SARD: A Statistical Approach for Ranking Database Tuning Parameters

Traditionally, DBMSs are shipped with hundreds of configuration parameters. To address a broad class of applications, such configuration parameters are set to default values. Since the database performance highly depends on the appropriate settings of the configuration parameters, DBAs spend a lot of their time and effort to find the best parameters values for tuning the performance of the application of interest. In many cases, they rely on their experiences and some rules of thumbs. However, as heuristics are used, time may be wasted by tuning those parameters which may have no or marginal effects.