Showing papers on "Network traffic simulation published in 1989"

Estimation of reliability for communication/computer networks simulation/analytic approach

Abstract: A general simulation/analytic approach for jointly assessing the reliability and performance of a communication/computer network is proposed. The advantages of this approach are the following: (a) it allows consideration of a variety of performance measures, (b) it allows consideration of a general structure of network components failure distribution, (c) it reduces complexity of the model and related calculations, (d) it is relatively simple to implement, and (e) it can derive sensitivity information about network performance with respect to various parameters-the first step for an effective network optimization algorithm. The method is illustrated in a number of examples using network connectivity and weighted network bandwidth as a measure of performance. >

ATM communications network control by neural network

Abstract: A learning control method using neural networks for service quality control in the asynchronous transfer mode (ATM) network is described. An ATM network is a high-speed packet-switching network for the data transmission layer of B-ISDN (broadband integrated services digital network) which provides multimedia services, including voice, data and video. Service quality control is one of the most crucial issues in realizing a flexible ATM network. It is a challenging research task to build an efficient network controller that can control the network traffic even when the precise characteristics of the source traffic are not known and the service quality requirements change over time. The proposed ATM network controller is flexible in function and simple in implementation because neural networks using backpropagation learn the relations between the offered traffic and service qualities. A training data selection method called leaky pattern tables is proposed for learning the accurate relations. The performance of the proposed controller is evaluated by simulation of a basic call regulation model. >

Management of sampled real-time network measurements

Abstract: An overview is presented of the statistical sampling of measurements from a computer network, a necessary task in all five facilities of the proposed ISO network management architecture. Network measurement, viewed from the standpoint of real-time status reporting, is performed either to approximate the condition of a network parameter over a given interval or to detect a significant change in system behavior over consecutive intervals. Appropriate statistical tools for sampling and analyzing traffic in real time are presented. >

M-architecture: a structural model of traffic management and control in broadband ISDNs

Abstract: A model of multiservice network architecture (M-architecture) is presented. An attempt is made to describe the traffic management and control schemes in a systematic way. The basic concepts are developed, and examples are given that indicate how the proposed framework can be used to solve multiplexing, routing, and flow control problems in the ISDN. The emphasis is on the traffic control schemes. The M-architecture can be regarded as either a reference architecture for future multiservice networks or an abstract model of traffic management and control in currently used networks. >

Development and potential of system optimized route guidance strategies

Abstract: The authors examine the traffic engineering potential and limitations of routeing strategies that are system optimized, as opposed to routeings that minimize the travel times of individual drivers This examination is based on a simple two-route traffic network The user- and system-optimum traffic assignments are illustrated for this network, and the sensitivity and magnitude of any differences between the two are examined An incremental traffic assignment technique that can find system-optimized routes in a traffic engineering context is introduced This can be utilized within existing traffic models or directly in a route guidance algorithm, where vehicles are assigned to the network as packets (increments) of size 1 It is concluded that there are potential savings in total network travel time if system- vs user-optimized routeing strategies are considered >

Simulation For Decision Making. An Introduction

Abstract: We give an overview of simulation modeling and analysis from the perspective of prospective users wanting to use simulation as a decision aid. Important considerations in building simulation models and analyzing their outputs are discussed. A brief overview of available software is given. At the end of this tutorial, you should have a general understanding of simulation and an understanding of its applicability to your situation.

Kalman filtering of traffic fluctuations for real-time network management

Abstract: Discrete- time stochastic models of sequential traffic records in telephone networks are considered. They appear to be well suited for a use of the Kaiman filter. A version of the filtering algorithm is adopted to deal with traffic fluctuations. The overall approach is tested against the real data gathered by the supervisor of the French toll network. The adaptive traffic estimator is inserted into a simulator of a testbed network operated in the urban zone of Paris. The traffic estimations provided by the filter are used to decide an overflow hierarchy.

Performance study of a tree LAN with collision avoidance

Abstract: The design is presented of a broadcast star network that uses a collision-avoidance switch. This network combines the benefits of random access (low delay when traffic is light; simple, distributed, and therefore robust protocols) with excellent network utilization. Performance analyses of a slotted broadcast star network are then presented to obtain distributions of the transmission delay: an exact analysis for a network with an infinite number of stations, and an approximate analysis for a network with a finite number of stations. >

Performance evaluation of adaptive traffic routing in a metropolitan network: a case study

Abstract: A performance analysis of adaptive traffic in the telephone network of Marseilles, France is presented. The authors give the main reasons for introducing flexible routing-especially, state-dependent routing-in modern telephone networks. They present the Marseilles network under study and briefly describe the routing algorithm and the network design method used. They consider a number of scenarios ranging from traffic overloads to network failures in order to evaluate the robustness and performance of the adaptively routed network. Simulations that emphasize the role of state-dependent routing in network efficiency are presented and analyzed. In some specific cases, the state-dependent routing is complemented with predictive control actions to limit low priority or overflow traffic. This results in an improved grade of service. It is concluded that adaptive traffic routing can cope with a large range of network conditions, such as traffic overloads and equipment failures. >

Traffic placement policies for multi-band network

Abstract: Recently protocols have been introduced that enable the integration of synchronous traffic (voice or video) and asynchronous traffic (data) and extend the size of local area networks without loss in speed or capacity. One of these is DRAMA, a multiband protocol based on broadband technology. It provides dynamic allocation of bandwidth among clusters of nodes in the total network. In this paper, we propose and evaluate a number of traffic placement policies for such networks. Metrics used for performance evaluation include average network access delay, degree of fairness of access among the nodes, and network throughput. The feasibility of the DRAMA protocol is established through simulation studies. DRAMA provides effective integration of synchronous and asynchronous traffic due to its ability to separate traffic types. Under the suggested traffic placement policies, the DRAMA protocol is shown to handle diverse loads, mixes of traffic types, and numbers of nodes, as well as modifications to the network structure and momentary traffic overloads.

The characterisation of network applications-parameters for artificial traffic generation

Abstract: The primary concern of the paper is the characterisation of traffic on a network which will lead to the development of a model of that traffic. This can be achieved by analysing the results from measurements of network traffic. Of especial interest is the traffic generated by each particular application using the network. The authors have investigated the traffic patterns produced from the use of an electronic mail system. This, together with results from other studies has led to the proposal of a set of criteria which characterise traffic on a per application basis.

Parallel simulation: practice: contrasting distributed simulation with parallel replication: a case study of a queuing simulation on a network of transputers

Abstract: As discrete event simulation programs become larger and more complex, the amount of computing power required for their execution is rapidly increasing. One way to achieve this power is by a employing a multiple processor network to run the simulation programs.Two approaches to the problem of assigning tasks to processors are described-environment partitioning distributed simulation, in which the tasks required to perform a simulation are assigned to processors in the network; and parallel replication, in which copies of the simulation program are assigned to processors, and the results of their execution aggregated. A simulation of an M/M/c queuing system has been implemented on networks of two and three transputers, using each approach. Heidelberger's statistical efficiency and the stabilization time of the system are used as metrics. The parallel replications tended to stabilize faster, but the statistical efficiencies were not significantly different.

Application of the product graph to a large multiexchange digital network

Abstract: The main properties of a network topology based on the Cartesian product of two complete graphs are briefly reviewed and the problems involved in the application of this type of structure to a multiexchange network are analysed. An approximate solution to the problem of assigning nodes of the network to exchanges (whose traffic matrix is known) is presented. This is obtained from a heuristic algorithm which involves solving a number of linear programming problems. Finally, a computational method for modelling a multiexchange digital network is applied to a network of very high traffic capacity such as might be the case for the top level of a digital national telecommunications network.

Contrasting Distributed Simulation With Parallel Replication: A Case Study Of A Queuing Simulation On A Network Of Transputers

Abstract: As discrete event simulation programs become larger and more complex, the amount of computing power required for their execution is rapidly increasing. One way to achieve this power is by a employing a multiple processor network to run the simulation programs. Two approaches to the problem of assigning tasks to processors are described--environment partitioning distributed simulation, in which the tasks required to perform a simulation are assigned to processors in the network; and parallel replication, in which copies of the simulation program are assigned to processors and the results of their execution aggregated. A simulation of an M/M/c queuing system has been implemented on networks of two and three transputers, using each approach. Heidelberger's statistical efficiency and the stabilization time of the system are used as metrics. The parallel replications tended to stabilize faster, but the statistical efficiencies were not significantly different.

Bounding delay on a token ring network with voice, data and facsimile applications: a simulation study

Abstract: A discussion is presented of the design issues in implementing a document retrieval system on a token-ring network. The presence of real-time traffic, namely, voice, on the same network makes the implementation complicated. The performance requirements of these traffic types are different. Voice creates stream traffic on a network, whereas data traffic is bursty. Voice packets need to be delivered within a limited time interval, whereas the data need emphasizes error-free delivery. A possible solution, using off-the-shelf components and supported by a simulation study, is discussed. >

Performance evaluation of a multichannel token bus local area network architecture

Abstract: A token-based protocol for multichannel local area network architectures is presented. The operation of the multichannel protocol is described, and its performance is evaluated by simulation. The simulation model is briefly described. Through simulation the behavior of the network is analyzed for varying network parameters and traffic load distributions. Discussions on the multichannel protocol's merits relative to the single channel protocol are provided. It is shown that the protocol offers improved throughput-delay performance with respect to the single channel token bus access protocol and that it also provides for high reliability and simpler network maintenance. >

Some issues related to simulation of the tracking and communications computer network

Abstract: The Communications Performance and Integration branch of the Tracking and Communications Division has an ongoing involvement in the simulation of its flight hardware for Space Station Freedom. Specifically, the communication process between central processor(s) and orbital replaceable units (ORU's) is simulated with varying degrees of fidelity. The results of investigations into three aspects of this simulation effort are given. The most general area involves the use of computer assisted software engineering (CASE) tools for this particular simulation. The second area of interest is simulation methods for systems of mixed hardware and software. The final area investigated is the application of simulation methods to one of the proposed computer network protocols for space station, specifically IEEE 802.4.

Nearly Poisson transfer traffic in real-time networks

Abstract: Consideration is given to a homogeneous distributed network of nodes with a quasi-dynamic random task-sharing discipline and identical Poisson hard real-time traffic at each node. The task-sharing discipline under consideration permits tasks to be shared randomly around the network as long as remaining task laxity permits. Analysis of task loss in such a network requires the assumption that the transferred task stream from all nodes form a Poisson stream. The authors study the requirements for such an assumption to be justified by using computer simulation to examine transferred traffic for both independence and its fit to an exponential distribution as a function of network size. A rough rule of thumb for estimating independence is to assume the transfers are independent after the system size exceeds the average number of transfers between nodes. >