Green Orchestration of Cloud-Edge Applications: State of the Art and Open Challenges

In recent years, Cloud-Edge computing continuum paradigms have received increasing attention to support the deployment of latency-critical and bandwidth-hungry applications enabled by the Internet-of-Things. Meanwhile, the dramatic in-crease in energy consumption of ICT calls for new methodologies to reduce the ecological impact of application orchestration, making it environmentally sustainable. This article conducts a systematic literature review of the state-of-the-art related to environmentally sustainable orchestration of containerised applications in Cloud-Edge settings. Ultimately, it points to open challenges and directions for future work in the field.

I. Introduction

In the last decade, Cloud computing has revolutionised the way applications are developed, deployed and delivered to their final users by relying on utility computing resources offered as a service on a pay-per-use billing scheme [1]. Since the advent of the Cloud, application orchestration methodologies have been thoroughly studied pursuing the need to automatise the deployment, migration, (re-)configuration, healing and undeployment of software services at scale [2]. In this context, containerisation technologies made it simpler for DevOps engineers to manage and monitor Cloud applications during their whole life cycle, usually through a declarative description of the desired target state and needed infrastructure resources [3]. This is the case, for instance, of Docker Swarm and Kubernetes orchestration engines.