Green Computing: An SLA-based Energy-aware Methodology for Data Centers

摘要

Cloud computing enables a pay-as-you-go model of utility-oriented services. It also enables the hosting of common applications for consumer and business use. However, the energy consumption, operational costs, and carbon emissions of data centers are very high. Therefore, a Green Cloud Computing solution is needed not only to reduce the energy consumption and environmental impact of cloud computing, but also to minimize operational costs. Conventional Green Cloud computing solutions reduce the performance of Cloud data center hosts. They may also incur penalties for Service-Level-Agreement (SLA) violations. This study therefore proposes a new framework for improving the energy efficiency of data centers by using efficient virtual machine allocation and selection mechanisms to reduce power consumption. A DVFS (Dynamic Voltage and Frequency Scaling) technique is also used to shorten host wakeup time. The proposed framework addresses the growing importance of SLA by decreasing total operational costs.