Matej Pavlović

researcher at Near One

Distributed consensus, state machine replication, and Byzantine fault tolerance

 

Abstract

Starting from the basics of distributed computing, we explain the most commonly used abstractions and algorithms. We show why consensus is such an important and fundamental problem in this field, why it is impossible to achieve the way we would most like it, and what we do about it in practice. We explore how distributed computer systems can be made to tolerate failures of some of their components, what happens if malicious actors actively try to disrupt those systems, and when and how we can still guarantee correct operation.

No specific prior knowledge is required.

Course material

• Slides [pdf]

---