The long-term ground state monitoring of the Manned spacecraft environmental control and life support system(ECLSS)is accumulating huge amount of measurement point data everyday,so how to manage and query the data of each measurement point efficiently is a key problem.Aiming at the inefficiency of traditional SQLServer database multi-table query logic based on union operation when the number of sub-tables in the database is relatively large,this paper designs an adaptive rowwise segmentation algorithm,which can select the optimal number of rowwise tangent queries according to different query time intervals,so as to make full use of the system's multithreading for parallel querying.The typical query scenario test proves that compared with the traditional query method,the designed query optimization method can obviously improve the query efficiency and significantly reduce the query time.