Research On Query Optimization Method Of Database Based On Cache Strategy

Distributed databases realize the horizontal expansion of data storage by storing data in multiple storage nodes.At the same time,due to the high degree of parallelism brought about by multiple physical nodes,the performance has also been greatly improved.However,it is precisely because of the decentralized nature of data that in the process of data retrieval,distributed databases often need to collect and transmit data or calculation results scattered across multiple storage nodes to the same node(interactive node)for unified integration.And filtering.In the process of gathering data to interactive nodes,a large number of network transmissions are unavoidable,which brings delays.In the database performance analysis and research,it is found that the delay caused by data transmission can offset the performance improvement brought by distributed computing to a certain extent.Therefore,based on the Volcano optimizer model and the Cascade optimizer model,this thesis proposes a query optimizer model based on cache optimization rules,and two specific scenarios of cache optimization rules(long transaction query optimization rules and primary key query optimization rules).This model is used to Reduce unnecessary network transmission between distributed databases in the process of data retrieval and query.The optimizer model and two optimization rules designed in this thesis can optimize data caching while ensuring data consistency and data validity in distributed scenarios.The specific research content is as follows:(1)By analyzing the various data collection modes in the distributed database,the application scenarios of the caching strategy are reasonably selected.The content includes,but is not limited to,the underlying data collection process of specific scenarios such as data insertion and update scenarios in a transaction,and data primary key query in a transaction.(2)In the selected two application scenarios(long transaction query and primary key query),a cache optimization rule that can ensure data consistency in distributed scenarios is proposed.(3)By analyzing the design architecture of the Volcano optimizer model and the Cascade optimizer model,as well as the two cache optimization rules designed in the second point,a query optimizer model based on the cache optimization rules is proposed.In this thesis,the query optimizer model based on cache optimization rules is applied to the domestic distributed database system,and the performance analysis and comparison are carried out through the SYSBENCH stress test program.The test results show that the query optimizer model based on cache optimization rules can ensure data consistency and data correctness in distributed scenarios while having better query performance advantages.