A Control Theory Foundation for Self-Managing Computing Systems

The high cost of operating large computing installations has motivated a broad interest in reducing the need for human intervention by making systems self-managing. This paper explores the extent to which control theory can provide an architectural and analytic foundation for building self-managing systems. Using the IBM autonomic computing architecture as a starting point, we show the correspondence between the elements of autonomic systems and those in control systems. For example, the sensors and effectors of the autonomic architecture provide the measured outputs and control inputs in control systems. The benefit of making this connection is that control theory provides a rich set of methodologies for building automated self-diagnosis and self-repair systems with properties such as stability, short settling times, and accurate regulation. This said, there remain considerable challenges in applying control theory to computing systems, such as developing effective resource models, handling sensor delays, and addressing lead times in effector actions. To illustrate these challenges, we present a case study in which control theory has been used in an IBM database management product. We believe that addressing these challenges requires a broader engagement of the research community. To this end, we propose a deployable testbed for autonomic computing (DTAC) that we believe will reduce the barriers to addressing key research problems in autonomic computing. The initial DTAC architecture is described along with several problems that it can be used to investigate.