An Evaluation of FaaS Platforms as a Foundation for Serverless Big Data Processing

Abstract

Function-as-a-Service (FaaS), offers a new alternative to operate cloud-based applications. FaaS platforms enable developers to define their application only through a set of service functions, relieving them of infrastructure management tasks, which are executed automatically by the platform. Since its introduction, FaaS has grown to support workloads beyond the lightweight use-cases it was originally intended for, and now serves as a viable paradigm for big data processing. However, several questions regarding FaaS platform quality are still unanswered. Specifically, the impact of automatic infrastructure management on serverless big data applications remains unexplored. In this paper, we propose a novel evaluation method (SIEM) to understand the impact of these tasks. For this purpose, we introduce new metrics to quantify quality in different big data application scenarios. We show an application of SIEM by evaluating the four major FaaS providers, and contribute results and new insights for FaaS-based big data processing.