Run-time Models for Self-managing Systems and Applications

The book topics focus on the adoption of run-time models in autonomic systems. Techniques and tools to estimate the performance of autonomic systems in transient conditions and at very fine-grained time scales will be investigated. Traditionally, performance models have a central role in the design, capacity planning, and management of autonomic systems. Models are used at design-time to support capacity planning of the physical infrastructure and to analyze the effects and trade-offs of different architectural choices, anticipating the discovery of potential bottlenecks which may degrade system performance. Models may also be used at run-time to assess the compliance of the running system with respect to the design-time model and to measure the real system performance parameters in order to fill the gap between design-time and run-time. Models at run time can also assess the compliance of service level agreements and trigger the run-time re-configuration of autonomic systems. Run-time models are receiving great interest from the research community, since, e.g., power management of modern CPU and resource allocation techniques in virtualized autonomic service centers can be actuated at very fine grain time scales. In such situations, traditional performance modeling techniques which evaluate the steady state of a system provide only a rough estimate of system behavior and are not effective to react to workload fluctuations.