Loading...

Holistic Energy Management in Distributed Data Center with Renewable Energy

Taheri, Somayyeh | 2020

266 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 54655 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Goudarzi, Maziar
  7. Abstract:
  8. In recent years, and with the development of information technology in various aspects of life, the growth of energy consumption in this sector has accelerated. With the advent of cloud computing and the development of its users, data centers have become one of the largest consumers of energy in the world. Adverse effects of fossil fuel consumption on the one hand and the global energy crisis as a limiting factor for development, on the other hand, have doubled the importance of managing the issue and reducing energy consumption in this sector. The two main energy-related concerns in data center construction are clean energy and energy management. The use of new energy and natural facilities in the environment to meet part of the data center's energy requirement and reduce energy consumption are among the attractive topics of many recent researches. However, in this context, the volatile and unpredictable nature of renewable energy, as well as its huge CAPEX, has been an obstacle to the expansion of its development.The geo-distributed architecture of data centers is one of the preferred structures in recent years. Higher flexibility and reliability, better disaster recovery, more availability, and finally scalability of productivity and goal-based improvement are among the benefits that drive data center operators to use several distributed data centers instead of the central one.Nevertheless, energy management and benefiting from new opportunities under this architecture are among the challenges that need to be discussed more.The subject of this research is energy management in data centers with renewable energy and free cooling. In this regard, considering the importance and challenges of using renewable energy and free cooling, we study the effect of these two decisions on the conventional evaluation parameters of the system and present a new parameter to evaluate the power usage efficiency in the data center. Then, given the challenge of investing time,we offer a solution with 44% less break-even time. In the following, considering the acceptability of geo-distributed architecture, we consider a cloud of distributed data centers with renewable energy and free cooling and examine the energy management challenges in it. In this context, we provide a solution for geographical load balancing among distributed data centers and thereby reduce the distributed energy consumption by 26%. Then, we examine the challenge of modeling and predicting the power consumption pattern in geodistributed data centers. For this purpose, we analyze the correlated parameters at the input to obtain an optimal set of important features through linear regression. Then, based on this set and using machine learning, we provide a predictive model of power consumption pattern with 87.2% accuracy. Then, we use the opportunity of predicting the pattern of power consumption to provide the opportunity of participating in emerging power markets.We formulate the power equations as well as energy billing in a contract of regulation services, and thereby reduce the electricity cost by 22%
  9. Keywords:
  10. Distributed Data ; Energy Management ; Free Cooling ; Renewable Energy Resources ; Uninterruptible Power Supplies ; Distributed Uninterruptible Power Source (UPS) ; Power Consumption Pattern Prediction

 Digital Object List

 Bookmark

No TOC