Course content

The course gives a broad introduction to the design and quantitative evaluation of dependability in systems based on information and communication technologies. The following issues are covered:
* Dependability attributes: availability, reliability, safety, QoS;
* The failure process, failure semantics, fault handling and maintenance;
* Methods for dependability modelling and analysis, with emphasis on block diagrams and state diagrams (Discrete state continuous time Markov models);
* Fault tolerance, types of redundancy, design principles and mechanisms, the properties of designs;
* Prediction of hardware failure rates;
* Modelling of software failure, prediction of software failure rates, impact of software failures on systems behaviour;
* Survivability of networks, dependability evaluation of networks taking the traffic carrying (QoS) requirements into account.

Learning outcome

A. Knowledge:
1) Terminology and concepts used in specification, design, operation and evaluation of dependable information and communication (ICT) systems. Overview of the ”entire” field of dependability in ICT systems.
2) The principles of fault tolerant/survivable design and operation of server systems and networks. Architectures and mechanisms used to achieve fault-tolerance/survivability/resilience/robustness.
3) Factors that influences HW and SW failure processes.
4) The relation between traffic handling and dependability in networks.

B. Skills:
1) Design of systems with appropriate use of fault tolerance/recovery mechanisms and redundancy to meet their dependability requirements.
2) To evaluate systems quantitatively with respect to availability and reliability by continuos time, discrete state Markov models and reliability block diagrams. Dependability dimensioning.
3) Design and implementation of a fault tolerant service in a distributed system using group communication middleware.

C. General competence:
1) Awareness of dependability as a important factor in the use and economics of systems; hereunder the potential impact of failures on commercial application, on the society, and on safety of humans and environment.

Learning methods and activities

Lectures. Exercises in modelling and evaluation. Laboratory project carry out small groups. If there is a re-sit examination, the examination form may be changed from written to oral.

Compulsory assignments

• Laboratory project/exercise

Recommended previous knowledge

Basic knowledge in ICT systems design. Basic knowledge in probability, Markov modelling and analysis, e.g. as covered by the course "TTM4110 Dependability and Performance with Discrete Event Simulation". Programming for lab. work.

Course materials

The material listed below is subject to change:
Bjarne E. Helvik, "An Introduction to the Design and Evaluation of Dependable Computing Systems and Communication Networks", Lecture notes published by TAPIR. Journal paper(s). Supplementary material for the laboratory project. Mathematica introductions and notebooks for dependability evaluation.

Credit reductions