As the edge computing era matures, it encompasses a broader range of applications that will be able to opportunistically use both the edge and the cloud mainly determined by the location constraints or the mission-critical reasons. Moreover, the convergence of verticals with the next generation mobile network will lead to an imminent edge-cloud collaboration that needs to be carefully addressed by Mobile Network Operators (MNOs). This paper proposes a novel approach to orchestrate the next-generation 5G core running as Cloud-Native Network Functions (CNFs) closer to the edge network while offloading the computation power required to ensure on-demand processing to a central cloud infrastructure. In this manner, full service isolation and network segmentation can be achieved independently from the geographical region with a high degree of scalability. Our main objective is to automate the CNFs provisioning in a declarative API-style in order to easily manage the deployment of applications and services across multiple clusters. Therefore, in our setup we considered crucial to assess the benefits of running the 5G Service-Based Architecture (SBA) in the cloud with multi-tenancy capabilities and evaluate the performance in terms of latency, while keeping the network control at the edge.

This paper presents a novel approach for scheduling the workloads in a Kubernetes cluster, which are sometimes unequally distributed across the environment or deal with fluctuations in terms of resources utilization. Our proposal looks at a hybrid shared-state scheduling framework model that delegates most of the tasks to the distributed scheduling agents and has a scheduling correction function that mainly processes the unscheduled and unprioritized tasks. The scheduling decisions are made based on the entire cluster state which is synchronized and periodically updated by a master-state agent. By preserving the Kubernetes objects and concepts, we analyzed the proposed scheduler behavior under different scenarios, for instance we tested the failover/recovery behavior in a deployed Kubernetes cluster. Moreover, our findings show that in situations like collocation interference or priority preemption, other centralized scheduling frameworks integrated with the Kubernetes system might not perform accordingly due to high latency derived from the calculation of load spreading. In a wider context of the existing scheduling frameworks for container clusters, the distributed models lack of visibility at an upper-level scheduler might generate conflicting job placements. Therefore, our proposed scheduler encompasses the functionality of both centralized and distributed frameworks. By employing a synchronized cluster state, we ensure an optimal scheduling mechanism for the resources utilization.