This letter presents an approximation method for calculating the blocking probabilities of loss systems with batch input. Using this method, a batch stream can be treated in the same manner as an overflow stream because both streams can be characterized by common parameters: the offered load and peakedness.

This paper presents a queueing network model for common memory access contention and access contention to common data that cannot be referenced by more than one processor in a multiprocessor-controlled switching system. The model takes into account the inter-dependence between these types of access contention, which is important for evaluating the performance of a single-bus multiprocessor system. Although this inter-dependence is incorporated into the queueing network model, an exact analysis is difficult. Therefore, an approximate analysis method is proposed based on decomposing the model into two submodels. Approximate analysis results are compared with simulation results and it is shown that the approximation method provides sufficient approximate values. This queuing network model is useful for estimating the call processing capacity of a multiprocessor-controlled switching system.

This paper reports an overload control method for the Intelligent Network (IN). The IN, which is being investigated as a future communication network, facilitates both rapid introduction of new services and easy modification of existing services. In the IN, the call processing functions and data needed to achieve IN services are distributed over several nodes. Therefore, traffic demand for the various services may cause varying patterns of node overloads. It is therefore important to develop effective overload control methods and to evaluate their characteristics. We propose an overload control method and evaluate its characteristics in comparison with other methods under various overload traffic patterns with a network simulator that models all nodes and their relationships in the IN. In particular, we focus on three aspects of overload control: how can high throughput be maintained, how can an overloaded node be stabilized, and how can fair access be guaranteed.

This letter compares an integration scheme with a segregation one for a loss system having two types of traffic. As a result of this comparison, the conditions under which the overall blocking probability of an integration scheme is lower than or higher than that of a segregation scheme are given.

This letter presents a method for analyzing single server queueing models having two classes of customers. This method, based on the conservation laws, makes it possible to analyze queueing models for various types of queueing disciplines.

This paper presents a new approach, based on the conservation law, for analyzing the following single server queueing system with two classes of customers. The arrival process of one class customers is a Poisson process, while that of the other is represented by a renewal process and therefore need not be a Poisson process. Both classes of customers have exponential service time distributions. These queueing models with two input streams are often encountered in communication systems. Utilizing the conservation law, it is shown that for various types of work-conserving queueing disciplines, the mean time in the system for each class of customers can be expressed in teams of the total workload in the system. When the total workload is known, these formulas can be used to obtain numerical solutions for the mean time in the system. The most important result in this approach is that the mean time in the system for each class of customers can be easily computed even when the service rates of both classes of customers are different. This is demonstrated by computing the mean time in the system using the derived formulas. The approximation formulas for the total workload are also presented based on the diffusion approximations, because it is difficult to obtain exact solutions for the total workload except for some special cases. Comparisons of exact and approximate results show that these approximation formulas provide good approximate values.

A partial buffer sharing scheme is proposed as loss-priority control for a finite buffer with batch inputs. A partial batch acceptance strategy is used for a batch arriving at a finite buffer. Customer loss probabilities for high- and low-priority customers are derived under this batch acceptance strategy, using a supplementary variable method that is a standard tool for queueing analysis. A comparison of the partial buffer sharing scheme and a system without loss-priority control is made in terms of admissible offered load.

A partial buffer sharing scheme is proposed as loss-priority control for a finite buffer with batch Poisson inputs under a whole batch acceptance rule. Customer and batch loss probabilities for high- and low-priority customers are derived under this batch acceptance rule using a supplementary variable method. A comparison of the partial buffer sharing scheme and a system without loss-priority control is made in terms of admissible offered load. Whole batch acceptance and partial batch acceptance rules are also compared in terms of admissible offered load.

Narrow-band ISDN services may experience nonstationary traffic conditions. Therefore, switch design should take account of these conditions. We propose new performance measures for switching systems and describe a traffic model, which is a mixture of stationary Poissonian traffic and momentarily focused traffic. On the basis of this model, performance measures are determined so as to satisfy grade of service requirements that are in effect during some short interval after the momentarily focused traffic enters the system. We also propose an overload control scheme that uses these new performance measures. Finally, we show practical and numerical examples for the performance measures and overload control scheme.

This paper presents a finite buffer M/G/1 queue with two classes of customers who are served by a combination of head-of-the-line priority and buffer reservation schemes. This combination gives each class of customers high or low priorities in terms of both delay and loss. The scheme is analyzed for the model in which one class of customers has high priorities over the other class of customers with respect to both delay and loss. First, steady-state joint probability distribution of the number of each class of customers in the buffer and remaining service time is derived by a supplementary variable method. Second, loss probability and mean waiting time for each class of customers are provided using this probability distribution. Finally, a combination of head-of-the-line priority and buffer reservation schemes is numerically compared with other buffer management schemes in terms of admissible offered load to show its effectiveness under differing QoS requirements.

This letter reports on an approximation of the mean waiting time in a finite buffer queue with delay priority and loss priority. Both priorities are controlled by head-of-the-line (HOL) priority scheduling and buffer reservation. The proposed approximation is based on the known results on a HOL-priority queue with infinite buffer and a finite buffer queue with FIFO scheduling and buffer reservation. The accuracy of the approximation is validated by comparing exact and approximate results. The approximation provides good estimates when the blocking probabilities at the buffer controlled by the buffer reservation are low.

This paper analyzes a finite buffer M/G/1 queue with two classes of customers who are served by a combination of head-of-the-line priority and push-out schemes. This combination gives each class of customers two different types of priorities with respect to both delay and loss. There are two models considered. The first one is that one class of customers has a higher priority over the other class with respect to both delay and loss; the second one is that one class has a higher priority with respect to loss and the other has high-priority with respect to delay. For both of these models, the joint probability distribution of the number of customers of both classes in the buffer is derived by a supplementary variable method. Using this probability distribution, we can easily calculate the loss probabilities of both classes, the mean waiting time for high-priority customers with respect to loss and the upper bound for mean waiting time for low-priority customers with respect to loss. Numerical examples demonstrate an effect of the combination of different types of priorities.

Modeling and performance analysis have played an important role in the economical design and efficient operation of switching systems, and is currently becoming more important because the switching systems should handle a wide range of traffic characteristics, meeting the grade of service requirements of each traffic type. Without these techniques we could no longer achieve economy and efficiency of the switching systems in complex traffic characteristic environments. From the beginning of research on electronic switching systems offering circuit-switched applications, Stored Program Control (SPC) technology has posed challenges in the area of modeling and performance analysis as well as queueing structure, efficient scheduling, and overload control strategy design. Not only teletraffic engineers and performance analysts, but also queueing theorists have been attracted to this new field, and intensive research activities, both in theory and in practice, have continued over the past two decades, now evolving to even a broader technical field including traditional performance analysis. This article reviews a number of important issues that have been raised and solved, and whose solutions have been reflected in the design of SPC switching systems. It first discusses traffic problems for centralized control systems. It next discusses traffic problems inherent in distributed switching systems.

We develop a method of dimensioning and managing the bandwidth of a link on which TCP flows from access links are aggregated. To do this, we extend the application of the processor-sharing queue model to TCP performance evaluation by using flow statistics. To handle various factors that affect actual TCP behavior, such as round-trip time, window-size, and restrictions other than access-link bandwidth, we extend the model by replacing the access-link bandwidth with the actual file-transfer speed of a flow when the aggregation link is not congested. We only use the number of active flows and the link utilization to estimate the file-transfer speed. Unlike previous studies, the extended model based on the actual transfer speed does not require any assumptions/predeterminations about file-size, packet-size, and round-trip times, etc. Using the extended model, we predict the TCP performance when the link utilization increases. We also show a method of dimensioning the bandwidth needed to maintain TCP performance. We show the effectiveness of our method through simulation analysis.