Virtual machines migration game approach for multi‐tier application in infrastructure as a service cloud computing. Issue 6 (13th October 2020)

• Record Type: Journal Article
• Authors: Thanh Bui, Khiet; Dac Ho, Hung; Pham, Tran Vu; Tran, Hung Cong
• Is Part Of: IET networks. Volume 9:Issue 6(2020)
• Page Start: 326
• Page End: 337
• Publication Date: 2020-10-13
• Subjects: fault tolerance -- fuzzy systems -- fault diagnosis -- cloud computing -- virtual machines -- virtualisation -- game theory; infrastructure‐as‐a‐service cloud -- physical machines -- near optimal migration solution -- virtual machines migration algorithm -- virtual machines migration game approach -- multitier application -- service cloud computing -- proactive fault tolerance -- virtual migration algorithms -- optimal virtual machines migration; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2019.0204
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : Virtual machines migration is an essential feature of virtualisation technology which brings many advantages in cloud computing management process such as proactive fault tolerance, load balancing, and power management. There have been many virtual migration algorithms that have been developed but it is hard to find one that best suits all applications as well as ensures the desire of all stakeholders. The optimal virtual machines migration of cloud computing is usually NP‐hard or NP‐complete. In addition, determining when virtual machines migrate is a challenge because of the difficulty of fault detection in cloud computing. In this study, the authors propose a virtual machines migration game approach for multi‐tier application in infrastructure‐as‐a‐service cloud computing where they first proposed fault detection model based on Takagi–Sugeno fuzzy system with metric's physical machines and then optimal or near optimal migration solution is approximated based on Nash equilibrium. The results of experiment demonstrate the efficiency of their proposal. The proposed virtual machines migration algorithm has been benchmarked in order to highlight their strength and feasibility.