End-to-end slices to orchestrate resources and services in the cloud-to-edge continuum

Fog computing, combined with traditional cloud offers an inherently distributed infrastructure – referred to as the cloud-to-edge continuum that can be used for execution of low-latency and location-aware IoT services. The management such is complex: resources in multiple domains need accessed by several tenants, while adequate level isolation performance has guaranteed. This paper proposes dynamic allocation end-to-end slices perform orchestration services a scenario. These require unified resource approach encompasses both data centre network resources. Currently, fog mainly focused on compute resources, likewise, slicing domain specifically centred solely creation isolated partitions. A strategy, able integrate selection, configuration part single abstracted object, missing. work aims minimise silo-effect, foundation comprehensive continuum, well acting basis system implementation. concept slice formally described via graph-based model allows discovery, selection mapping different algorithms optimisation goals; working presented way build across dynamically, based model. are independently accessible objects abstract various providers traded Marketplace slices, allocated using bare-metal approach, being interconnected each other connectivity slices. Experiments, carried out real testbed, demonstrate three features slices: selected, controlled softwarised fashion; tenants instantiate those transparently; service absolutely not affected status share same infrastructure.