic Network Assignment Simulation Model for Advanced Road Telematics) is a program that we developed earlier that captures these key components in a flexible simulation framework that models individual vehicle movements using macroscopic traffic relationships and the driver decisions based on information at the network nodes. In this paper, we report on the recent enhancements that we have made to this model to make it more user-friendly, as well as the additional capabilities that we have incorporated to it.

ILS: A system of learning distributed heterogeneous agents for network traffic management

Abstract: Integrated Learning System (ILS) is a framework for integrating several distributed heterogeneous learning agents that cooperate to improve problem-solving performance. The agents learn both independently and cooperatively. Each agent has been tested independently using telecommunications traffic control as a domain, and each has demonstrated that it can learn by interacting with that domain. At present, ILS is being extensively tested using the same domain, and the initial results are very promising. >

A simulation model of dynamic traffic flow in networks

Abstract: In this paper, a dynamic model for time dependent traffic flow in networks with multiple sources and sinks is presented. The main goals of the model are a real-time simulation of larger transportation networks and a planning tool for traffic networks.

A method for automatic evaluation of fault effects in the advanced intelligent network

Abstract: The method described here provides network operators with criteria for deciding the priorities with which services degraded by network faults should be restored. This method consists of four processes: the first lists the unavailable services of customers affected by the fault, the second predicts the mean time needed to repair the fault, the third predicts the traffic trends for the affected services, and the fourth calculates the criteria used to decide the service restoration priorities. This method enables network operators to choose suitable means restoring services and to avoid future congestion due to faults. It also lets them respond more quickly and precisely to customer claims. We are introducing this method into our advanced IN operations systems. >

A short-term demand forecasting model from real-time traffic data

Abstract: Developing real-time traffic diversion strategies is a major issue of Advance Traffic Management Systems (ATMS), a component of Intelligent Vehicle Highway Systems(IVHS). Traffic diversions attempt to maximize the use of available capacities in the roadway systems during congestion-causing events. In selecting the diversion route, the decision is based that the route may not become congested by the time the diverted drivers reach that part of the network. Thus the forecasting of expected traffic flow on various parts of the network in a prompt and accurate fashion would help determine the efficient alternate routes. In this research, a prediction model has been developed to provide traffic forecasts. It has two components. One component is an Auto Regressive Integrated Moving Average (ARIMA) model based on the previous traffic data. The other component is the average traffic flow for that period as obtained from previous days. These two components are combined to represent the fluctuations in the traffic flow behaviour.

Simulating network traffic flows with a massively parallel computing architecture

Abstract: This paper presents a traffic network simulation model for real-time applications in IVHS. The proposed model has incorporated three key features essential for IVHS operations: (1) the capability of simulating both freeways and surface street networks as an integrated network; (2) a path-processing capability for representing drivers' route choice behavior at an individual/vehicle level; and (3) the capability of simulating different subnetworks at different levels of detail so as to increase the execution speed for real-time operations. Three simulation methodologies are implemented in the real-time traffic simulation model to satisfy the above requirements. These three methodologies are macroparticle traffic simulation model (MPSM), modified NIPSM (M-MPSM), and microscopic (MICRO) model. Due to the real-time operating requirement, the proposed simulation model has been parallelized to take advantage of the parallel computers since they can offer the required computational power at an economical cost/performance ratio. Several simulation experiments have been carried out to compare the execution speed of each methodology. The preliminary research results indicate that parallel computing architecture offers a very promising alternative for the implementation of real-time traffic simulation.

Method for modeling and simulating data traffic on networks.

Abstract: A method for modeling traffic on a network according to a Client/Server paradigm is disclosed. The method includes the steps of exctracting the model parameters from known or measured interactions between a client node running a particular application and a server node with no other activity on the network. The method is repeated for each nodal configuration and for each application until the traffic for each node-application combination has been modeled with no other load on the network. Scripts are created from the Paradigm parameters and are used in conjunction with a traffic generator to create discreet events mimicking the traffic which would be generated from each node. The model will insure that the traffic generated by each traffic source will be adjusted as a function of the response of the network under the load created by all of the traffic loading the network.

Efficient Simulation of DEDS by Means of Standard Clock Techniques: Queueing and Integrated Radio Network Examples

Abstract: : In this report we discuss the application of the Standard Clock (SC) approach to the simulation of several examples of discrete event dynamic systems (DEDS), and demonstrate the improved efficiency that can be achieved by using this approach. In SC simulation, a common clock (event time, event type) sequence is used by a large number of simulations running in parallel, thereby resulting in a significant decrease in the most costly aspect of the simulation process, namely the generation of events. In studies of the M/M/1/K queue, we quantify the time spent in the various aspects of the simulation, and compare predicted performance improvement with that which is actually measured. This approach scales well when applied to larger problems, including queueing networks and multihop radio networks. We develop a SC simulation model for a multihop, integrated voice/data radio network, and demonstrate its use for the determination of admission control policies. Ordinal optimization techniques are shown to provide good, although not optimal, control policies on the basis of short simulation runs. Simulation, Voice/data integration, Discrete event dynamic systems, Standard clock, Communications network

Traffic congestion analysis of the public switched network

Abstract: This paper presents a model for performing traffic congestion analyses of the entire public switched-voice network (PSN) in periods of network stress The model considers the assets of the local exchange carriers and the three major interexchange carriers - AT&T, MCI, and Sprint These assets include over 20,000 telecommunications switches interconnected by approximately 50,000 trunk groups Off-the-shelf network analysis packages and call-by-call simulators cannot model such a network within practical computing limits The TAMI model (Traffic Analysis by Method of Iteration) was developed to address this need It was designed specifically for the size and structure of the PSN Central to the model is the way it uses nested iterations of the traffic loading process to determine steady state convergence of trunk group blockages and switch overloads These results are combined to measure the performance of the network with a single parameter-traffic-weighted average end office-to-end office blockage TAMI was designed as a crisis analysis tool It readily accepts different network stress scenarios It models special handling of limited amounts of priority traffic Glare and switch matching loss are considered, as well as the effects of network management controls >

Estimation of queue lengths in urban road networks

Abstract: In this paper we present a new scheme for the surveillance of traffic state on a link of an urban road network. It is shown that measurements of traffic counts and speed values from a single detector may produce reliable estimates if the detector is placed at an appropriate distance from the traffic signal. Traffic state estimation then is performed using the Extended Kalman Filter Which uses a dynamic model for link flows. The high of this scheme is underlined by some simulation results.

The self-traffic-engineering network

Abstract: Characterizes several techniques for the distributed, dynamic management of telecommunications network transmission facilities. Several methods are presented for changing trunk-group sizes in order to respond to extraordinary network loads. These methods are given the generic title of self-traffic-engineering (STE), because the network effectively reengineers itself to adapt to unforeseen circumstances. Self-traffic-engineering is a new network management application for telecommunication facilities networks. One of these methods is based on a new interpretation of network reliability, which is embodied in a metric called connectability. The paper reports the results of a simulation study in which self-traffic-engineering is used to enable the Telecom Canada network to adapt to a focused overload.

Data-driven simulation of networks with manufacturing blocking

Abstract: This paper presents a data-driven simulation of a network using manufacturing blocking to control work-in-process, and this simulation is c2aled BNS for Blocking Network Simulation. The model is object oriented and is constructed in C++ using an object-oriented simulation environment called HOSE. The structure of the data- driven network simulator is illustrated with examples of the underlying code. An important feature of the model is a sequential procedure for determining run length to achieve statistical precision in an output performance measure. An illustrative example compares the performance of BNS with a queueing network approximation. The principal conclusion is that the data-driven network simulation can be applied by users with minimal simulation knowledge as a rapid modeling tool.

Traffic Control and Resource Management

Abstract: The primary role of traffic control and resource management procedures is to protect the network so that it can achieve the required network performance objectives. The uncertainties of broadband traffic patterns and the complexity of resource management suggest a step-wise approach for defining these parameters and procedures. An initial set of traffic-control and resource-management capabilities is currently defined by the CCITT in Recommendation 1.371. Further sets of capabilities may subsequently be defined to achieve increased network efficiency. The following four functions are considered: network resource management, connection admission control, usage parameter control and network parameter control. Their location in an ATM network is also discussed.

Experience in massively parallel discrete event simulation

Abstract: Discrete event simulation of complex systems is one of the most important problems in scientific and industrial computing. It is not uncommon for a single simulation run to take hours on a high speed workstation. Massively parallel computers provide opportunities for simulating complex systems significantly faster. At the heart of the computational difficulties in parallel simulation are (i) random, datadependent events that couple subsystems, and (ii) asymmetries wherein some subsystems receive a disproportionate number of events, leading to severe load imbalances. In this paper, we discuss a parallel simulation of a complex system—a large asymmetric circuit-switched network, with state-dependent control—and methods we have applied to overcome both types of difficulties just mentioned. In a worst case sense, we show that the problem is hard: In particular, simulation of a network with just three nodes is P-complete. However, we show that a general purpose synchronous relaxation method uncovers the parallelism inherent in practical, realistic cases. In addition, we show that special purpose (specific to circuit-switched simulations) parallel methods allow us to profitably assign more processors to subsystems receiving more events, and thereby handle load imbalances. Promising performance results are reported for a 16384 processor MasPar MP1 implement ation, on a realistic 114 node network simulation scenario.

Concentrated traffic in non-hierarchical networks under alternate routing scheme: a behavior analysis

Abstract: Behavior characteristics of alternate routing schemes under homogeneous traffic in fully connected nonhierarchical networks have been investigated in the past. The fundamental behavioral characteristic of these networks has been shown to be that of a fold catastrophe. In this paper, we investigate the behavioral aspects when the traffic Intensity towards a node is much higher when compared with the traffic intensity towards other nodes. We call the high intensity traffic the concentrated traffic. In relation to the traffic intensity corresponding to the fold point under homogeneous traffic, the concentrated traffic intensity may be much higher while the intensity of traffic other than the concentrated one is much lower. For the behavioral characterization of the node under concentrated traffic, the tool of analysis employed in this paper is the same as that employed in the earlier behavioral analyses under uniform traffic: formulating the flow balance function, then obtaining the potential function from the flow balance function and finally finding the degeneracy of the potential function based on the results from catastrophe theory. The model presented in this paper is only for the node under the concentrated traffic. The analysis, though based on the average values of traffic parameters, yields satisfactory results when compared with the results obtained from simulation when the number of trunks per link is large. >