Research On Load Balancing Problem Of Complex Cloud Service System Based On Transaction Scheduling And Data Placement

As cloud computing is widely accepted,more and more traditional services to cloud services change.The cloud service system is deployed in the cloud,providing large-scale online applications of business services,such as the Alibaba Taobao,in accordance with the idea of cloud computing software as a service.Cloud services are characterized by on-demand services and on-demand extensions.The first characteristics of large cloud service system are users large scale,request changes varied;then large data size,data storage multi-center distribution.In order to meet the needs of users,cloud services system set aside a lot of free resources,which makes the system utilization is very low.Through in-depth analysis,cloud service system load balance is particularly important.Based on transaction scheduling and data placement of the complex cloud service system load balancing problem refers to the complex cloud service system,how the cloud service system based on the workload and user needs,on the one hand to consider the high concurrent transaction balance problem,to achieve high concurrent affairs Scheduling,on the other hand,consider the complex features such as the relevance,diversity and real-time of heterogeneous multi-source data.In distributed data storage environment,dynamic data distribution is realized,and the interrelated data are agile and flexible,Real-time control of data placed to meet the diverse needs of users of high quality.Load balancing technology in cloud service system has become one of the key research problems in current and future computer field.The problem of transaction scheduling and data placement in complex cloud service system affects the performance and reliability of the whole system.Therefore,this.paper studies the load balance of complex cloud service system based on transaction scheduling and data placement.In this paper,through the in-depth analysis,cloud service system has two important factors affecting the imbalance and data storage is not balanced.Due to the complex situation of the transaction and the location of the data makes the cloud service system load imbalance situation,which greatly affected the user experience.Therefore,this paper proposes two methods of transaction scheduling and data placement to implement the load balance of complex cloud services.The main contents of this paper are as follows:1.Transaction scheduling method of load balancing in complex cloud service system.In the transaction scheduling method,the complex cloud service system due to the existence of complex transactions,cloud service system performance,such as response time,will have a lot of fluctuations,which greatly affected the user experience.In this paper,the causes of cloud service system performance fluctuation are analyzed in detail,and a scheduling method of complex transaction is given to smooth the fluctuation of cloud service system performance.2.The method of placing data in load balancing of complex cloud service system.In the data placement method,the data in the complex cloud service system is not put on a server node,and distributed data storage is a feature of the cloud service system.The strategy of data placement will directly affect the performance of cloud service system.This paper focuses on the problem of multi-dimensional data placement in cloud service system,and proposes a method of data placement.Previous work focused on the indexing of multidimensional data in the cloud database,which was designed to provide quick lookup services.The purpose of this paper is different,this paper aims to promote the relevance of data to facilitate the retrieval process.This paper presents a costing method for data packetization and proposes a new method to get the approximate optimal data grouping strategy.This paper builds a load balancing experiment environment of cloud service system,which is used to verify the load balancing method of cloud service system proposed in this paper.In this paper,we first verify the existence of the response time fluctuation phenomenon caused by the complex transaction and verify the effectiveness of the scheduling algorithm for the complex transaction of the gentle response time fluctuation proposed by this experiment.In addition,this paper presents a data placement method,and also through experiments to prove that the data placement algorithm is effective.The above two aspects demonstrate the accuracy of the framework of the load balancing method proposed in this paper.