A Framework for Predicting Query Response Time at Application Development Stage

In a typical database application environment, emphasis has been given on promising Service level Agreements (SLAs) for perceived query elapsed response time; the SQL queries are tested on the small size of database at application developmentstage, which may be a fraction of the production database. As time progresses the database grows and the earlier optimized queries may not hold SLA anymore. Once the application is deployed, it becomes difficult to modify the application or alter the production system. In this paper, we have discussed a framework for predicting the SQL query response time with growth of the database while being transparent to the production hardware, storage subsystem and DB Server. We have discussed various factors which can impact query response time and are also affected by the increase in the data size. We have presented a theoretical model for predicting the elapsed response time of SQL queries and also discussed a case study of Oracle 10g for implementing the proposed framework.