DCR Tools A quick tour of tools for Dynamic Condition Response graphs Thomas T. Hildebrandt & Søren Debois IT University of Copenhagen (and joint work with R. Mukkamala, T. Slaats, M. Marquard, F. Zanitti) Dagstuhl Seminar 17051 Theory and Applications of Behavioural Types February 1st, 2017 IT UNIVERSITY OF COPENHAGEN
IT systems increasingly control and support critical processes & interactions between humans and machines vebido.de Thomas T. Hildebrandt IT UNIVERSITY OF
IT systems increasingly control and support critical processes & interactions between humans and machines vebido.de Run-time adaptability, Correctness & compliance with legal, safety and security regulations is getting more and more critical! Thomas T. Hildebrandt IT UNIVERSITY OF
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Key points of this talk Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Key points of this talk • Correctness and security: Need for formalisation of processes & data-flow involving both humans and IT Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Key points of this talk • Correctness and security: Need for formalisation of processes & data-flow involving both humans and IT • Effectiveness, maintainability & compliance: Need for “intelligent”, flexible & adaptable IT systems Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Key points of this talk • Correctness and security: Need for formalisation of processes & data-flow involving both humans and IT • Effectiveness, maintainability & compliance: Need for “intelligent”, flexible & adaptable IT systems • The Dynamic Condition Response (DCR) Graphs process technology is a promising approach developed jointly by industry and academia and supported by tools DCRGraphs.net & dcr.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies condition Rule 1: The Seller must have made an offer before the Buyer can buy Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies response condition Rule 1: The Seller must have made an offer before the Buyer can buy Rule 2: The Seller must eventually make an offer if the Buyer ask Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies response condition event Rule 1: The Seller must have made an offer before the Buyer can buy Rule 2: The Seller must eventually make an offer if the Buyer ask Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies response condition event activity Rule 1: The Seller must have made an offer before the Buyer can buy Rule 2: The Seller must eventually make an offer if the Buyer ask Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies role response condition event activity Rule 1: The Seller must have made an offer before the Buyer can buy Rule 2: The Seller must eventually make an offer if the Buyer ask Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What are DCR Graphs ? Dynamic Condition Response (DCR) graphs are declarative run-time adaptable choreographies role response condition event activity Rule 1: The Seller must have made an offer before the Buyer can buy Rule 2: The Seller must eventually make an offer if the Buyer ask Let’s us look at the example in DCR.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Parser Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Visualiser Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 What was the point? • The DCR graph language has an operational semantics with state given as a marking of events with three booleans (executed,pending,included) • We do not need to unfold the entire transition system! • Distinguish may (enabled) & must (pending) eventually • Flexibility in execution • Conflicts ? • Adaptable ? Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Merging Choreographies Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Merged Choreography Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed DCR Graphs Eventually is often not good enough…. Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed DCR Graphs Eventually is often not good enough…. and delays may be required Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed DCR Graphs Eventually is often not good enough…. and delays may be required Timed DCR Graphs introduce [JLAP82,2013, delays on conditions, and deadlines on responses CSF2016] Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed choreographies DCR.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed choreographies response deadline DCR.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed execution Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Advancing time by 7 Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed choreographies: Delays DCR.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Timed choreographies: Delays delayed conditions DCR.tools Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Sub processes Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Sub processes Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Sub processes Infinite state system, and indeed reachability and refinement becomes undecidable [FM 2015] Look for decidable approximations! Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Enforceability & Escalation Some events are uncontrollable in particular progress of time and human activities Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
A quick tour of tools for Dynamic Condition Response (DCR) graphs February 1st, 2017 Enforceability & Escalation Some events are uncontrollable in particular progress of time and human activities Need compensation/escalation & pro-active enforcement [CSF2016] Thomas T. Hildebrandt (hilde@itu.dk) IT UNIVERSITY OF COPENHAGEN
Recommend
More recommend
