Support vector machine approach for virtual machine migration in cloud data center

Fan Hsun Tseng, Xiaojiao Chen, Li Der Chou, Han Chieh Chao, Shiping Chen

Research output: Contribution to journalArticlepeer-review

16 Scopus citations


The social media services are popular with Internet services today, such as Facebook, YouTube, Plurk and Twitter. However, the enormous interactions among human beings also result in highly computational costs. The requested resources and demands of some specific social media services are changing severely, and the virtual machines (VMs) exhaust the computing resource of physical machine (PM). Thus this will lead to VM migration. Many researchers investigate how to stabilize the average utilization of virtual machines and physical machines in cloud data center. In this paper, we formulated the VM migration problem in cloud data center based on mixed integer linear programming (MILP). Then, the VM allocation algorithm was proposed to allocate the VMs among the PMs, which is based on the Support Vector Machine (SVM). According to the training process during a specific time, the minimum numbers of VM migration and maximum resource utilization of PMs were accomplished. As the allocation case and simulation results showed, we achieved the stable and low-cost for social media services in cloud data center.

Original languageEnglish
Pages (from-to)3419-3440
Number of pages22
JournalMultimedia Tools and Applications
Issue number10
StatePublished - 16 May 2015


  • Cloud data center
  • Load balance
  • Mixed integer linear programming
  • Social media service
  • Support vector machine


Dive into the research topics of 'Support vector machine approach for virtual machine migration in cloud data center'. Together they form a unique fingerprint.

Cite this