Model And Scheduling Strategy Of Network Intensive Applications In Cloud

With the development of information science and technology,the amount of information on the Internet is increasing rapidly.With the emergence of technologies such as Internet commerce,Internet of thought,social networking and self media,data transmission and storage are going on.In the face of the huge amount of data and storages,cloud computing emerges as the time requires.As a new computing model,cloud computing has caught more and more attention recently.With the efficiency and convenience,there are many problems,energy consumption is too large,the resource utilization efficiency is not high,etc.It is based on this consideration,a lot of researches are carried out about network intensive applications in the cloud,find the applications some common characteristics,and based on the characteristics,we establish mathematical model of network intensive.Based on the mathematical model,the efficiency enhanced scheduling strategy of network intensive applications is proposed.The main contributions of this paper are as follows:First,the cloud computing network intensive applications have been done a lot of analysis,find out some common features of network intensive applications,including CPU utilization rate,the number of context switches,memory usage,IO usage and network load,etc.Based on these parameters,this paper proposes a network intensive application five tuple model,which provides a basis for further research on this specific type of processing strategy.Second,on the basis of the above mentioned model of network intensive applications,this paper puts forward the corresponding scheduling strategies to improve the efficiency.When a number of CPU and IO intensive applications are executed concurrently,the traditional resource allocation strategy does not take the characteristics of the applications into account,results in low resource utilization,low applications execution efficiency.In view of this situation,this paper analyzes the IO intensive applications and network intensive applications,according to their characteristics,they can be quantified,and puts forward scheduling strategy of IO and network intensive applications based on the priority.Improve the priority of small applications,to get larger CPU time slices,so that small applications can complete as soon as possible,and then all the CPU time slices are allocated to large applications,to reduce the process of switching between scheduling to improve the efficiency.In addition,the paper also puts forward the pipeline packing transmission strategy to further improve the performance of CPU.Third,this paper makes a lot of experiments on the network intensive application model and the corresponding model scheduling strategy.The experimental results have achieved expected goals,the accuracy of the model is proved,and the scheduling strategy proposed in this paper can effectively improve the efficiency of network intensive applications.