Skip to Main Content
Autonomic computer systems adapt themselves to cope with changes in the operating conditions and to meet the service-level agreements with a minimum of resources. Changes in operating conditions include hardware and software failures, load variation and variations in user interaction with the system. The self adaptation can be achieved by tuning the software, balancing the load or through hardware provisioning. This paper investigates a feed-forward adaptation scheme in which tuning and provisioning decisions are based on a dynamic predictive performance model of the system and the software. The model consists of a layered queuing network whose parameters are tuned by tracking the system with an Extended Kalman Filter. An optimization algorithm searches the system configuration space by using the predictive performance model to evaluate every configuration.