Performance Engineering based on UML & SPN's: A software performance tool

The increasing relevance of UML as a semi-formal modelling paradigm has entailed the need for an adjustment of the classical performance evaluation methods within the scope of the new working environment. Under these circumstances, a formal semantics for the UML language and a strong mathematical substratum are required in order to be able to compute performance estimates and validate logical properties in the first stages of the software life-cycle. We believe that stochastic Petri nets are specially suited for this aim. A compositional approach for the translation of several UML diagrams into analyzable Petri net models has therefore been considered in previous papers. Following this approach, we will focus here in the depiction of a model case study from the perspective of our new performance-oriented CASE tool