Adopting Background Tasks in Serverless Web Applications

Abstract

In recent years, serverless computing has gained much popularity in software development among developers. It has brought many benefits to the industry. From the developer's point of view, it has removed the burden of maintaining infrastructure with many repetitive tasks. With automated infrastructure and less maintenance, developers have an opportunity to redirect their actions toward the implementation of logic. However, it has not been smooth sailing since serverless has also introduced a new set of challenges that were not present in traditional server-based web development. These challenges include limitations like timeouts, necessitating the adoption of new design patterns to overcome them. Another interesting aspect is that serverless computing is constantly evolving and these limitations are also changing over time as cloud vendor's services are becoming more flexible. New runtime environments, like managed container orchestration services, are not incapacitated from the limitations present in the older Function-as-a-Service platforms.

The research goal of this thesis is to explore the present status of serverless computing, focusing on its limitations and possible mitigation strategies. The research is limited to the context of background tasks execution in web applications. The research methods used include a literature review on the subject and the development of prototypes for background tasks using Amazon Web Services and Google Cloud Platform environments. These prototypes aim to validate the potential findings from the literature review. The prototypes are also validated through interview with experts of the commissioner.