Brief Introduction to Stochastic Petri Nets
Stochastic Petri Nets are a formalism developed in the field of computer science for modeling system performance. Software has been written to implement and solve models defined using SPN formalism.
Definition of SPNs:
SPNs consist of places and transitions as well as a number of functions. The basic functions are input, output and weight functions. The initial state of the system is represented by the initial marking. SPNs can be represented graphically, with places represented as circles and transitions as rectangles, and input and output functions as directed arcs.
An example of an SPN might be a computer system consisting of two processers which receive jobs from a common buffer. This systems might have 3 'places' (buffer, processer 1, processor 2) and 5 transitions (jobs arriving in buffer, jobs being transferred to one of the processors, jobs being completed by processors). The graphical representation of this SPN would be as follows:
Properties of SPNs:
SPNs have discrete state spaces, defined by the number of objects in each place (the marking). Places can be linked to transitions as input places, and transitions can be linked to output places. Transitions are said to be enabled when there are enough objects in each of the input places. Enabled transitions can fire, removing objects from their input places and adding objects to their output places. Enabled transitions fire according to exponential distributions, characteristic of Markov Processes.
Types of solution available for SPN models:
Last updated on June, 23,97