Resource Allocation Strategy Research In Cloud Computing Environment

Take 2010 as the boundary, we can put the past into the PC era and the Internet era, andthe future is Cloud computing, Internet of Things and Three Networks Convergence era. Thebirth of new calculation model of Cloud computing lead to Revolutionary change in the ICTindustry .It became the hot spot in our industry when Cloud computing appeared. Its designconcept is that users obtain a variety of services like use of water and electricity accordingthey need to use and pay to use. This calculation mode let every one can contact with verylow cost to enjoy top ICT technology. As the world of ICT business giants like Microsoft,Google, IBM, Amazon, Salesforce, etc have launched their own cloud computing products,considering Cloud computing as the future development of one of the main strategic goal.How to allot resource efficiently is an important issue to be resolved in cloud computingenvironment. It is also a hotspot in cloud computing environment. In order to use resourceefficiently, a resource allocation and adjusted strategy based on modified particle swarmalgorithm was proposed in this paper. The improved particle swarm algorithm particle iscalled dynamic particle swarm optimization algorithm. Because in cloud computingenvironment the number of resources is huge, the number of user requirements is huge, howto allocate the resources in a short time is what we expected. Compare to the traditionalparticle swarm algorithm, the time complexity of dynamic particle swarm optimizationalgorithm is far better than traditional particle swarm optimization algorithm, thus greatlysaving the time of resource allocation in cloud computing environment. According toworkload characteristic, a QoS utility function was constructed. The prices of the resourceswere dynamically adjusted by the corresponding resource agents based on the requirement ofall of the workloads. At the same time particle swarm algorithm was employed to maximizeprofit of a certain workload. The results of simulation experiments validate that modifiedparticle swarm algorithm in resource allocation is effective, feasibility, robustness andshortest computing timeCloud computing data center is constituted by the huge of heterogeneous computingresources pool, storage resources pool and network resources pool, etc. When all servers arestarted, they will consume large amounts of electricity energy, discharge a lot of waste gascontaminants in our environment cause great pollution. Therefore, we need to find greenenergy saving solutions, not only can minimize the execution of the cost but also can reducethe impact on the environment. According to the above question, in this paper we present theresources provisioning and allocation algorithm for high efficiency energy saving. The basicidea of this algorithm is that through the support of operating on different VMs betweenphysical nodes, and then according to the needs of the execution, dynamically migrating the VMs. When not using the resources that host provided, it will dynamically change orintegrate into a minimum number of physical node, and the idle physical node will be turnedoff and eliminate the energy, at last reduce the total energy consumption. Finally compare toother allocation of resources algorithms the algorithm which presented in this paper isvalidity and optimal in saving energy.No matter in which calculation mode, resource allocation are need to be urgentlyaddressed question。It is no exception in cloud computing environment. This paper presenttwo aspects of the research about resource allocation. The first aspect is the resourceallocation strategies based on market mechanism. Consider service providers, services Qosand balanced load, we put forward a set of effective resource allocation methods. The secondaspect is from green energy-saving Angle and real life inspired, we puts forward how tointegrate resources and save energy, and finally reach the purpose of environmentalprotection.