A Review on Dynamic Consolidation of Virtual Machines for Effective Energy Management and Resource Utilization in Data Centres of Cloud Computing

John, Nimmol P.; Bindu, V. R.

doi:10.1109/iccmc48092.2020.iccmc-000114

Cited by 6 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Numerous studies have been performed to optimize the VMP problem in CDCs. [12][13][14] Various objectives in these studies such as maximizing revenue, reducing service costs, maximizing QoS parameters, minimizing network traffic, reducing energy consumption, reducing delay, maximizing utilization, and non-violation of SLA have been considered. 10 In this regard, various methods have been proposed to address the VMP problem such as deterministic and meta-heuristic algorithms.…”

Section: Literature Reviewmentioning

confidence: 99%

“…10 In this regard, various methods have been proposed to address the VMP problem such as deterministic and meta-heuristic algorithms. 12,13 Two types of energy saving algorithms 12 and energy efficiency algorithms 13 are more important to solve VMP problem in CDCs. The main idea of these algorithms is to reduce energy consumption by considering SLAs.…”

Section: Literature Reviewmentioning

confidence: 99%

“…However, migration can lead to SLA violation and QoS degradation, which must be addressed through a compromise between energy consumption and SLA violation. 13 VMP is an important problem for achieving dynamic consolidation of VMs in the minimum number of PSs, which deals with the problem of live migration of VMs. 13 In the VMP problem, there is a set of M migrated VMs as VM = { vm 1 , vm 2 , … , vm j , … , vm M } that must be assigned to a set of N PS available as PS = { ps 1 , ps 2 , … , ps i , … , ps N } .…”

Section: Problem Formulationmentioning

confidence: 99%

“…13 VMP is an important problem for achieving dynamic consolidation of VMs in the minimum number of PSs, which deals with the problem of live migration of VMs. 13 In the VMP problem, there is a set of M migrated VMs as VM = { vm 1 , vm 2 , … , vm j , … , vm M } that must be assigned to a set of N PS available as PS = { ps 1 , ps 2 , … , ps i , … , ps N } . Let vm mg be a migrated VM with the required resources including processor (P mg ), memory (M mg ), storage (S mg ), and bandwidth (B mg ).…”

Section: Problem Formulationmentioning

confidence: 99%

“…Due to different and contradictory objectives, VMP problem is often formulated as a hybrid multi-objective problem. 13,22 The following are some of the most important objectives of the VMP problem.…”

Section: Optimization Objectivesmentioning

confidence: 99%

See 4 more Smart Citations

Dynamic placement of virtual machines using an improved multi‐objective teaching‐learning based optimization algorithm in cloud

Wang¹,

Osmani

Mirzaei

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Recently, cloud computing is growing rapidly and contributing to the realization of other technologies. Cloud data centers use a lot of energy to host applications, which leads to high operating costs and increased carbon dioxide emissions into the environment. Dynamic consolidation of virtual machines (VMs) into the minimum number of physical servers is an efficient approach to managing energy consumption in cloud. Virtual machines placement (VMP) is an important problem in the proper consolidation of VMs and is the most common way to improve resource utilization. In VMP problem, VMs are consolidated to minimize the number of active physical servers. This consolidation is done using migration of VMs. After placement, the VMs execute on selected physical servers and the underloaded physical servers are off to optimize energy consumption. Basically, the purpose of VMP is to allocate a set of VMs to a subset of physical servers, taking into account some objectives such as reducing energy consumption and violating the service level agreement (SLA). In this article, an improved teaching-learning based optimization (TLBO) algorithm is formulated to solve VMP as a multi-objective problem, which we call VMP-TLBO. VMP-TLBO can perform optimal placement of migrated VMs to physical servers with the purpose of reducing energy consumption and SLA violation. VMP-TLBO is implemented in MATLAB and the experimental results show that proposed algorithm does not violate SLA and compared to the best equivalence algorithm (ie, PAPSO), it improves energy consumption and SLA violation by 1.8% and 5.6%, respectively. In addition, VMP-TLBO consolidates migrated VMs into the minimum number of physical servers to minimize the number of overloaded hosts as much as possible.

show abstract

Section: Literature Reviewmentioning

confidence: 99%