Research On Virtual Machine Placement Method Oriented To Resource Fragment Optimization

As an important support for cloud computing infrastructure,data centers have been increasing the scale of production.However,huge energy consumption and low resource utilization are the fatal shortcomings of data centers.Therefore,reasonable and effective placement of Virtual Machines(VMs)in the data center can effectively reduce energy consumption and improve resource utilization effectively.Domestic and foreign Scholars have achieved fruitful results in the research of VMs placement methods,but there are still some problems that need to be optimized and solved.In this paper,we focus on the problem of low resource utilization and energy waste in data centers caused by a large number of resource fragments,and conduct in-depth research on virtual machine placement methods.Two VMs placement methods for resource fragments optimization under cloud environments are given.The experimental results show that our proposed the VMs placement method can reduce resource fragments and improve resource utilization efficiently.As a result,the energy consumption is reducing.The main contribution of this paper is as follows:(1)A virtual machine placement method based on artificial bee colony algorithm MABC-VMP(Modified Artificial Bee Colony-Virtual Machine Placement)is proposed.MABC-VMP comprehensively considers the two resource dimensions of CPU and memory,and aims to reduce the number of resource fragments and the number of active physical machines..We classify the physical machines in the data center into active physical machines and idle physical machines,and discretizes the continuous artificial bee colony algorithm to solve the virtual machine placement problem.Simulation experiments show that the MABCVMP can reduce the number of resource fragments and improve resource utilization under the premise of ensuring lower energy consumption.(2)In order to improve the shortage of current virtual machine placement methods in optimizing resource fragment,a virtual machine placement method based on resource fragment optimization is proposed.In this paper,we proposes the BFD-RB(BFD-Resource Balance)method based on the Best Fit Decreasing(BFD).BFD-RB focuses on the internal resource balance of the physical machine.Firstly,physical machines are divided into active physical machines and idle physical machines.Secondly,the resource balance value is calculated according to the requested resources of the virtual machine and the remaining resources of the physical machine.And finally the appropriate target physical machine is selected based on the energy consumption awareness strategy.Simulation experiments show that BFD-RB has good performance in reducing resource fragments and the number of active physical machines,improving resource utilization and reducing energy consumption.Additionally,the BFD-RB is simple and easy to use,robust and adaptable.