مقاله انگلیسی رایگان در مورد کمینه سازی هزینه انرژی با تعهد امنیت شغلی – الزویر 2017

Cloud computing, comprises of infrastructure platforms called Internet data center (IDC), is a large-scale distributed computing to meet the skyrocketing demand of big data applications and services. As an IDC typically consists of tens of thousands of servers, the energy consumption or energy cost is one of the critical problems. For example, many IDCs (e.g., Microsoft, Google, Akamai and INTEL) spend millions of dollars on electricity costs every year, which result in a large portion of operation expense [1, 2]. Hence, a considerable cost can be saved even reducing a few percent energy cost. In cloud computing environment, the jobs or application requests from the cloud users can be submitted to IDC, which are also considered as virtual network (VN) requests [3, 4]. These jobs or applications may be delay tolerant big data, such as scientific computing and data intensive MapReduce applications [2]. Generally, jobs or applications are arrived at IDC randomly. Therefore, the scheduling problem, which is also can be considered as the energy cost problem, is key issue for IDC to ensure the QoS of jobs and reduce the energy overhead [5, 6]. Recently, a great attention has been paid to IDC energy management by both academia and industry. Extensive research has been developed to minimize the energy cost by utilizing the electricity price dynamics across geographically distributed regions [7, 8], and apply VM migration to achieve the goals of saving energy [9, 10]. Especially, the electricity price manifests spatial and temporal diversity in the real life. For instance, in North America, owing to the different power generation profiles and electricity markets have been deregulated, the electricity prices are not constant but vary on the basis of an hour or 15-min [6]. To consider the temporal diversity of electricity price, the energy storage for energy cost saving is studied [11, 12], and both service delay and energy cost are taken into account in geographically distributed data center [2]. Security is another critical concern and even ranked as the greatest challenge in cloud computing environment. A survey from international data corporation shows that security is one of the greatest concerns in cloud computing [13]. Many works tackle the security problem on clusters [14], grid computing [15], heterogeneous distributed system [16, 17], cloud computing [18-21] and real-time embedded systems [22]. Unfortunately, because cloud computing environment is used to execute various applications of users, applications and users all may be the sources of malicious attack [23]. Furthermore, security mechanism is overlooked and has not been employed to counter any security threats [24, 25]. Therefore, it is necessary to deploy security services to protect various applications running in the IDC. However, security workload is incurred by adding security services to applications. Hence, it is a big challenge to tradeoff energy cost and service quality.