Introduction
Services in that section are computed using GreatSPN,
a well known software package for the modeling, validation, and performance
evaluation of distributed systems using Generalized Stochastic Petri Nets
designed in the universities of Torino and Genova (Italy).
For the conputation of invariants on Colored Petri nets, it has been
coupled with our Colored Petri net unfolder.
The P/T net invariants menu
When the Petri net you have designed is syntactically correct, you can
reach the services (in red).
Invariants
|
|
Colored invariants
|
|
P-invariants
|
|
T-invariants
|
|
P-invariants by unfolding
|
|
Linear characterization (verbose)
|
|
Linear characterization
|
Display of results
All results in GreatSPN are displayed the same way. Objets in
the input Petri net are outlined and a floating window allows you to go
from one structural property to the other one.
Display of a P-invariant by GreatSPN in CPN-AMI.
For P-invariants and T-oinvariants, ponderations are shown as temporary
labels. In the previous figure, the Invariant has to be read : 1*B
+ 1*C + 1*G + 1*F + 1*D = constant.
The floatting window to go from one result to another.
The Floatting window allows the user to go from one result to an other
(for example, one P-invariant to another one). In the example above, there
are three inviariants. The third one is currently outlined in the model
window. No floatting widow appear if there are no result (for example,
no P-invariant in the Petri net model).
Please note that if the computation is tool long, you can cancel the
execution by clicking on the STOP button in the execution
window.
Display of Results
for Colored Nets
Results are displayed in the historic window. Let us have a look on the
typical philosopher example (model below).
The philosopher model.
The computation of P-invariants provide the following result. Please
note that name of original colored places is postfixed with the color
unfolding to increase readability.
Computation of P-invariants on the philosopher model.
|