Aims to deliver an infrastructure management system for the holistic management of computing, storage, and networking resources, encompassing techniques for runtime adaptations of all BigDataStack operations.
The validation of the system will be done through the following use cases:
- Real-time ship management: Optimise and help cut costs on maintenance and spare parts inventory planning and dynamic routing.
- The connected consumer: Improving consumer shopping experience with optimal insights into consumer preferences for retailers.
- Smart insurance: A multi-channel scenario will facilitate data analytics-powered smart insurance, providing a 360-degree view of the customer and personalised services.
Our team participates in the project coordinating the development of one of the key capabilities of the platform: the data-driven infrastructure management, working alongside companies (Red Hat, NEC, and IBM) and universities (University of Glasgow and University of Piraeus) with the aim to deliver a self-adaptive infrastructure management system for the holistic management of computing, storage, and networking resources, encompassing techniques for runtime adaptations of all BigDataStack operations.
Performance data at three different layers of the software stack (resource clusters, applications, and data services) are collected and evaluated, individually or aggregated, against levels of Quality of Service (QoS) specified by the data scientists using the platform.