To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
UnPMD: An Effective Power Saving Underloaded Physical Machine Detection Method in the Cloud Computing Environment |
|
Author |
Deafallah Alsadie |
| Citation |
Vol. 22 No. 4 pp. 710-720 |
|
Abstract |
Workload consolidation is a significant task in effectively using Cloud Computing resources over a minimum number of the physical machine. It enables reduction in cloud data center’s power consumption while meeting service level agreements between cloud users and service providers. It requires accurate determination of over utilized and underutilized physical machines that allows transfer of suitable virtual machines to other machines and ensure the least count of active physical machines. To that end, this work proposes a new method for detecting underutilized physical machines (UnPMD) based on the overall workload of the cloud data center. Besides, it also determines the coutn of active physical machines that can be vacated. It also avoids overloading the active physical machines while shifting the virtual machines from under utilized physical machines. The proposed (UnPMD) method is experimentally validated through simulations using cloudsim software. The obtained results demonstrate the superiority of the proposed (UnPMD) method over existing methods in terms of power consumption, Virtual Machine migration, service level agreement violation, and the number of physical machine shutdowns. The proposed (UnPMD) method can result into a 62% improvement in power consumption, a 109 % decrease in the number of service level agreement violations, and a 54% increase in the number of physical machine shutdowns over the existing methods under different scenarios. |
|
Keywords |
Cloud computing; Energy efficiency; VM consolidation; Underloaded server; Utilization. |
|
URL |
