Critical Time Distributed Systems: Qualitative and Quantitative Analysis Based on Stochastic Timed Petri Nets

Critical time distributed systems cannot be considered from a time point of view as Markovian processes and classical analysis methods cannot then be used.We present for analysing such systems a methodology based on the concept of the worst case(s). Such methodology is implemented thanks to the so-called “Stochastic Timed Petri Net” which allows to combine different transition firing rules.As application example, we consider a distributed producer-consumer process (the production process is periodic with a jitter; the produced data have a validity time window; the consumption request process is periodic with a jitter).

[1]  Guy Juanole,et al.  Methodology for obtaining abstract views of state graphs labeled with probabilities and times. An example of application to a communication protocol , 1995, MASCOTS '95. Proceedings of the Third International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems.

[2]  Guy Juanole,et al.  Functional and Performance Analysis Using Extended Time Petri Nets , 1987, Petri Nets and Performance Models.

[3]  Robin Milner,et al.  A Calculus of Communicating Systems , 1980, Lecture Notes in Computer Science.

[4]  Joseph Sifakis,et al.  Automatic Verification Methods for Finite State Systems , 1989, Lecture Notes in Computer Science.

[5]  Marco Ajmone Marsan,et al.  On Petri Nets with Stochastic Timing , 1985, PNPM.