Abstract
Cloud computing has innovated the IT industry in recent years, as it can delivery subscription-based services to users in the pay-as-you-go model. Meanwhile, multimedia cloud computing is emerging based on cloud computing to provide a variety of media services on the Internet. However, with the growing popularity of multimedia cloud computing, its large energy consumption cannot only contribute to greenhouse gas emissions, but also result in the rising of cloud users’ costs. Therefore, the multimedia cloud providers should try to minimize its energy consumption as much as possible while satisfying the consumers’ resource requirements and guaranteeing quality of service (QoS). In this paper, we have proposed a remaining utilization-aware (RUA) algorithm for virtual machine (VM) placement, and a power-aware algorithm (PA) is proposed to find proper hosts to shut down for energy saving. These two algorithms have been combined and applied to cloud data centers for completing the process of VM consolidation. Simulation results have shown that there exists a trade-off between the cloud data center’s energy consumption and service-level agreement (SLA) violations. Besides, the RUA algorithm is able to deal with variable workload to prevent hosts from overloading after VM placement and to reduce the SLA violations dramatically.
Highlights
IntroductionThe cloud computing paradigm [1] has rapidly attracted much of the public’s attention in recent years, as it can delivery subscription-based services to users in the pay-as-you-go model
The cloud computing paradigm [1] has rapidly attracted much of the public’s attention in recent years, as it can delivery subscription-based services to users in the pay-as-you-go model. It is convenient for users who come from different places in the world to access the cloud services which are usually described as infrastructure as a service (IaaS), platform as a service (PaaS)
The analysis of the results shows that the combination of minimum migration time (MMT) for virtual machine (VM) selection and local regression (LR) for finding overloaded hosts has better performance in energy consumption, service-level agreement (SLA) violation and the number of VM migrations when compared to other combinations
Summary
The cloud computing paradigm [1] has rapidly attracted much of the public’s attention in recent years, as it can delivery subscription-based services to users in the pay-as-you-go model. It is convenient to take advantage of the live migration technology to consolidate VMs into a few servers in order to improve the resource utilization of the physical server and to switch off the idle host to reduce the cloud data center’s energy consumption. The algorithm can effectively find proper a underutilized host and migrate out all of the VMs running on the host, switches the host to sleep mode for power saving It has taken different kinds of PMs into consideration and can be applied to large heterogeneous cloud data centers. It is more robust to deal with the variable workload when placing VMs on the physical server These two algorithms have been combined and applied to the cloud data center for reducing both energy consumption and SLA violations.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
