Author: Andreas Dittrich Ph.D. Thesis, submitted to the Faculty of Mathematics and Natural Sciences, Humboldt University of Berlin, December 9, 2014. Oral defense March 20, 2015. Grade: magna cum laudae. Reviewers: Prof. Dr. Miroslaw Malek, Lugano, Switzerland Prof. Dr. Alexander Reinefeld, Berlin, Germany Prof. Dr. Jörg Kaiser, Magdeburg, Germany Download: published version, meta information

Dependability of service provision is one of the primary goals in modern networks. Since providers and clients are part of a connecting Information and Communications Technology (ICT) infrastructure, service dependability varies with the position of actors as the ICT devices needed for service provision change. We present two approaches to quantify user-perceived service dependability. The first is a model-driven approach to calculate instantaneous service availability. Using input models of the service, the infrastructure and a mapping between the two to describe actors of service communication, availability models are automatically created by a series of model to model transformations. The feasibility of the approach is demonstrated using exemplary services in the network of University of Lugano, Switzerland. The second approach aims at the responsiveness of the service discovery layer, the probability to find service instances within a deadline even in the presence of faults, and is the main part of this thesis. We present a hierarchy of stochastic models to calculate user-perceived responsiveness based on monitoring data from the routing layer. Extensive series of experiments have been run on the Distributed Embedded Systems (DES) wireless testbed at Freie Universität Berlin. They serve both to demonstrate the shortcomings of current discovery protocols in modern dynamic networks and to validate the presented stochastic models. Both approaches demonstrate that the dependability of service provision indeed differs considerably depending on the position of service clients and providers, even in highly reliable wired networks. The two approaches enable optimization of service networks with respect to known or predicted usage patterns. Furthermore, they anticipate novel service dependability models which combine service discovery, timeliness, placement and usage, areas that until now have been treated to a large extent separately.

	Authors: Andreas Dittrich Rafael Rezende 14th European Workshop on Dependable Computing (EWDC), Coimbra, Portugal, May 15-16, 2013 Download: accepted version, the final publication is available at link.springer.com

Service-oriented architecture (SOA) has emerged as an approach to master growing system complexity by proposing services as basic building elements of system design. However, it remains difficult to evaluate dependability of such distributed and heterogeneous functionality as it depends highly on the properties of the enabling information and communications technology (ICT) infrastructure. Moreover, every specific pair service client and provider can utilize different ICT components, constituting for the user-perceived view of a service. We provide a model-driven methodology to automatically create reliability block diagrams of such views. Given a service description, a network topology model and a pair service client and provider, it identifies relevant ICT components and generates a user-perceived service availability model (UPSAM). We then use this UPSAM to calculate the steady-state availability of different views on an exemplary mail service deployed in the network infrastructure of University of Lugano, Switzerland.

	Authors: Andreas Dittrich Jon Kowal Miroslaw Malek International Workshop on Dependable Network Computing and Mobile Systems, DNCMS 2008, in conjunction with IEEE SRDS 2008, Naples, Italy, October 2012 Download full paper

Service-oriented architectures focus mainly on the automatic configuration of the attributes that describe the different layers involved in service communication and treat service instances monolithically – they either exist in the network which means that they are fully usable or they do not. This approach does not work well in environments where services are insufficiently dependable and the types of services used are not well known or standardized. This paper proposes a model to compose complex services from independent components with basic functionality that are organized as minimal services in the same service-oriented architecture. The approach promises to better handle run-time diagnostics and on-the-fly (re-)composition of service functionality in networks with highly dynamic capabilities.

	Autor: Andreas Dittrich Exposé zur Dissertation HU Berlin Paper: (mail)

Herkömmliche, dienstbasierte Architekturen gehen von idealisierten Bedingungen aus, innerhalb derer sie die Konfiguration der verschiedenen Schichten der Dienstnutzung zuverlässig automatisieren. In diesem Exposé wird das Konzept der Survivability eines serviceorientierten Systems diskutiert, damit es in Umgebungen überlebensfähig und vorhersagbar bleibt, in denen Ressourcen nicht nur stark begrenzt sind, sondern auch unzuverlässig zur Verfügung stehen. Dies ist besonders in Katastrophenszenarien der Fall. Der hier beschriebene Ansatz basiert darauf, dass ein System durch die permanente Überwachung des eigenes Zustands Self-Awareness erreicht, aufgrund dieser Erkenntnis intelligente Anpassungen vornehmen und mittels definierter Szenarios vorausschauend handeln kann. Die geeigneten Modelle, Metriken und Simulationen für die vollständige Erforschung der Problemstellung sollen innerhalb des interdisziplinären Graduiertenkollegs METRIK im Rahmen einer Dissertation erfolgen.