Generalized Stochastic Petri Nets (GSPN) are a prominent tool for the performance analysis of many real systems, but are prone to a state space explosion phenomenon that limits the applicability of the methodology. In this paper we describe CM-2 based algorithms dealing with the most computing-intensive part of the solution of GSPN models. Their implementation is fully integrated with a well-known tool for GSPN analysis (GreatSPN) and allows the solution of nets with up to several 106 states. The efficiency of the approach varies with the structure of the model under analysis.
Experiences on SIMD massively parallel GSPN analysis / Caselli, S.; Conte, G.; Bonardi, F.; Fontanesi, M.. - 794:(1994), pp. 266-283. (Intervento presentato al convegno 7th International Conference on Modelling Techniques and Tools for Computer Performance Evaluation, 1994 tenutosi a aut nel 1994) [10.1007/3-540-58021-2_15].
Experiences on SIMD massively parallel GSPN analysis
Caselli S.
;Conte G.;
1994-01-01
Abstract
Generalized Stochastic Petri Nets (GSPN) are a prominent tool for the performance analysis of many real systems, but are prone to a state space explosion phenomenon that limits the applicability of the methodology. In this paper we describe CM-2 based algorithms dealing with the most computing-intensive part of the solution of GSPN models. Their implementation is fully integrated with a well-known tool for GSPN analysis (GreatSPN) and allows the solution of nets with up to several 106 states. The efficiency of the approach varies with the structure of the model under analysis.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
