AI-empowered Edge Cloud Continuum for self-aware cognitive computing environments

Next generation enablers, such as IoT, AI and cloud computing, open new opportunities to deal with world’s current and future societal, environmental, and economic challenges. However, they come with significant data management challenges. According to IDC, the total amount of data generated only by connected devices will exceed 40 trillion gigabytes by 2025. In most of the current systems, data storage and analysis happen on centralized locations on the cloud. This is pushing network capacity to its limits. Furthermore, centralized storage and processing lead to single point of failure situations, which are critically inefficient in case of disasters and crisis. Our world is continuously living natural disasters, health crisis, climate change and security threats that show the importance of resilient and energy-efficient information systems. COGNIFOG targets those challenges and proposes to build a Cognitive Fog Framework to: (i) reduce energy consumption and latency in next generation IT systems by reducing the network traffic, by analysing data at the edge in a distributed manner, closer to where they are generated, rather than routing them through the communication networks to a data center; (ii) reduce OPEX and faster service provisioning by providing a cognitive, self-adaptive framework with minimum or no human intervention, with dynamic provisioning of computing, storage and networking resources along the far-edge-to-edge-to-cloud path; (iii) ensure European leadership by providing an open interoperable framework with open APIs for application developers to rapidly create and deploy applications benefiting the edge-cloud continuum on top of heterogenous IoT/IT systems. COGNIFOG will validate project results in three representative application domains: critical collaboration missions, smart health and smart industry. With a consortium of 12 European partners, leaders in their domains, COGNIFOG will be a cornerstone in the cognitive fog computing domain