6 August 2016
Reliability of Computer Systems
1. Concepts: common systems and reliable, fault-tolerant systems, overview on fault-tolerance techniques, structural system models
2. Measures and their interpretation: reliability, failure probability, availability, mean time to failure, mean time to repair
3. Models: Boolean models and their variants (fault-trees, reliability-block-diagrams), calculation of reliabilities based on component parameters and on the system structure
4. Models of systems with repair/reconfiguration: Markov models, simulative analysis of Markov models
Prof. Dr.-Ing. habil. Peter Sobe (Dresden)
Students and professionals in the field of informatics, computer science or electrical engineering,
Attendance requirements: Computer architecture, basics in combinatorics, probabilistic evaluation
This lecture introduces reliability issues of computer hardware, networks and distributed applications. The students obtain an overview of techniques to improve the reliability of systems (e.g. use of stand-by resources, failover, fault-tolerant network topologies and routing). Measures for evaluation of systems (failure probabilities, availability, probabilities of operation) are introduced and modeling techniques to obtain them from component parameters and from the system structure are explained.
A result of the lecture should be an intensified attention on reliability, the ability to assess risks and to interpret measures, and the ability to model systems and their structural and operational variants.
Credits are allocated, based on students workload (ECTS-system). On successful completion of the Summer School courses, a certificate, which confirms the ECTS-credits earned as well as the grade, is awarded. The certificate is signed by all lecturers and includes a detailed description of the content and number of the courses attended.
EUR 0: No fee
- Free accommodation (including breakfast) and lunch for all participants