首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0719113-162126


URN etd-0719113-162126 Statistics This thesis had been viewed 1567 times. Download 338 times.
Author Cheng-Sin Guo
Author's Email Address No Public.
Department Computer Science and Enginerring
Year 2012 Semester 2
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 40
Title Dynamic Placement Strategy for Virtual Machine in Virtual Cluster Environment
Keyword
  • OpenNEbula
  • Virtual Cluster
  • Virtualization
  • Dynamic Placement Strategy
  • Dynamic Placement Strategy
  • Virtualization
  • Virtual Cluster
  • OpenNEbula
  • Abstract Virtualization technology is a facility that allows multiple operating systems to share same physical resource, the benefits of virtualization are improved the resource utilization and energy-efficient, therefore, the cloud computing relies on virtualization technology, however, due to the abstraction of cloud computing environment, how to allocate hardware resources to the virtual machine will directly affect the system performance.
    OpenNEbula(ONE) is an open-source virtual infrastructure , Aims to provide rich functionality, flexible approach and better interoperability and scalability to simplify the management of large data centers and existing virtual infrastructure. Thus cloud service providers can build their own cloud easily. ONE also supports most of hypervisors. However, three default placement policies which ONE provides do not have dynamic placement when server is overloaded, resulting in considerable waste of resources. Therefore, we provide dynamic placement policy to improve this problem, using the built-in attributes by ONE to filter high load node so as not to continue into the virtual machine scheduling. Also adding a dynamic monitoring mechanism in ONE system, make each nodes’ CPU usage in a reasonable range to maintain stable services. Whenever a node’s CPU usage is overloaded, using the virtual machine live migration techniques to alleviate the burden of high load node, makes cloud service more stable.
    Advisor Committee
  • Chia-Ying Tseng - advisor
  • Liang-Teh Lee - co-chair
  • Trong-Yen Lee - co-chair
  • Files indicate in-campus access immediately and off-campus access at 2 years
    Date of Defense 2013-07-04 Date of Submission 2013-07-22


    Browse | Search All Available ETDs