Research And Implementation Of Data Access Optimization Technology In Cloud Platform

Now it has entered the era of cloud computing.Data access in cloud environment can not be separated from the support of database.A database system with excellent performance can greatly improve the speed of data access in cloud platform.At present,many databases used by cloud platforms are clustered,which is mainly due to the following reasons.First of all,database clusters often achieve load balancing.Secondly,the cluster contains more than one data node,which greatly improves the security of data.In addition,when a node in the cluster can not continue to perform tasks,it can be assigned to other nodes to execute the tasks and achieve failover.As can be seen,database cluster has many advantages.However,in the cloud environment,the amount of user access is changing in real time,and it is likely that there is a sharp increase in data access.Data access technology based on traditional database cluster can not solve the above phenomena very well.Redis has been widely used in recent years.Compared with disk,data access speed has been greatly improved.At the same time,the existing database cluster system load balancing and data synchronization and cache replacement strategies also have many drawbacks,and can not adapt to the characteristics of data access in the cloud environment.Therefore,through the research of existing related technologies,this paper finally realizes a database cluster system based on Redis and MySQL,and studies and optimizes the technology for data access,mainly including the optimization of centralized cache,reasonable load balancing,the speed of query results feedback caused by read-write separation,and the improvement of system reliability caused by dual-agent and fissure resolution.Optimize the performance of parallel data synchronization within the cluster.The system is divided into proxy subsystem and data access subsystem.Firstly,the agent subsystem contains the following four modules.Access access module realizes thread pool mechanism and speeds up the processing speed of user access requests.In the data cache module,a reasonable cache decision algorithm and a cache replacement algorithm are designed,which greatly improves the cache hit rate.Fissure resolution module,designed a reasonable voting algorithm,so that double agents can avoid the competition between agents and ensure the normal operation of the system.In the load balancing module,a performance quantization unit circle partition load balancing algorithm is designed.The weight distribution is more reasonable,and it has the ability of self-adjustment,which makes the load distribution more uniform.Secondly,the data access subsystem includes the following two modules.Data synchronization module realizes data parallel synchronization between cluster nodes based on performance linked list,which greatly speeds up the synchronization speed.Data access module achieves read-write separation and database connection pool mechanism,which further speeds up the query speed of the database.Through the cooperation of different modules,the resources of the whole system are fully utilized.On the basis of realizing high availability and high efficiency,the speed of data access is greatly accelerated,the reliability of the system is enhanced,the overall performance of the system is improved,and the purpose of data access optimization is realized.