An efficient virtual machine allocation using single stage weapon target assignment model in cloud software‐defined network environment. (17th January 2022)

• Record Type: Journal Article
• Authors: Padmavathy, T.; Anitha, R.
• Is Part Of: International journal of communication systems. Volume 35:Number 6(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-01-17
• Subjects: cloud computing -- job assignment -- optimal network -- software‐defined networks -- virtual machine; Telecommunication systems -- Periodicals; 621.382
• DOI: 10.1002/dac.5086
• Languages: English
• ISSNs: 1074-5351

Abstract

Summary: With the rapid growth of the Internet, technological advancements, and the increasing number of cloud users, a paradigm transformation towards separate data and control planes with a centralized control point is taking place. As a result of this separation, software‐defined networks (SDNs) have distinct software controllers for routers and the capacity to manage network behavior. Virtual machine placement (VMP) and virtual machine migration (VMM) are two examples of SDN usage in data centers. The issue of VMP in cloud environments is a significant challenge that has been widely researched but not fully resolved. In this research, we used the single stage weapon target assignment (SSWTA) to model the virtual machine allocation (VMA) for jobs in order to maximize the resource utilization of the physical machines (PMs) by reducing idle time. Furthermore, this research proposes three algorithms: job assignment algorithms, PM allocation (PMA) using the least recently used (LRU) policy, and optimal VMA. The proposed VMA's performance is compared to other related VMA models such as first fit decreasing, and it is determined to be efficient in terms of resource utilization. Abstract : Single stage weapon target assignment (SSWTA) is used to model the virtual machine allocation (VMA) for jobs in order to maximize the resource utilization of the physical machines (PMs) by reducing idle time. Furthermore, this research proposes three algorithms: job assignment algorithms, PM allocation (PMA) using LRU policy, and optimal VMA. The proposed VMA's performance is compared to other related VMA models, and it is determined to be efficient in terms of resource utilization.