Loading...

Game Theory-Based Approach for Reliability and Power Management in Fog Computing

Younesi, Abolfazl | 2024

2518 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56829 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Ejlali, Alireza; Fazli, Mohammad Amin; Ansari, Mohsen
  7. Abstract:
  8. With the increasing development of Internet of Things (IoT) devices, issues such as establishing effective communication, optimizing energy consumption, ensuring reliability and improving the quality of services provided by these devices become more and more complex and critical challenges. With the introduction of fog computing model by Cisco in 2012, some of these challenges were successfully managed. Fog computing is a distributed computing paradigm that acts as a middle layer between cloud data centers and IoT-based devices/sensors. By distributing computing resources closer to edge devices, fog computing enables real-time data processing and analysis. However, one of the key aspects in fog computing revolves around ensuring efficient power management and reliability. Efficient power management is critical in fog computing due to the limited resources available in edge devices. Implementing power consumption management methods and improving resource allocation can help reduce energy consumption and increase the life of these devices. Furthermore, maintaining reliability is equally important as fog computing relies on the collaborative efforts of multiple interconnected devices. Implementing fault-tolerant approaches and ensuring seamless communication between devices is essential to guarantee the overall reliability and stability of the fog computing infrastructure. To address these challenges, we have introduced a new model that proposes a strategy based on non-cooperative game theory to strike a balance between power consumption and reliability. Our proposed model uses a master/backup strategy to guarantee the reliability objective by re-executing tasks to different nodes when a fault occurs, while dynamically scaling voltage and frequency to reduce power consumption. Power is used during task execution and maximizing overall efficiency. Non-cooperative game theory plays an essential role in our model, as it facilitates the development of strategies and solutions that maintain reliability while reducing energy consumption. By considering the trade-off between power and reliability as a non-cooperative game, our proposed method brings significant energy savings, with up to 35% reduction in energy consumption, 41% reduction in waiting time, 31% shorter completion time. and perform about 20% more tasks compared to other methods, and we use iFogSim software to evaluate this method
  9. Keywords:
  10. Fog Computing ; Game Theory ; Reliability ; Task Scheduling Algorithm ; Power Consumption ; Energy Consumption Reduction

 Digital Object List

 Bookmark

  • چکیده
  • فهرست جدول‌ها
  • فهرست تصویرها
  • فصل1 : مقدمه
    • 1-1 بیان مسئله
    • 1-2 اهمیت پژوهش
    • 1-3 ساختار پایان‌نامه
  • فصل2 مفاهیم اولیه
    • 2-1 نظریه بازی‌ها
    • 2-2 رایانش ابری
      • 2-2-1 ویژگی‌های رایانش ابری
      • 2-2-2 مزایا و محدودیت‌های رایانش ابری
    • 2-3 رایانش مه
      • 2-3-1 ویژگی‌های رایانش مه
      • 2-3-2 کاربرد‌های رایانش مه
      • 2-3-3 زمان‌بندی وظایف در رایانش مه
      • 2-3-4 شبیه‌ساز‌های رایانش مه
  • فصل3 کار‌های پیشین
  • فصل4 روش پیشنهادی
    • 4-1 تعریف مسئله و بیان فرضیات
    • 4-2 مدل سامانه
      • 4-2-1 توان مصرفی
      • 4-2-2 قابلیت اطمینان و تحمل‌پذیری اشکال
      • 4-2-3 راهکار پیشنهادی
        • 4-2-3-1 الگوریتم راهکاری پیشنهادی
        • 4-2-3-2 پیچیدگی زمانی تقریبی الگوریتم
    • 4-3 تجزیه و تحلیل ریاضی
    • 4-4 خلاصه فصل
  • فصل5 نتایج شبیه‌سازی
    • 5-1 ارزیابی خروجی
      • 5-1-1 الگوریتم‌های زمان‌بندی
      • 5-1-2 انرژی مصرفی
      • 5-1-3 نرخ اجرای موفق وظایف
      • 5-1-4 زمان تکمیل وظایف
      • 5-1-5 زمان انتظار وظایف
      • 5-1-6 متوسط توان مصرفی
  • فصل6 نتیجه‌گیری
    • 6-1 کار‌های آتی
  • منابع یا مراجع
  • پیوست1 مقالات مرتبط به پژوهش
  • پیوست2 آزمایشگاه‌های معروف حوزه رایانش مه
...see more