Concurrency and Distributed Systems

The language of Communicating Sequential Processes (CSP) is a powerful tool for understanding concurrent behaviour. It models parallel execution and inter-process communication, it supports a practical theory of correctness and refinement, and it is attractive and intuitive to use in specification.

Course dates

Objectives

At the end of the course, students will have a deeper understanding of concurrency. They will be better equipped to design and develop distributed systems.

Contents

Events and processes:

models of concurrency; representing communication and interaction; components and interfaces; behavioural and state-based specification; levels of abstraction.

Interaction:

state machines; internal and external choice; nondeterminism; safety and liveness constraints; algebraic properties; traces and refusals; process refinement.

Concurrency:

parallel combination; specifying interfaces; static alphabets; refinement in context; dynamic process creation and mobile code.

Abstraction:

changing levels of abstraction; mechanisms: interleaving, renaming, and hiding; abstraction and nondeterminism; encapsulation and refinement.

Communication:

shared channels; synchronous and asynchronous communication; modelling values and state; input and output; indexing and replication; data independence.

Sequential composition:

distributed termination; sequential composition of parallel components; abstract data types; processes and state; polling, interrupts, and timeouts.

Applications:

case studies and exercises, including: communication protocols; critical systems; caching algorithms; objects, classes, and threads.

Requirements

There are no prerequisites for this course, although some familiarity with discrete mathematics would be useful: Software engineering mathematics would be an ideal preparation.