ارائه مکانیسمی کارآمد برای تعیین تابعی برای توزیع بارکاری در محاسبات مه با استفاده از سیستم های دسته بند یادگیر

نوع مقاله : مقاله پژوهشی

نویسندگان

1 کارشناسی ارشد دانشگاه بوعلی سینا همدان، همدان، ایران

2 دانشیار دانشکده فنی و مهندسی، دانشگاه بو علی سینا همدان، همدان، ایران

چکیده

با توسعه سریع اینترنت اشیا ، الگوی محاسبات مه به عنوان یک راه حل جذاب برای پردازش داده‌ها‌ در برنامه‌ها‌ی اینترنت اشیا ارائه شده است. در محیط مه ، برنامه‌ها‌ی اینترنت اشیا توسط گره‌ها‌ی محاسباتی میانی در مه و همچنین کارگزارهای فیزیکی در مراکز داده ابری اجرا ‌می‌شوند. از این رو، مسائل مربوط به مدیریت منابع و مدیریت انرژی به عنوان یکی از مشکلات چالش برانگیز در محاسبات مه باید مورد توجه قرار گیرد. اخیرا پژوهش هایی برای ایجاد تعادل بین انرژی و هزینه در محاسبات مه صورت گرفته است. در این پژوهش ضمن بررسی این رویکردها روشی کارآمد برای تقریب تابع توزیع بار با دو روش مبتنی بر سیستم های یادگیر دسته‌بند به نام XCSF و BCM-XCSF در گرههای پردازشی مه به منظور بهینه سازی هرچه بیشتر رویکردهای قبلی و مدیریت منابع پردازشی مه ارائه شده است. این دو روش در داشتن یک حافظه برای نگهداری بهترین دستهبند‌ها‌ با هم متفاوت هستند. نتایج آزمایش‌ ها نشان می دهند که این دو روش همانند XCS و BCM-XCS توزیع بار مناسبی دارند. این دو روش بخصوص روش BCM-XCSF افزون بر این که سربار محاسباتی را کم ‌می‌کنند روش اخیر، حدود 60 درصد تاخیر را کاهش میدهد و مصرف انرژی را بهینه تر ‌می‌کند.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

An efficient mechanism to determine a function for load distribution in fog computing based on the use of learning classification systems

نویسندگان [English]

  • Bahare HamidiMoheb 1
  • Mahdi Abbasi 2
1 Computer Engineering Department, Boali Sina University, Hamedan, Iran
2 Associate Professor, Technical and Engineering Faculty, Boali Sina University, Hamedan, Iran
چکیده [English]

In recent years, the Internet of Things is one of the most popular technologies that facilitate new interactions between people and humans to increase the quality of life. With the rapid development of the Internet of Things, the fog computing model is emerging as an attractive solution for data processing of Internet of Things applications. In the fog environment, IoT applications are run by intermediate computing nodes in the fog as well as physical servers in cloud data centers. On the other hand, due to resource limitations, resource heterogeneity, dynamic nature and lack of energy, it is necessary to consider resource management and energy management issues as one of the challenging problems in fog computing. Recently, some researches have been done to create a balance between energy and cost in fog computing. In this research, while examining these approaches, an efficient method for approximating the load distribution function with two methods based on batch learning systems called XCSF and BCM-XCSF in fog processing nodes in order to optimize the previous approaches as much as possible and manage fog processing resources. These two methods differ in having a memory to store the best classifiers. Experiments indicate that these two methods, like XCS and BCM-XCS, have a suitable load distribution. These two methods, especially the BCM-XCSF method, in addition to reducing the computational overhead; It reduces the delay by about 60% and optimizes energy consumption.

کلیدواژه‌ها [English]

  • Internet of things
  • fog computing processing
  • function estimation
  • learning classification systems
  • Ni, K. Zhang, . X. Lin and . X. . S. Shen, "Securing fog computing for internet of things applications: Challenges and solutions," IEEE Communications Surveys & Tutorials, vol. 20, pp. 601-628, 2017.
  • Chiang and . T. Zhang, "Fog and IoT: An oveBrview of research opportunities," IEEE Internet of Things Journal, vol. 3, pp. 854-864, 2016.
  • Al-Fuqaha, M. Guizani, M. Mohammadi, . M. Aledhari and . M. Ayyash, "Internet of things: A survey on enabling technologies, protocols, and applications," IEEE Communications Surveys & Tutorials, vol. 17, pp. 2347-2376, 2015.
  • -H. Hong and . B. Varghese, "Resource Management in Fog/Edge Computing: A Survey," arXiv preprint arXiv:1810.00305, 2018.
  • Zhang, . M. Zhou and . G. Fortino, "Security and truäst issues in Fog computing: A survey," Future Generation Computer Systems, vol. 88, pp. 16-27, 2018.
  • Shi, J. Cao, Q. Zhang, Y. Li and . L. Xu, "Edge computing: Vision and challenges," IEEE Internet of Things Journal, vol. 3, pp. 637-646, 2016.
  • M. Talaat, . M. S. Saraya, . A. I. Saleh, H. A. Ali and S. H. Ali , "A load balancing and optimization strategy (LBOS) using reinforcement learning in fog computing environment," Ambient Intell Human Comput, vol. 11, p. 4951–4966, 2020.
  • W. Wilson, "Classifiers that approximate functions," Natural Computing, vol. 1, p. 211–234, 2002.
  • Holland et al., "What Is a Learning Classifier System?," in Learning Classifier Systems, 2000.
  • Bartin, "Use of learning classifier systems in microscopic toll plaza simulation models," IET Intelligent Transport Systems, vol. 13, pp. 860-869, 2019.
  • R. Karlsen and S. Moschoyiannis, "Evolution of control with learning classifier systems," Applied network science, vol. 3, p. 30, 2018.
  • Xu, . L. Chen and S. Ren, "Online learning for offloading and autoscaling in energy harvesting mobile edge computing," IEEE Transactions on Cognitive Communications and Networking, vol. 3, pp. 361-373, 2017.
  • Kanapram, . G. Lamanna and . M. Repetto, "Exploring the trade-off between performance and energy consumption in cloud infrastructures," in Second International Conference on Fog and Mobile Edge Computing (FMEC), 2017.
  • Wu, L. Chen, C. Shen, . W. Wen and . J. Xu, "Online geographical load balancing for energy-harvesting mobile edge computing," in 2018 IEEE International Conference on Communications (ICC), 2018.
  • Wan, B. Chen, S. Wang, M. Xia, D. Li and C. Li, "Fog Computing for Energy-aware Load Balancing and Scheduling in Smart Factory," IEEE Transactions on Industrial Informatics, 2018.
  • Zeng, L. Gu and H. Yao, "Towards energy efficient service composition in green energy powered Cyber–Physical Fog Systems," Future Generation Computer Systems, 2018.
  • Abbasi, M. Yaghoobikia, M. Rafiee, A. Jolfaei and M. R. Khosravi, "Efficient resource management and workload allocation in fog–cloud computing paradigm in IoT using learning classifier systems," Computer Communications, vol. 153, pp. 217-228, 2020.
  • W. Wilson, "Get Real! XCS with Continuous-Valued Inputs," in International Workshop on Learning Classifier Systems, 2000.
  • W. Wilson, "Classifiers that approximate functions," Natural Computing, vol. 1, p. 211–234, 2002.
  • Lanzi, "XCS with Computed Prediction for the Learning of Boolean Functions," in IEEE Congress on Evolutionary Computation, 2005.
  • fchauvel, "XCSF," [Online]. Available: https://github.com/fchauvel/XSCF.Optimization of Electrical and Electronic Equipment (pp. 167–173).