The Research On Dynamic Forecast Schedule Algorithm For Virtual Machine Placement In Cloud Computing Environment

Cloud computing is a popular trend in current computing,which promote the development of information technology.And more and more applications will be transferred to the cloud environment.These applications will share use the cloud environment resources with each other.So,how to effective management cloud environment's multiple applications is a challenging for current cloud computing.The core technology of cloud computing is virtualization,through use of virtualization technology multiple virtual machines can be consolidate.In most cloud computing platforms,the virtual machine quotas are seldom changed once initialized.Usually,use the virtual machine's peak memory consumption,and the current allocated resources are not efficiently utilized.So,for static deployment policy the virtual machine's initial memory is not fully used,and will cause cloud environment low rates of memory resource utilization.Through study the virtual machine deployment in cloud environment,the average utilization of cloud servers in most data centers can be improved through virtual machine placement optimization.This paper proposes a scheduling algorithm called VM-DFS(virtual machine dynamic forecast scheduling)to deploy virtual machines in a cloud computing environment.In this algorithm,through analysis of historical memory consumption,the most suitable physical machine can be selected to place a virtual machine according to future consumption forecast.This paper formalizes the virtual machine placement problem as a bin packing problem,which can be solved by the FFD(first-fit decreasing)scheme.Through this method,for specific virtual machine requirements of applications,we can minimize the number of physical machines.In order to ensure that virtual machines on each server's memory consumption to meet SLA requirements,each server machine's memory will set an upper threshold Lm,and for the overload exceeds the threshold rate will set a balance factor r.Experimental results show that,VM-DFS algorithm can meet Lm requirements under the premise of improving server memory resource usage.To ensure that the virtual machine to the requirement of memory resource SLA,when physical machine's memory consumption exceeds the set threshold rate greater than a pre-set value r,for this moment,the cloud environment should to start the live migration of virtual machine.For virtual machine live migration problem,this paper proposes a dynamic virtual machine migration model VM-DFM(virtual machine dynamic forecast migration),which the migration is based on dynamic prediction.The VM-DFM algorithm gives some suggestion for"hot spots" server migration of virtual machine list,choose the appropriate virtual machine to live migration.Experimental result show that dynamic prediction algorithm can obtain batter performance,and can reduce physical machine's memory consumption SLA violation,meanwhile the VM-DFM algorithm can reduce the number of virtual machine migration in the cloud environment.