Lightweight Resource Exposure Framework for Efficient Service and Resource Orchestration in the Cloud-Edge Continuum

The Cloud Edge Continuum (CEC) opens up new opportunities for deploying applications across a wide array of resources, ranging from centralized cloud infrastructures to edge and far-edge nodes. However, orchestrating resources in this heterogeneous and dynamic environment requires innovative approaches beyond traditional cloud or network function virtualization (NFV) systems. In this paper, we introduce a novel orchestration framework for the CEC that separates service orchestration from resource orchestration. This separation is crucial for managing the ever-changing and diverse nature of CEC resources, particularly at the edge and far-edge levels. To facilitate this, it is essential for the orchestration and management plane of the CEC to incorporate a Resource Exposer (RE) component, which is the primary contribution of this work. Additionally, a Central Resource Discovery (RD) Module aggregates data from multiple REs, providing a comprehensive global view of the registered infrastructures and available resources. All components of the framework run in containers, ensuring compatibility with containerized environments and seamless integration across Kubernetes, openshift, KubeEdge, and K3S platforms. The experimentation results demonstrate that the proposed system is highly efficient, with minimal CPU and memory consumption, even when deployed on low-resource devices like edge nodes. The results show that the RE achieves low-latency responses and scales well under high-frequency data collection, making it a viable solution for orchestrating resources in the cloud-edge continuum.