Doctor's Theses (authored and supervised):

Ph. Leitner:
"On Preventing Violations of Service Level Agreements in Composed Services Using Self-Adaptation";
Supervisor, Reviewer: S. Dustdar, F. Casati; Institut für Informationssysteme, AB Verteilte Systeme, 2011; oral examination: 2011-11-09.

English abstract:
Providers of composite Web services face the challenge of having to comply to Service Level Agreements (SLAs), which are agreements governing the minimum performance that customers can expect from a composite service. SLAs contain Service Level Objectives (SLOs), concrete numerical Quality of Service (QoS) objectives, which the service needs to fulfill. If objectives are violated, agreed upon consequences (usually taking the form of penalty payments) go into
effect. However, fulfilling SLAs can also lead to costs for the service provider (e.g., because the composite service provider needs to use more expensive services itself, or because of the costs
inherent to optimizing the composition). Therefore, it is not trivial for the provider to decide to what extend the service´s SLAs should be fulfilled, or which SLAs should (temporarily) be violated for economical reasons. Even more so, these decisions should ideally be automated, to allow for fast reactions to changes in the business environment. In this thesis, a framework for optimizing adaptations of service compositions with regards to SLA violations has been developed. The framework, dubbed PREvent (Prediction and Prevention
of SLA Violations Based on Events), uses techniques from the areas of machine learning and heuristic optimization to construct models for prediction of SLA violations at runtime, and to decide which adaptation actions may be used to improve overall performance in a composition instance. An optimizer component decides, whether applying these changes makes sense economically (i.e., whether the costs of violating the SLAs are bigger than the adaptation costs).
If this is the case, the respective actions are applied in an automated way. At its core, Prediction and Prevention of SLA Violations Based on Events (PREvent) is a self-optimizing system in the sense of autonomic computing, with the target of minimizing the total costs of adaptations and SLA violations for the service provider. The outcomes of the thesis are explained and evaluated based on an illustrative case study. Results show that, using the PREvent system, providers of composite Web services are able to
reduce their total costs. Furthermore, for service clients, the advantage of PREvent is that the number of SLA violations is reduced. This in turn increases client satisfaction.

Created from the Publication Database of the Vienna University of Technology.