A software cybernetics approach to self-tuning performance of on-line transaction processing systems

Self-tuning performance of On-Line Transaction Processing (OLTP) Systems is a challenging and time-consuming task since multiple performance parameters are needed to be automatically configured in Database Management Systems (DBMSs). In this paper, we present a software cybernetics approach to self-tune the performance of DBMSs. A DBMS is designed with an adaptive control based on fuzzy logic such that it has the capability to control objects, i.e., the performance parameters, and update the controller itself, i.e., a set of fuzzy rules in our case. The principles and concepts in software cybernetics are applied to guide the synthesis of software controllers for monitoring and adapting system behaviors. Experimental results for On-Line Transaction Processing using TPC-C, a benchmark of the Transaction Processing Performance Council, show that the proposed method is feasible and effective.