This work proposes the integration of distributed computing concepts into BPMN process simulations to evaluate the scalability of distributed applications. The extension allows physical resources to be associated with each activity and dynamically calculates task durations based on hardware performance, supporting infrastructural design and evaluation as early as the design phase. The working approach is demonstrated by experimental results that evaluate the scalability of a federated learning application. Furthermore, the extended simulator exhibits an almost linear growth in execution times as the number of nodes varies, while maintaining a limited overhead even in complex multi-node configurations.
Simulation-Based Scalability Evaluation of Distributed Applications Using BPMN
Errico, Palma
;Venticinque, Salvatore;Aversa, Rocco
2026
Abstract
This work proposes the integration of distributed computing concepts into BPMN process simulations to evaluate the scalability of distributed applications. The extension allows physical resources to be associated with each activity and dynamically calculates task durations based on hardware performance, supporting infrastructural design and evaluation as early as the design phase. The working approach is demonstrated by experimental results that evaluate the scalability of a federated learning application. Furthermore, the extended simulator exhibits an almost linear growth in execution times as the number of nodes varies, while maintaining a limited overhead even in complex multi-node configurations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


