Intelligent operator: Machine learning based decision support and explainer for human operators and service providers in the fog, cloud and edge networks

Abstract

The growing volume of cloud-based applications, services and cyber-physical IoT solutions presents vital challenges linked to resource allocation, misconfiguration, scaling, and running software updates. Various solutions and applications have different hardware and energy requirements of the involved elements. Hence, the recent technology trends suggests transferring some more complex computational tasks on devices that leverage fog computing concept. Obviously, the tangible outcome in that case is the distribution of the processing capacity closer to the data sources. This often gives the user better control over the data with respect to the privacy, integrity, and security. The Intelligent Operator proposal introduced in this paper addresses application misconfiguration problems in a cloud environment. Its main goal is to provide better maintainability and scalability of the deployed applications.