Applicability of serverless computing in fog computing environments for IoT scenarios

Abstract

With new paradigms to deliver software, a programmer’s utopia is close to becoming reality, where she only focuses on the realization of an application without messing around with infrastructural limitations and deployment considerations. Currently, this vision is supported by a paradigm shift of cloud providers’ service models, where new abstraction layers enable in particular serverless computing. Besides, the Internet of Things (IoT) requires a shift from the cloud paradigms to a fog computing perspective, where the functionality of a system needs to be allocated in the cloud-to-fog-continuum. In this regard, we analyze the applicability of a Function as a Service (FaaS) framework on an IoT service platform - SensIoT, which actually monitors environmental factors. Additionally, we deliver functions to cheap, energy-efficient Single Board Computers, which nowadays rapidly emerge as nodes of the IoT. We evaluate our approach by analyzing the resource usages of a FaaS enabled SensIoT and give an outline whether the combination of serverless computing, fog computing, and the IoT is going to enable the era of cloudless computing.