The test data challenge for database-driven applications

Business applications rely typically on databases for storing and processing their data (database-driven applications, or DBAPs). Testing DBAPs requires testing the application logic plus the interaction between the application logic and the database. Thus, DBAP test cases consist of input and output parameter values, the function to be tested, and an initial database state (i.e., DBAP test data). Various test data provisioning methods exist, such as manual test data design, generators for synthetic test data, and live-system snapshots. Many criteria and factors influence which method is optimal for a given project setting, such as costs, quality, data privacy, etc. This paper presents our methodology for guiding software development projects towards the DBAP test data provisioning method best suited for them