Title :
Energy efficient disaggregated servers for future data centers
Author :
Ali, Howraa M.Mohammad ; Lawey, Ahmed Q. ; El-Gorashi, Taisir E.H. ; Elmirghani, Jaafar M.H.
Author_Institution :
School of Electronic and Electrical Engineering, University of Leeds, LS2 9JT, United Kingdom
fDate :
June 30 2015-July 2 2015
Abstract :
With the dawn of cloud computing, data centers´ power consumption has received increased attention. In this paper we evaluate the energy efficiency potential of exploiting the concept of Disaggregated Server (DS) design in data centers for efficient resource provisioning. A DS, is a new approach for future racks where servers are disaggregated and resources, such as processors, memory and IO ports are arranged in resource pools constructing processing pools, memory pools and IO pools. We developed a mixed integer linear programming (MILP) model for energy minimization of the virtual machine (VM) placement problem in data centres implementing DS approach. The results show that the average power savings are up to 49% for the different VM types considered.
Keywords :
Data models; Energy efficiency; Memory management; Ports (Computers); Power demand; Program processors; Servers; Resource provisioning; VM placement; data center; energy efficiency; rack scale server;
Conference_Titel :
Networks and Optical Communications - (NOC), 2015 20th European Conference on
Conference_Location :
London, United Kingdom
DOI :
10.1109/NOC.2015.7238630