ScaFi: Integration and Performance Analysis with Scala Native

Aggregate Computing is an emerging paradigm for complex distributed systems where a vast number of distributed devices are involved in a global computation and must cooperate to produce a collective result. This situation is common in the Internet of Things, large-scale urban events, drone coordination and smart cities. Modern Aggregate Computing APIs are normally based on the Field Calculus that offers the basis for the global-to-local computation abstraction, providing Computational Fields. Moreover, these APIs also rely on abstraction layers that hide the complexity of the environment from the sight of the developer (complexity "hidden under the hood"), offering a simple and friendly way to develop this kind of applications. An Internal Domain-specific language that offers these features is Scala with Computational Fields (ScaFi), a Scala framework implementing aggregate programming mechanisms. A critical concept for these types of libraries is portability since their nature implies the possibility of being run over a wide range of different devices. The work shown in this thesis offers a solution to improve the portability and flexibility of ScaFi integrating Scala Native, a Scala ahead-of-time compiler that makes it possible to directly compile Scala code over devices that do not support the JVM (enabling the so-called Cross-compilation). Cross-compilation between different platforms is a very desirable feature for a programming language because it makes the language much more flexible. For this reason, it is often included in many modern languages such as Kotlin and Rust. To conclude, several tests are done to validate the stability and the performance of the integration and in order to prove that the implementation proposed can efficiently extend the number of devices on which ScaFi can be run.