Research And Implementation On Query Optimization Based On Data Partition In Massive Data Management

With the rapid development of modern science and information technology, applications in network environment keep on growing. More and more applications present characteristics such as huge data amount and consistent writing data flow. The huge amount of data makes the traditional queries often inefficient. Based on the current soft-and-hardware environment, we propose a better solution for query optimization specialized in query request of massive data. This article's main jobs is:1. Compared and analyzed the advantage and disadvantage of the solution for queryoptimization of massive data, and the suitable situation of the solution.2. Propose a query optimization method for massive Data based on Multi-dimensionalData Partition. This method partitions the data in a table in multi-dimension, changes the storage mode and optimizes the query performance on massive databases.3. By implementing this method base on the parallel query middleware, we bring forwarda method of data partition, in order to reduce the scanning data size of query request and win the ability of the highest efficiency.4. Based on the method of Data Partition, further research on parallel query and queryoptimization for improve entity query performance of the whole system.5. Finally, we verify its feasibility and efficiency by experiment.