Search

This paper investigates the power-saving mechanisms of Discontinuous Reception (DRX), a technique used in wireless communication networks to reduce energy consumption. By employing a discrete-time Geo/Geo/1 queueing model with differentiated vacations and system disasters, we aim to more accurately capture the intermittent nature of data arrivals, often overlooked in continuous-time models. Our research addresses the existing gap in the literature by providing a more realistic representation of DRX behaviour. Understanding the performance and characteristics of DRX is crucial for optimizing energy efficiency and improving the overall performance of wireless networks. This paper contributes to this understanding by deriving steady-state probabilities, calculating key performance metrics, and visualizing the system behaviour through graphical analysis. [ABSTRACT FROM AUTHOR]

In this paper, we propose a binary feedback scheme for congestion control in computer networks using a gradient-based mechanism which employs neural network modelling of the system dynamics. The optimal direction for rate adjustment at the source is based on a single bit feedback signal which depends upon the sign of the sensitivity of the system performance index with respect to queue input rate. The paper presents the scheme and the analysis that went into the choice of the various decision mechanisms. Simulation results are presented to show the performance of the gradient based technique as compared to the conventional queue-based approach. Copyright © 2000 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Computer networks are by far the most popular way of transmitting information between users, despite the distance and technical and software differences. There are a large number of topologies of such networks, but the standard is client-server architecture with one server and many client machines, which will be considered and analysed in this paper. The purpose of this study is to analyse and evaluate the efficiency and speed of such a network by mathematical calculations considering such network parameters as the number of channels, the number of network nodes, the allocated space for pending requests, the frequency of requests, the average time allocated for processing one network request. The investigation and analysis of this topic will be based on analytical models created using the mathematical theory of queues applied to the principle of operation of the client-server architecture of a computer network containing only one server node. In the course of the study, the possible maximum speed of information transfer in a client-server architecture computer network with a priority of requests to a single server was mathematically calculated and considered, mathematical models were constructed, the formulas describing the dependencies of variables among themselves and their impact on the result were derived, and graphs based on them were constructed. The efficiency of the network was determined at different loads based on the data obtained. In particular, the calculation was made of the number of request packets in the network for its different loads for different percentages of the operation time of such networks. The work carried out will provide a better understanding of the methods of optimising and improving the performance of computer systems of such architecture, which will be useful in the course of their further design and physical implementation. [ABSTRACT FROM AUTHOR]

The behavior of customers plays a vital role in realizing the nature of a queue. If there is a favor for customers from the side of service facility the arrival rate increases than usual. Also the positive perspective about the service providers also encourages more number of customers to join the system. The arrival rate of the customers follow Poisson distribution. This paper analyses a queuing model with those encouraged customers who urges to join the system. Here the customers are served in batches according to the general bulk service rule along with the phenomenon that the servers undergo repeated vacations until they find minimum number of customers to start the service. In addition this paper interprets the scenario that if there is a breakdown in the service facility, the waiting line of the customers increases which causes a greater impact on the effectiveness of the service providers favoring the customers. On account of this situation the steady state probability solutions and some performance measures are evaluated along with a numerical illustration. [ABSTRACT FROM AUTHOR]

This paper investigates a single-server queuing system with heterogeneous service, failure, and maintenance. The proposed model features a server acting as both the main and backup server. System failure can occur at any stage. When a failure happens, instead of stopping the service entirely, the main server functions as a backup, providing service at a reduced rate. Once all jobs in the system have been serviced, the backup server enters the maintenance state. Following the repair process during maintenance, the server transitions to an idle state, awaiting incoming jobs. Explicit expressions for both transient and steady-state behaviours of the system are derived. Additionally, key system performance metrics are discussed in this paper, accompanied by graphical illustrations to visualize system size probabilities and performance indices. [ABSTRACT FROM AUTHOR]

A paper published in 1983 in this journal by A. de Palma et at. uses a deterministic queueing supply model in combination with a random utility demand model to predict the pattern of traffic flows during a peak period. Two different versions of the supply model are included, a "basic model" which is identical to the supply model used by a number of other authors investigating the same general problem (e.g., M. J. Smith, 1984), and a more complicated "extended model." As detailed below, there is a mathematical discrepancy in the analysis. Because of this discrepancy, some of the conclusions about the traffic flow pattern in the case of the extended model are incorrect and all are unproven. For the simpler-and seemingly more realistic-basic model the discrepancy vanishes and the conclusions remain valid. [ABSTRACT FROM AUTHOR]

This paper investigates a multiprocessor queueing system where tasks require varying numbers of processors for service i.e., unlike classical multiserver queuing systems, here, to service each task needs a specific required number of service nodes. It presents a detailed model and derives analytical solutions for the steady-state probabilities of the system. The study addresses the challenges of organizing computations in multiprocessor environments, considering factors such as task arrivals, service times, and resource constraints. Through rigorous mathematical analysis, this work provides insights into the dynamics of task scheduling and resource utilization in such systems. Our findings contribute to advancing queueing theory applications in multiprocessor systems and offer valuable implications for optimizing resource allocation and system performance in practical computing scenarios. [ABSTRACT FROM AUTHOR]

In the dynamic world of cloud computing, auto-scaling stands as a beacon of efficiency, dynamically aligning resources with fluctuating demands. This paper presents a comprehensive review of auto-scaling techniques, highlighting significant advancements and persisting challenges in the field. First, we overview the fundamental principles and mechanisms of auto-scaling, including its role in improving cost efficiency, performance, and energy consumption in cloud services. We then discuss various strategies employed in auto-scaling, ranging from threshold-based rules and queuing theory to sophisticated machine learning and time series analysis approaches. After that, we explore the critical issues in auto-scaling practices and review several studies that demonstrate how these challenges can be addressed. We then conclude by offering insights into several promising research directions, emphasizing the development of predictive scaling mechanisms and the integration of advanced machine learning techniques to achieve more effective and efficient auto-scaling solutions. [ABSTRACT FROM AUTHOR]

The couple between the power network and the transportation network (TN) is deepening gradually with the increasing penetration rate of electric vehicles (EV), which also poses a great challenge to the traditional voltage control scheme. In this paper, we propose a coordinated voltage control strategy for the active distribution networks considering multiple types of EV. In the first stage, the action of on-load tap changer and capacitor banks, etc., are determined by optimal power f low calculation, and the node electricity price is also determined based on dynamic time-of-use tariff mechanism. In the second stage, multiple operating scenarios of multiple types of EVs such as cabs, private cars and buses are considered, and the scheduling results of each EV are solved by building an optimization model based on constraints such as queuing theory, Floyd-Warshall algorithm and traffic flow information. In the third stage, the output power of photovoltaic and energy storage systems is fine-tuned in the normal control mode. The charging power of EVs is also regulated in the emergency control mode to reduce the voltage deviation, and the amount of regulation is calculated based on the fair voltage control mode of EVs. Finally, we test the modified IEEE 33-bus distribution system coupled with the 24-bus Beijing TN. The simulation results show that the proposed scheme can mitigate voltage violations well. [ABSTRACT FROM AUTHOR]

The long queuing time for electric vehicles to charge under intelligent network connection leads to low distribution efficiency. Therefore, this paper proposes a strategy to predict the probability of queues forming for electric vehicles arriving at charging stations under intelligent network connection. Both the dynamic demand of customers and the characteristics of the alternating influence of charging vehicles should be considered when studying such problems. Based on the above problem characteristics, a real-time dynamic charging selection strategy is developed by predicting the probability of other vehicles in the region going to the charging station. A distribution path optimization model based on intelligent network connection and queuing theory is proposed for electric logistics vehicles in charging mode, taking into account the time window constraint and the influence of charging vehicles when using intelligent network connection for path planning. The objective is to minimize the total cost, and the route for electric logistics vehicles is adjusted in real time. This is solved by an improved hybrid genetic-annealing algorithm. The experimental results show that this paper obtains real-time dynamic road information and charging information with the help of intelligent network connection. It predicts the queuing probability of electric vehicles by combining with queuing theory, which can help select a more suitable charging location and timing for electric logistics vehicles. This can effectively avoid peak periods and reduce waiting times. By comparing with other models, this paper's model can save the distribution cost of electric vehicles. [ABSTRACT FROM AUTHOR]

Computation offloading and service migration are two major research hotspots in the mobile edge computing (MEC) environment. However, in the existing MEC architecture, the idle computing resources of offsite edge servers are not fully utilized, which leads to the problem of high overall system time and energy costs. In this paper, we propose a multi-edge collaborative computation offloading strategy for this problem. The strategy analyzes and calculates the energy consumption and latency cost of task execution for local terminals, edge servers and central cloud, constructs a computation offloading model with the weighted sum of latency and energy consumption as the optimization objective, and then solves the model using an improved genetic algorithm to obtain the best computation offloading decision. On the other hand, the mobility of users in the MEC environment leads to service migration, which leads to unbalanced load on the edge servers and network congestion, etc. This paper proposes an energy threshold-based task migration strategy. The strategy analyzes the time and energy consumption of service execution and data transmission, designs an edge server selection algorithm based on the energy consumption threshold, constructs a service migration model, and finally solves the optimal service migration strategy by improving the genetic algorithm. Experimental results show that the multi-edge collaborative computation offloading strategy proposed can significantly improve the performance of data transfer cost, energy consumption, and task completion time. The proposed migration strategy based on energy consumption threshold can significantly improve the performance of mobile server energy consumption, service completion time, and data transfer energy consumption. [ABSTRACT FROM AUTHOR]