Data and Reference Semantic-Based Simulator of DB-Nets with the Use of Renew Tool

Abstract

Complexity of software systems is constantly growing, which is even more aggravated by concurrency of processes in systems, so modeling and validating such systems is necessary for detecting and eliminating failures. One of the most well-known formalisms for solving this problem is Petri nets and their extensions such as colored Petri nets, reference nets, and similar models. Many software systems use databases for storing persistent data. However, Petri nets and their mentioned extensions are not designed for modeling persistent data manipulation since these formalisms aim at modeling a control flow of the considered systems. DB-nets, as a novel formalism, aim to solve this problem by providing three following layers: (1) the control layer represented by a colored Petri net with extensions, (2) the data logic layer, which allows to retrieve and update the persistent data, and (3) the persistence layer representing a relational database for storing the persistent data. To date, there are no publicly available software tools that implement simulation of db-net models with reference semantics support. The paper presents a novel software tool for the db-nets simulation. The simulator is developed as a pure plugin for the Renew (Reference Net Workshop) software tool without modifying existing Renew source code. Such an approach for development of the simulator allows to reinforce existing Renew reference semantics. The SQLite embeddable relational DBMS is used as a base tool to implement the db-net persistence layer. In the paper, theoretical foundations and architecture of the developed simulator are described. The results of the work can be used in research projects that involve modeling complex software systems with persistent data for both academic and industry-oriented applications.