An efficient approach for resource allocation to real-time traffic flows in fog environment using SDN

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: An efficient approach for resource allocation to real-time traffic flows in fog environment using SDN

ارائه دهنده: Provider: محمد صادق زاده - رشته کامپیوتر

اساتید راهنما: Supervisors: dr mohammad nassiri , dr reza mohammadi

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: dr hatam abdoli , dr mehdi sakhaei nia

زمان و تاریخ ارائه: Time and date of presentation: 20, September , 2022

مکان ارائه: Place of presentation: seminar 2 omran

چکیده: Abstract: The Internet of Things has various applications in the field of medicine, industry, and transportation, and in the field of Internet Objects, each of the Internet of Things devices has limited computing power, storage, and network. Cloud computing provides a suitable infrastructure for transferring calculations of Internet of Things applications to cloud servers that have high processing power and storage. The use of cloud computing in the Internet of Things, with many advantages, also has limitations such as high bandwidth consumption, high latency, and lack of proper scheduling of tasks. Fog computing is a computing model that has been introduced to deal with the challenges of the Internet of Things and the cloud so this architect extends cloud computing services to the edge of the network and processes IoT applications at the edge of the network. Fog computing includes servers that are located at the edge of the network, and in this way, the processing is done with less delay and less cost at the edge of the network, taking into account criteria such as energy consumption and the low capacity of fog servers, the allocation of resources and the decision to assign tasks to fog nodes are the main challenges in The quality of services causes the level of services such as delay and energy consumption, so it is important to decide to provide a comprehensive mathematical model and also provide an optimal task allocation algorithm. In this thesis, the mathematical model is presented by considering the service quality criteria of the service level and the limitations of fog nodes and the Internet of Things, and in the next step, the hybrid algorithm of genetics and gray wolf is presented, which is an improved hybrid algorithm for solving the problem of resource allocation. In the genetic algorithm, the solution space is searched comprehensively, so there will be less possibility to converge to a locally optimal point and the gray wolf algorithm also from the second half of the iterations due to the weakening of the role of exploration. It may get stuck in local optima. The combination of these two algorithms provides various solutions. It should be noted that the implementation of the mentioned algorithms has a processing cost and a computational delay, but due to the improvement of the service level quality criteria, this cost can be ignored, finally, with the help of simulation, the performance The research will be evaluated and compared with existing methods. The obtained results indicate that by using the mathematical model and the presented limitations, as well as combining and simultaneously using the positive points of the two algorithms, The results show that proposed algorithm improve the running time of 18.30% and completion time of the last task of 15.14%, as well as reducing energy consumption of 10.21% have better performance relative to the basic article

فایل: ّFile: Download فایل