Implementation of a serverless application

Serverless computing is an execution model where the cloud provider is responsible of provisioning and managing servers. Serverless applications can leverage several paradigms, like Function as a Service, in which code is executed in ephemeral stateless containers, or Backend as a Service, in which the server-side logic relies on third-party services, generally used for providing authentication, cloud storage, and more. Serverless is meant to help IT teams to build easily scalable architectures, economically advantageous and with low effort in maintenance. Furthermore, it allows organizations to reallocate time and people to problems unique to the product. It is a relatively new model, which still requires experience in order to enlarge the general background, to identify patterns, to build a real culture and this can happen thanks to the joint research of universities and companies. Absence of server management does not mean absence of operations, as claimed by the NoOps philosophy. Tasks like maintaining of infrastructures, collecting metrics, monitoring, and logging are still primary responsibilities. In this sense, serverless is a game-changer, because it requires a new way to look at DevOps automation practices, which tend to move up the technical stack and raise the abstraction level. The goal of this thesis is to design a complex serverless application, made of two independent areas, and which is able to integrate components that are specifically thought for serverless environment. Afterwards it will be integrated in a DevOps CI/CD pipeline. The final result is aimed to be part of a very large B2B project.