Feedback control resource management using a posteriori workload characterizations

Certain real-time applications must operate in highly dynamic environments, thereby precluding accurate characterization of the applications´ workloads by static models. Thus, guarantees of real-time performance based on a priori characterizations are not possible. However, potential benefits of a posteriori approaches are significant, including the ability to function correctly in dynamic environments (through adaptability to unforeseen conditions), and higher actual utilization of computing resources. In this paper, we consider a control theoretic framework that is appropriate for systems which experience large variations in workload. The goal is to manage a distributed collection of computing resources by continuously computing and assessing QoS and resource utilization metrics that are determined a posteriori. Potential benefits of developing a control theoretic framework for resource management include: reuse of the large body of results in control theory, focus on problem characteristics instead of control infrastructures, and stability analysis