PEPA
PEPA
Performance Evaluation Process Algebra
PEPA
In memoriam: Robin Milner (1934--2010) PEPA Club
  • PEPA club will meet on Friday (13rd of September) at 11:00am in room 3.02 of the Informatics Forum. In this meeting, the speaker will be Daniel Reijsbergen (title and abstract here). The volunteer baker this week is Dimitris.
  • The schedule of upcoming talks can be found here. If you would like to give a talk please e-mail the current PEPA Club organiser Anastasis Georgoulas.
  • PEPA Club usually meets on Fridays at 11:00 in Room 3.02 of the Informatics Forum at the University of Edinburgh.
  • Anyone can join the PEPA Club mailing list.

pyPEPA Software News: Spring 2013
  • A new PEPA implementation project has just released its first beta version. The pyPEPA project is creating an open-source implementation of PEPA in the Python programming language.
  • The pyPEPA implementation contains a library for PEPA, libpepa, a command-line tool for graphing and solving PEPA models, and support for map/reduce distributed computation of PEPA models on clusters.
  • The pyPEPA project is being led by Dariusz Dwornikowski of the Poznań University of Technology.
  • Project code and documentation are available from https://github.com/tdi/pyPEPA.

QUANTICOL

QUANTICOL

  • The main objective of the QUANTICOL project is the development of an innovative formal design framework that provides a specification language for collective adaptive systems (CAS) and a large variety of tool-supported, scalable analysis and verification techniques..
  • In this project, PEPA and its extensions have been used to construct formal models of the bus transport systems to study their resource provisioning.
  • The project home page is here

XXL The XXL Project
  • The XXL project (Advanced Tools and Techniques for the Specification and Verification of Systems with Elevated Granularity) has used PEPA to model emergency egress where people need to evacuate from buildings quickly in case of fire or other emergencies.
  • Models were built in PEPA and evaluated using the PEPA Eclipse Plug-in, making use of its ability to analyse systems with very large state-spaces.
  • More details on the work appear in the papers Reconciling Population and Agent Models for Crowd Dynamics and A Scalable Fluid Flow Process Algebraic Approach to Emergency Egress Analysis.
  • The PEPA models are available on-line here.

SENSORIA

SENSORIA

  • The development of the PEPA Eclipse Plug-in was supported by the four-year European project, SENSORIA (Software Engineering for Service-Oriented Overlay Computers).
  • The SENSORIA project provided funding for the development of the PEPA Eclipse Plug-in and the XSP query language for PEPA, but it also provided the significant technical challenge of modelling service-oriented computing, large-scale systems where computing resources are bound at run-time.
  • This provides a number of challenges for quantitative modelling which we addressed through SRMC (The SENSORIA Reference Markovian Calculus), a structured process calculus for service-oriented computing which compiles to a set of PEPA models.

PEPA Eclipse Plug-in

Software

  • The PEPA Eclipse Plug-in is a development environment for PEPA, supporting Markovian analysis, continuous-space analysis and simulation. The PEPA Eclipse Plug-in is available for free download from the download site.
  • For passage-time analysis the International PEPA Compiler provides a sophisticated query language for probing PEPA models to obtain response-time distributions along passages of interest through the model's behaviour. The International PEPA Compiler is available for free download from the download site.
  • A LaTeX style file for PEPA is available (example). A Beamer LaTeX theme file for PEPA is available.
  • logopepa can be used on Linux systems to generate PEPA logos in a variety of graphics formats, see "logopepa --help" for details.

Informatics Forum Visit us
  • PEPA has a new home. The PEPA group have now moved to the Informatics Forum in Crichton Street, Edinburgh. The Informatics Forum is an exciting new venue for our work on stochastic process algebras. It is located in the heart of Edinburgh city. Take a video tour of the Forum.
  • Come and visit us! The PEPA group welcomes visitors interested in stochastic process algebra, performance modelling, simulation and other methods of quantitative analysis. We have plenty of office space in our beautiful new building and are happy to host visitors.
  • Are you interested in doing a PhD on PEPA? We always welcome applications from interested students. Please send email to Jane.Hillston at ed.ac.uk in the first instance.

Sir Michael Atiyah and Jane Hillston

Awards

  • Congratulations to Professor Jane Hillston, personal chair in quantitative modelling at The University of Edinburgh and inventor of the PEPA language, who was made a Fellow of the Royal Society of Edinburgh at their award ceremony on 30th April 2007.
  • Congratulations to Richard Hayden of Imperial College, London who was presented with the Microsoft Research Award for the Best Computational Science Student at the Science, Engineering & Technology Student of the Year awards held in Alexandra Palace London in September 2007. Richard won the award for his dissertation on PEPA, Addressing the state space explosion problem for PEPA models through fluid-flow approximation.

Selected recent papers

( ... all PEPA papers, ... PEPA bibliography )


PEPA, the one

New to PEPA?

The following papers provide an introduction to PEPA and the concepts involved.

About PEPA

Jane Hillston's PEPA is a stochastic process algebra which is used for modelling systems composed of concurrently active components which co-operate and share work. PEPA allows the modeller to study either behavioural or performance properties. This archive gives access to papers on PEPA, example models, and software tools which support the PEPA language. A LaTeX style file for PEPA is available (example). A TrueType font for PEPA is available. You can also search the site.

Pronunciation: PEPA is pronounced Pep-Ah, not Pea-Pah or Pay-Pah.

Contact addresses: Jane Hillston and Stephen Gilmore, {jeh,stg}@inf.ed.ac.uk.
    Laboratory for Foundations of Computer Science, The University of Edinburgh.
Last modified: Friday 29 March 2013