A Set of Performance and Dependability Analysis Components for CADP

Abstract

This paper describes a set of analysis components that open the way to perform performance and dependability analysis with the Cadp toolbox, originally designed for verifying the functional correctness of Lotos specifications. Three new tools (named Bcg Steady, Bcg Transient and Determinator) have been added to the toolbox. The approach taken fits well within the existing architecture of Cadp which doesn't need to be altered to enable performance evaluation.

CADP 2011: a toolbox for the construction and analysis of distributed processes

“…-Determinator [7] takes as input an extended Ctmc encoded in the Bcg format and tries to extract on-the-fly a pure Ctmc (i.e., containing only stochastic transitions). Doing so, the tool checks a sufficient condition ensuring that the resulting Ctmc is unique, or returns an error otherwise.…”

CADP 2006: A Toolbox for the Construction and Analysis of Distributed Processes

“…Besides Bcg Min, the Exp.Open tool [37] now supports also the parallel composition of extended Markovian models, implementing maximal progress of internal transitions in choice with stochastic transitions. New tools have been added, namely Determinator [32], which eliminates stochastic nondetermin-ism in extended Markovian models on the fly using a variant of the algorithm presented in [10], and the Bcg Steady and Bcg Transient tools, which compute, for each state s of an extended Markovian model, the probability of being in s either on the long run (i.e., in the "steady state") or at each time instant t in a discrete set provided by the user.…”

Ten Years of Performance Evaluation for Concurrent Systems Using CADP

“…A number of these tools (namely Bcg Steady, Bcg Transient, and Determinator [7]) have been developed specifically to support performance evaluation. Other tools already existed but were already compatible or have been extended to be compatible with the proposed approach.…”

“…The Determinator tool [7] eliminates stochastic nondeterminism in extended Markovian models on the fly. It takes as input an extended Markovian model M (encoded in the Bcg graph format) containing probabilistic and/or stochastic transitions and attempts at translating M to a Ctmc (Continuous Time Markov Chain), i.e., an Lts (encoded in the Bcg graph format) that contains (labeled) stochastic transitions only.…”