Reference Models

This chapter describes four case studies of varying difficulty that are simplified versions of complex projects used to study real world problems. Despite their simplicity, the implemented models are still able to capture the key aspects of the phenomena analyzed. Initially, theIntranet models capacity planning of a facial recognition surveillance system based on edge computing architecture is described. Next, the design of a digital infrastructure that dynamically scales the computational capacity to handle workload fluctuations is presented. The implemented multi-formalism model consists of both Queueing Networks and Petri Nets components. The problem of performance forecast of a web app is tackled by simulating a concise version of the workflow of an e-commerce app. The impact of different authentication protocols for payment security is also considered. The last case study concerns the modelling of a platform for crowd computing. The originality of this model lies in the fact that its customers that flows in and out are the computational nodes, which are added and removed from the crowd platform. The resources that resides inside a Finite Capacity Region of the model manage the computational servers that may or may not be available to users.