Showing a Progress Bar While Executing Stored Procedures!

In this paper, we present a novel dynamic scheme to estimate the execution time of stored procedures inside the Database Management Systems (DBMS). Our estimation model can provide users an accurate estimation of the execution time of stored procedures. The proposed estimation model adopts the basic idea of Radial Basis Function (RBF) network to accurately predict the execution time of the stored procedures based on their input parameters. Our model is self-trained and does not need a separate training set. Moreover, it can automatically adjust itself to adapt to the changes of the database. Furthermore, our proposed model can be implemented by SQL to embed directly in the database which is an important advantage over previous systems. Extensive experimental results show that the proposed approach can accurately predict the execution time of stored procedures with error rate below 10% after getting executed for as least as 20 times. Moreover, the new model can effectively adjust itself to dramatic database updates.