Temperature-aware Power Dissipation Analysis in Hyperscale Data Centers

Rezaei Mayahi Nejad, Mehdi | 2019

483 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: English
  3. Document No: 51909 (52)
  4. University: Sharif University of Technology, International Campus, Kish Island
  5. Department: Science and Engineering
  6. Advisor(s): Sarbazi-Azad, Hamid
  7. Abstract:
  8. Hyperscale cloud data center (HCDC) (a.k.a., hyperscale data center) is the backbone of a wide variety of Internet services such as web-hosting, ecommerce, social networking, software as a service (SaaS), platform as a service (PaaS), application as a service (AaaS), and cloud computing.HCDC platforms consist of massive parallel processing, high utilization rate, and high volume storage that cost hundreds of million dollars.The increased demand for various Internet services and the subsequent dramatic growth in HCDC platforms, have caused an exponential increase in utilization of energy. With energy costs on the rise and global attention focused on carbon footprints, organizations are looking for new ways to detect, assess, and reduce the power dissipation in order to increase eciency and decrease cost of services. Current computational methods for evaluating the total power consumption of HCDC use prede ned parameters such as total number of equipment, unit equipment power consumption, average utilization rate, and power usage eectiveness (PUE). These methods are temperature- ndependent.However, the temperature of HCDC servers is generally higher than that of traditional data center platforms because of their higher utilization rate, which as a result, increases the overall amount of power dissipation.To address this challenge, we rst demonstrate in what ways high-temperature environment can impact the total power/energy consumption. Second, we propose a temperature-aware power consumption model, which calculates the server power consumption value using server temperature and idle (or peak) server power.Finally, we use this model to show the impacts of high temperature on metrics such as, total power consumption, operational cost (OpEx), total cost of ownership (TCO) and lifecycle cost analysis (LCCA). This model is the rst attempt to estimate server power consumption by taking servers' temperature into account
  9. Keywords:
  10. Hyper Scale Data Center ; High Utilization Rate ; Power Consumption ; Temperature ; Operational Expenditure (OPEX) ; Total Cost ; Life Cycle Cost (LCC)

 Digital Object List