Software Engineering and Information Systems
Narges Akhound; Sahar Adabi; Ali Rezaee; Amir masoud Rahmani
Articles in Press, Accepted Manuscript, Available Online from 27 September 2022
Abstract
The advent of the Internet of Things (IoT) technology has made it possible for different devices to be widely connected to the Internet and interact. It has led to the production of large amounts of heterogeneous data. On the other hand, cloud computing is a convenient and efficient processing model ...
Read More
The advent of the Internet of Things (IoT) technology has made it possible for different devices to be widely connected to the Internet and interact. It has led to the production of large amounts of heterogeneous data. On the other hand, cloud computing is a convenient and efficient processing model for storing and processing data. Still, the increasing demand for real-time and delay-sensitive applications is increasing day by day. Due to network bandwidth limitations, these problems cannot be solved using cloud computing alone. A fog layer located between the IoT devices and the cloud computing layer has been proposed to overcome the problem of resource constraints in mobile devices. delay-sensitive applications run that require more volume and power resources. In this paper, end-to-end architecture for integrating IoT, fog, and cloud layers into a large-scale dispatched application is proposed to support high availability to make efficient use of fog-cloud resources and achieve the appropriate quality of service (QoS) in terms of delay and failure probability criteria. The mentioned architecture consists of three hierarchal layers: IoT devices, fog nodes, and cloud data centers. Depending on the processing power of each layer's resources, user requests may be executed on the same layer or sent to a higher layer. Then, quality characteristics such as availability, performance, and interoperability for the proposed architecture are evaluated by the ATAM scenario-based method. The basis of architectural evaluation and analysis in this method is the study of the requirements and the quality characteristics of the system architecture.
Computer Networks and Distributed Systems
Sara Mohammadi; Parvaneh Asghari; Amir Masoud Rahmani
Articles in Press, Accepted Manuscript, Available Online from 02 December 2022
Abstract
As a new technology, cloud computing is a key part of making systems more efficient and better and improving the Internet of Things. One of the significant challenges in fog computing is trust management, taking into account the processing, storage, and network constraints of fog devices. This study ...
Read More
As a new technology, cloud computing is a key part of making systems more efficient and better and improving the Internet of Things. One of the significant challenges in fog computing is trust management, taking into account the processing, storage, and network constraints of fog devices. This study suggests that a multi-objective imperialist competitive optimization algorithm be used to increase trust and decrease response time in fog environments. After formulating trust, delay, and accuracy, the multi-objective imperialist competitive optimization algorithm is developed and evaluated for fog server selection. Evaluations show that the proposed method is more efficient and works well in terms of accuracy, delay, and trust than other algorithms.
Computer Networks and Distributed Systems
Alireza Enami; Javad Akbari Torkestani
Volume 7, Issue 1 , February 2021, , Pages 19-34
Abstract
Fog computing is being seen as a bridge between smart IoT devices and large scale cloud computing. It is possible to develop cloud computing services to network edge devices using Fog computing. As one of the most important services of the system, the resource allocation should always be available to ...
Read More
Fog computing is being seen as a bridge between smart IoT devices and large scale cloud computing. It is possible to develop cloud computing services to network edge devices using Fog computing. As one of the most important services of the system, the resource allocation should always be available to achieve the goals of Fog computing. Resource allocation is the process of distributing limited available resources among applications based on predefined rules. Because the problems raised in the resource management system are NP-hard, and due to the complexity of resource allocation, heuristic algorithms are promising methods for solving the resource allocation problem. In this paper, an algorithm is proposed based on learning automata to solve this problem, which uses two learning automata: a learning automata is related to applications (LAAPP) and the other is related to Fog nodes (LAN). In this method, an application is selected from the action set of LAAPP and then, a Fog node is selected from the action set of LAN. If the requirements of deadline, response time and resources are met, then the resource will be allocated to the application. The efficiency of the proposed algorithm is evaluated through conducting several simulation experiments under different Fog configurations. The obtained results are compared with several existing methods in terms of the makespan, average response time, load balancing and throughput.
Computer Networks and Distributed Systems
Yaser Ramzanpoor; Mirsaeid Hosseini Shirvani; Mehdi GolSorkhTabar
Volume 7, Issue 1 , February 2021, , Pages 67-80
Abstract
Fog computing is known as a new computing technology where it covers cloud computing shortcomings in term of delay. This is a potential for running IoT applications containing multiple services taking benefit of closeness to fog nodes near to devices where the data are sensed. This article formulates ...
Read More
Fog computing is known as a new computing technology where it covers cloud computing shortcomings in term of delay. This is a potential for running IoT applications containing multiple services taking benefit of closeness to fog nodes near to devices where the data are sensed. This article formulates service placement issue into an optimization problem with total power consumption minimization inclination. It considers resource utilization and traffic transmission between different services as two prominent factors of power consumption, once they are placed on different fog nodes. On the other hand, placing all of the services on the single fog node owing to power reduction reduces system reliability because of one point of failure phenomenon. In the proposed optimization model, reliability limitations are considered as constraints of stated problem. To solve this combinatorial problem, an energy-aware reliable service placement algorithm based on whale optimization algorithm (ER-SPA-WOA) is proposed. The suggested algorithm was validated in different circumstances. The results reported from simulations prove the dominance of proposed algorithm in comparison with counterpart state-of-the-arts.