CAMINO - Contextually Aware Mediation of Intent for Network Orchestration
摘要
Conflicting goals of self-organizing components in O-RAN open, modularized architectures present a major issue for the management of telecommunications networks. The use of autonomic network control results in individual management applications making real-time reconfigurations to the network without user intervention. A major unresolved challenge presents as each autonomic agent operates with its own specific goals, resulting in potential conflicting actions. Many orchestration techniques have been proposed to avoid deploying components that do not work together harmoniously due to competing goals and effects. However, efficient operation of a telecommunications network involves trade-offs between competing goals. This paper proposes the CAMINO - Contextually Aware Mediation of Intent for Network Orchestration architecture as an extension to the O-RAN Software Management and Orchestration (SMO) function. CAMINO preemptively detects conflicts between the individual reconfiguration actions of competing components. CAMINO implements a novel, contextual, enhanced intent-based architecture utilizing information from non-traditional sources to determine how to resolve the conflict. CAMINO facilitates the operator’s intent for the network through analysis of both telecommunications data sources and external data sources, i.e., weather, traffic, planned events, incident information. CAMINO preemptively detects conflict between the competing actions of autonomic network configuration functions. Reconfiguration actions in the network that are predicted to cause degradation of Key Performance Indicators (KPIs) can be over-ridden using preemptive conflict detection. CAMINO specifically targets predicted network degradations that conflict with the current intent of the network operator, are contra-indicated by telecommunications data sources, or are assessed as being incompatible with current conditions in the operating environment.