Analiza in nadgradnja platforme Kubernetes za robno-oblačni kontinuum

As the digital landscape evolves, the demand for high-performance, low-latency computing has surged, pushing traditional cloud computing to its limits. This has led to the development of edge-cloud continuum, which integrates and leverages the strengths of cloud, fog and edge computing. This master’s thesis presents a comprehensive system design and evaluation of a Kubernetes-based platform that facilitates seamless integration between the cloud and fog layers. Key contributions include a comparative analysis of single versus multi-cluster Kubernetes architectures, federation solutions, and cluster network interconnection strategies. The introduction of workload descriptors enables infrastructure engineers to provide deeper insights into the edge-cloud continuum requirements and preferences of various workloads. By integrating the Federation Scheduler with the Federation Data Collector and workload descriptors, workload placement is significantly improved compared to the standard Kubernetes scheduler, as demonstrated by a 46% improvement in the average edge-cloud continuum placement score without sacrificing scheduling time. This score evaluates factors such as CPU, memory, uptime, bandwidth, latency, fault tolerance, and unscheduled pods. Additionally, Federation Deschedulers maintain a high and stable edge-cloud continuum placement score despite environmental changes, ensuring that edge-cloud continuum objectives are met at any point in time.