Query Execution Plan Cache Based On DM Embedded Database

With the development of embedded technology, more and more users need more effective management of the data. Embedded database is an effective data management tools. However, due to the limited resources of embedded database system, the efficiency of the system is of great concern to users.In order to improve the efficiency of the embedded database, we expand a new query optimization - Execution plan reuse mechanism on origin optimizer ,which based on in-depth study of virtual machines and embedded database query optimization mechanisms. There are the theorems of identical SQL statement reuse and similar SQL statement reuse included in the new query execution plan cache mechanism, and also the query execution plan hash key generation algorithm, buffer query execution plan reuse and file-level query execution plan reuse and so on.For the determine the reuse implementation plan, if it is exactly the same SQL statement, only need to do a simple comparison in ASCII; if it is similar SQL statement reuse, we need to handle the case, invalid characters, constant value with the segmentation algorithm. If an appropriate treatment plan existed in the buffer, reuse it directly.For the limited query cache size, lowly plan reuse rate, we design file-level cached plan mechanism, which is store the buffer plan to file, or load the file-level cached plan into memory buffer, which can improve the execution plan reuse. The loading and storage time are open and close the database, it will not bring additional burden to the system.For the identical SQL statements and similar SQL statements, the test result indicates that the query processing time is efficiently shortened; the experimental results show that the new reuse mechanism significantly improved the system efficiency.