الفهرس 
Chapter 1 IntroductionOnly 14 pages are availabe for public view
 |  | from | 95 |  |  |
| | | from | 95 | | |
Abstract
Virtualization technology has inspired cloud computing to emerge as a promising trend to provide users with services over the Internet. With the widespread usage of cloud computing to benefit from its services, cloud service providers have invested in constructing large-scale data centers. Consequently, a tremendous increase in energy consumption has arisen in conjunction with its results including the remarkable rise in costs of operating and cooling servers. Besides, increasing energy consumption has a significant impact on the environment due to emissions of carbon dioxide. Nowadays, consolidation of Virtual Machines (VM)s into the minimal number of servers, without degrading services provided to users, has attracted both industry and academia. Specifically, VM placement problem is considered a critical issue according to that research area. This thesis proposes a Power-Aware VM placement technique depending on Particle Swarm Optimization (PAPSO) to determine the optimal placement for migrated VMs. A discrete version of Particle Swarm Optimization (PSO), which is based on a decimal encoding, is adopted to map the migrated VMs to the best appropriate servers. PAPSO can consolidate migrated VMs into the minimum number of servers with a major constraint to decrease the number of overloaded hosts as possible. Therefore, it can reduce power consumption without violating Service Level Agreement (SLA). PAPSO has been implemented in CloudSim and tested using different test cases under random workloads. Experimental results in comparison to the Power Aware Best Fit Decreasing (PABFD) algorithm show that PAPSO outperforms PABFD in terms of consumed energy, number of migrations for VMs, number of server shutdowns and the combined metric Energy SLA Violation (ESV), respectively.