Cloud–edge microservices architecture and service orchestration: An integral solution for a real-world deployment experience

Abstract

After some years in which information management has been centralized in cloud platforms, the need to implement architectures that perform more efficient and intelligent management of such information has become essential. In addition, the growth of Smart City solutions, which involve the deployment of large-scale network infrastructures, has revealed the benefits of processing data as close as possible to the source, a key principle of Edge Computing. However, the benefits of service orchestration in real-world applications have not been demonstrated to date given the novelty and difficulties related to large-scale deployments. To overcome this barrier, this work presents the design, deployment, and validation of a cloud–edge microservice-based architecture for service orchestration on top of a real-world network consisting of 175 nodes. These nodes were deployed as part of a regional network of touristic terminals to raise tourism’s potential in the region of Badajoz, Spain. The integration of the openBalena open-source cloud platform into the architecture enables remote and centralized management. The experiments carried out revealed an orchestration latency time to deploy new applications from scratch throughout the entire region between 22 s and 292 s depending on the size of the application image. In addition, CPU load and memory usage have been monitored and the lack of suitability of virtualized environments for interactive applications has been overcome, and a step-by-step guide from a successful deployment is provided. As a result, this integral solution contributes valuable insights into how to bring the benefits of orchestration to reality.