Title: SQL Query Optimization on Cross Nodes for Distributed System
Abstract: Query terms are also expanded from several dimensions of simple query expansion to complex query of multi dimension. It is becoming more and more difficult to extract, store and analyze the massive data by using the traditional database software. So the better query strategy is the key to optimize the database query operation. As the important method of database analysis data, SQL query plays an important role in analyzing and processing data. Through SQL inquiries, users can get their most concerned information quickly. With the continuous development of the Internet industry, the data need to deal with in database continues to expand its scale. Database shows its characteristics as large amount, multiple types, fast processing speed and low density value. This paper proposed a new global adaptive optimization processing method in order to achieve the purpose of reducing the number of data I/O and load balance when SQL query is carried out in the distributed parallel system. Its characteristic is: to build a multi-factor decision fuzzy evaluation model for each sub-query path optimization decision and to definite the optimized global cost function for adaptive optimization on global query path which will satisfy the total cost minimum requirements for the purpose of query. The experiments show that the global query method has the faster query speed. At the same time, the total cost of query is controllable by defining the global optimized cost function.