Communicating Process Architectures

In 2000 we officially took the name of Communicating Process Architectures (CPA) but retained an affiliation with our founding organization, namely, the World Transputer User Group (WoTUG), which dates back to 1984 when the first technical meeting was held in Bristol. We have since 1984 met at technical meetings or conferences a total of 39 times - 2018 will be our 40th anniversary conference. Our main focus is concurrency and formal verification - all kinds of concurrency - but many papers have dealt with or been based on Sir Tony Hoare's Communicating Sequential Processes (CSP). There has been many papers on languages or libraries with a foundation of CSP (examples include occam and occam-π, but also newer libraries like PyCSP). We have highly theoretical papers on semantics, about formal verification, high-performance computing (HPC), on the implementation of runtime systems or compilers, and many other aspects of concurrency.

Search our paper database or look through a link of resources of projects we have worked on or work with.

Latest News

Pictures and Awards

Conference pictures and CPA 2018 awards

 August 23, 2018

Pictures of the presentations, the fringes as well as the award winners.

CPA 2018 Dresden

CPA 2018 in Dresden has started

 August 20, 2018

CPA 2018 in Dresden as started at the Faculty of Computer science. This is our 40th anniversary conference.


CPA 2018 in Dresden announced

 October 14, 2017

We are happy to announce that the 40th anniversary conference will be held in late august of 2018 in Dresden at the Technical University.

CPA 2017 Awards

CPA 2017 Awards

 August 23, 2017

Congratulations to Uwe Mielke (Best Fringe Presentation), Antoon Boode (Best Student Paper), and Kevin Chalmers (Best Paper).

Our Work

KRoC and occam-π

The Kent Retargetable occam Compiler (KRoC) is an occam/occam-π language platform, comprised of an occam compiler, native-code translator and supporting run-time system.

occam-π and KRoC are the result of on-going research extending the (CSP) ideas of occam by judicious inclusion of the mobility features of the π-calculus. In the interests of proveability, we have been careful to preserve the distinction between the original simple static point-to-point synchronised communication of occam and the dynamic asynchronous multiplexed communication of the π-calculus; in this we have been prepared to sacrifice the elegant sparsity of the π-calculus. We conjecture that the extra complexity (and discipline) introduced will make the task of proving concurrent and distributed programs easier.

Go to GitHub Repository