Research On Large Data Storage / Priority Scheduling Model Under Cloud Environment

With the continuous increase of business volume and supply in small and medium-sized enterprises,People's demand for big data storage and resource scheduling in e-commerce procurement system has become harder and harder.Enterprises need a scientific,sustainable and efficient management tools to assist management.The procurement management system based on cloud computing can just solve these problems.This paper analyzes the domestic and international reseach status on the big data storage and big data resource scheduling of cloud computing,Aim at the problems of the lowrate of big data storage and network delay,the paper proposes storage/priority scheduling model for big data under the cloud environment using technologies such as virtualization,cloud storage,and resource scheduling.According to the demands of current market and then combining with the model,the paper designs an e-commerce procurement management system based on cloud computing.The paper will study from the following three aspects.Firstly,build modeling combined with UML and describe the relevant constraint condition of resource class in resource pool.Then,the analysis and modeling design are performed.Secondly,aim at the problems of resource storage and scheduling in cloud environment,the paper builds storage/priority scheduling model for big data based on cloud environment.And it also shows the hierarchy of the model structure.The model can be divided to two parts.In terms of cloud storage,this paper focus on the issue of large batches of files being sent from the client to the server.This is usually done after all file packages are received.The package will continue to consume a lot of time and it will have problems of transmission efficiency.Therefore,this paper improves the transmission efficiency by optimizing the parallel processing algorithm.In terms of priority scheduling,the operational research's priority M/M/1 queuing theory model is used to perform network delay analysis on virtual machine requests and change the order of files.Based on Map-Reduce modeling,corresponding metrics of resource scheduling are introduced to improve the algorithm's performance under the cloud environment.At the cloud computing SaaS(software as a service)layer,an e-commerce procurement system based on cloud storage/priority scheduling model is designed and developed.And the module of procurement order management in the procurement management system is taken as a typical example to analyze the use case modeling and business flow chart in details.The procurement of business modules and database are designed by using Hadoop distributed processing framework.Thirdly,the design of cloud storage and application of scheduling algorithm are conducted based on Hadoop distributed framework.Applying Web Service technology and J2 EE technology can realize the research and development and application of e-commerce procurement system.The part of the expriment shows the relevant tests to the system in order to verify the practicality of the proposed model.Through the analysis of the results from the experimental test the system,the storage/prioritized scheduling model for big data under the cloud environment not only breaks through the constraints of traditional sequential transmission,but also reduces the storage coupling and greatly improves the load balance,speeds up the total time for task scheduling and makes the load efficiency of the scheduling time improved.It is of great significance to the continuous improvement and popularization of the e-commerce procurement system.