Energy Consumption Optimization in Cloud Data Centers
Dzmitry Kliazovich, Pascal Bouvry, Fabrizio Granelli, Nelson L. S. da Fonseca · 2015
This chapter introduces the energy consumption of computing and communication devices, emphasizing how efficient energy consumption can be achieved, especially in communication networks. It addresses issues related to scheduling, load balancing, data replication, virtual machine placement, and networking that can be capitalized on to reduce the energy consumption in data centers. Enabling the sleep mode in idle computing servers and network hardware is the most efficient method of avoiding unnecessary power consumption. Consequently, load balancing becomes the key enabler for saving energy. Energy consumption virtualized data centers can be reduced by an appropriate decision on which physical server a virtual machine should be placed. Costs and operating expenses have become a growing concern in the cloud computing industry, with energy consumption accounting for a large percentage of the operational expenses in the data centers used as backend computing infrastructure.