Developing Distributed Programs For The Cloud-Edge Computing Combining Multitier And Aggregate Programming

In the digital era, the increasing number of interconnected devices caused by the advent of Internet of Things and cyber-physical systems, has made distributed systems more and more pervasive as well as difficult to develop. This has brought about new challenges in terms of complexity and heterogeneity in the infrastructure on which these systems are deployed. For instance, in a scenario where the devices are distributed across the cloud and the edge. To tackle this complexity, two programming paradigms emerged in the last years: aggregate and multitier programming. The former is a paradigm that addresses the development of large-scale distributed systems by considering the system as a collection of devices (i.e. aggregates). Those devices can be programmed by using functions composition. The latter is a paradigm that allows the development of distributed systems by abstracting from the communication layer and developing every component of a distributed system in a single code base. The integration of these two paradigms can be beneficial for the development of the so-called Cloud-Edge continuum since aggregate computing can be used to program the logic of the devices at the edge, and the multitier paradigm can be used to address specific nodes in the cloud as well as easily deploy the system's components. The work described in this thesis aims to investigate the integration of the two programming models and to provide a middleware that uses the aggregate programming language ScaFi and the multitier programming language ScalaLoci to allow the development of distributed systems that can be deployed on the cloud, the edge, or a combination of both.