Distributed Algorithms

Lecturer s : Alexandros Koliousis (homepage) , Alexander Wolf (homepage)
For course notes click on the lecturers homepages.

Models of distributed computing
Synchrony, communication and failure concerns
Synchronous message-passing distributed systems
Algorithms in systems with no failures - Leader Election and Breadth-First Search algorithms
The atomic commit problem
Consensus problems - the Byzantine Generals Problem
Asynchronous message-passing distributed systems
Logical time and global system snapshots
Impossibility of consensus
Fault-tolerant broadcasts
Partially synchronous message-passing distributed systems
Failure detectors

The Labelled Transition System Analyser (LTSA) tool http://www.doc.ic.ac.uk/~jnm/book/ltsa-v2/index.html is used throughout the course for modelling and demonstrating the execution of various algorithms.

Learning Outcomes:

By the end of this course students will be able to understand and explain the concepts behind distributed algorithms, including the assumptions made and the potential benefits and shortcomings. Students will also be able to assess the applicability of distributed algorithms to a particular circumstance.