Efficient Algorithms for Performance Management of Computing Systems∗

This paper addresses the problem of managing computing systems using an integration of model-based control tech-niques and efficient AI search strategies. The control ap-proach uses the system model to forecast all future system behavior up to a certain horizon and then searches for the best path for the system based on a given utility function. In practical computing systems, however, the large number of control (tuning) options directly affects the computational overhead of the control module which executes in the back-ground at run-time, and ultimately slows down the overall system. To handle this problem, several search algorithms are introduced to improve the controller’s performance. A case study of an online processor power management is used to demonstrate the effectiveness of the new search techniques for the model-based control approach