Architecture and performance evaluation of distributed computation offloading in edge computing

Abstract

Edge computing has been proposed to cope with the challenging requirements of future applications, like mobile augmented reality, since it shortens significantly the distance, hence the latency, between the end users and the processing servers. On the other hand, serverless computing is emerging among cloud technologies to respond to the need of highly scalable event-driven execution of stateless tasks. In this paper, we first investigate the convergence of the two to enable very low-latency execution of short-lived stateless tasks, whose computation is offloaded from the user terminal to servers hosted by or close to edge devices. We tackle in particular the research challenge of selecting the best executor, based on real-time measurements and simple, yet effective, prediction algorithms. Second, we propose a performance evaluation framework specifically designed for an accurate assessment of algorithms and protocols in edge computing environments, where the nodes may have very heterogeneous networking and processing capabilities. The proposed framework relies on the use of real components on lightweight virtualization mixed with simulated computation and is well-suited to the analysis of several applications and network environments. Using our framework, we evaluate our proposed architecture and algorithms in small-and large-scale edge computing scenarios, showing that our solution achieves similar or better delay performance than a centralized solution, with far less network utilization.

A Decentralized Framework for Serverless Edge Computing in the Internet of Things

“…We have used our own performance evaluation framework with real applications running in Linux namespaces interconnected via a network emulated with Mininet 6 . The interested reader may find full details in [10], which also describes the details of the implementation of the e-routers and e-controller. As executors we have used both OpenWhisk (as in [4]) running in Docker containers 7 and emulated e-computers, which provide responses to lambda requests based on the simulation of processing of tasks in a multi-processing system, also illustrated in [10].…”

Dynamic Edge and Cloud Service Integration for Industrial IoT and Production Monitoring Applications of Industrial Cyber-Physical Systems

“…Offloading computation to the edge has been studied in [9] and [8], and edge computing and related paradigms have been surveyed by [10], [14], [28] among others. In [29] the authors propose a method for the dynamic management of service resources for cloud, fog and edge based CPPS.…”

“…The challenge is to analyze data efficiently when striving for data-driven operations [1], [7]. In cloud computing, transmission of data, allocation of processing resources, and finally making results available typically includes delays and transfer costs [8], [9].…”

Self-Serviced IoT: Practical and Private IoT Computation Offloading with Full User Control

“…Serverless Offloading. Several research efforts have also explored how to leverage serverless computing to augment local devices' limitations [8,9,28,50]. Our design choice of using the FaaS paradigm was partly inspired by DIY [50], which explores executing private web services such as E-mail on serverless platforms.…”