Key Technology Of Historical Data Query And Extraction For Multi-period Backup Database

Mold is a typical machinery manufacturing enterprise in accordance with the order of non-repetitive production, with the development of Mold enterprise, a large number of new business production orders are produced every year, which results in back-end database data surging its ERP system, the system is running slowly, which has great impaction on normal use of mold enterprise resource planning, product data management and shop floor execution layer and so on; and with the gradual increase of the amount of mold enterprise data, this situation will become more serious. Therefore, to explore a suitable mold enterprise data processing program is very necessary.To apart the historical databases, accelerating the system, this paper classify the data by the frequency of accessing the databases and the increasing pattern of data, based on the background of mould enterprises. Then this paper proposes the rules of the orders belong to the running data in the backup database, analyzes the factors impacting period of backup, and proposes the solution of data backup for the mould enterprises.Then this paper studies the inquiring historical data problem in the multiple-periods-backup databases. To improve the efficiency of inquiring historical databases, this paper proposes a theorem to judge the result of selection. If the result is null, this operation is done, or propose three rules based on the theorem. Finally, this paper proposes an algorithm based on the four rules, inquiring the historical databases, and designs an experiment to estimate the performance of the algorithm proposed in this paper.In addition, this paper proposes an solution on extracting the data in the multiple-periods-backup databases with complex distribution, base on the background of mould enterprises. Analyzes the situation under the change of dimension table and fact table when the index is added. Propose the extracting solution based on the algorithm of inquiring data, from the multiple-periods-backup databases when the index is added.Finally, this paper develops and designs a system with the functions of backup, apart, inquiring and extracting data, based on the algorithm proposed in this paper and software engineering, which verifies the feasibility of this paper.