Performance evaluation of serverless edge frameworks

Abstract

The rapid development of the Internet of Things (IoT) resulted in a significant increase in connected devices producing a high amount of data. Edge computing was introduced to reduce the overall latency by bringing computing resources closer to the users, while at the same time, the serverless paradigm emerged to automate the scalability and resource management of cloud servers. Serverless edge computing is a concept that proposes the utilization of the serverless paradigm on edge servers. The emergence of serverless edge raises concerns about the performance and reliability of serverless frameworks operating on edge devices with limited resources. Due to their popularity, newer versions of those frameworks are intended for the edge. This thesis evaluates the performance of two of such frameworks—FaaSD and Lean OpenWhisk—on edge devices by benchmarking different categories of applications on serverless edge environments. The performance evaluation yields a few results: 1) FaaSD outperformed Lean OpenWhisk on IO-intensive tasks and had better stability; 2) Lean OpenWhisk showed a better performance for other types of tasks compared to FaaSD; and 3) FaaSD was easier to install and deploy than Lean OpenWhisk for ARM-based CPUs due to its native support for this architecture. The output of this research provides valuable guidance for the selection of a serverless edge framework for edge environments.

Date of Award	2023
Original language	English
Awarding Institution	Western Sydney University
Supervisor	Rodrigo Neves Calheiros (Supervisor) & Bahman Javadi Jahantigh (Supervisor)