The ever-growing resource needs of modern-day applications regarding guaranteed low latency and the massive data transfer rate are constantly pushing the boundaries of technologies and requiring a paradigm shift. To cater for these escalating resource needs, modern IT computing platforms have evolved beyond the more traditional central cloud/DC with bleeding-edge processing powers and high-capacity networking infrastructure to extend their coverage all the way to the network edge, which may also include the far-edge nowadays. This creates a new paradigm called cloud edge computing continuum (CECC), whereby the services span from core cloud to edge and far edge. To efficiently manage and continuously optimize resources through this new model using the CECC approach, we propose an Agile and Cognitive Cloud-edge Continuum (AC3) management framework. This framework will play a critical role in providing scalability, agility, effectiveness, and dynamicity in service delivery over the CECC infrastructure. AC3 will offer a common and secure federated platform that manages data source, CECC resources, and application behaviour in a unified and harmonized manner to ensure the desired SLA and save energy consumption. Moreover, the AC3 platform can adapt to a different context and events happening in the network, such as lack of resources, data deluge, or mobility of data source, by managing (i.e., deploying or migrating) micro-services across CECC infrastructures. AC3 will leverage AI, ML, and semantic and context awareness algorithms to provide an efficient life cycle management system of services, data sources, and CECC resources for ensuring low response time and high data rate while saving energy consumption.

Cloud-based services have grown from basic computing services to complex ecosystems, comprising (virtual) infrastructure, business processes and application code. These advanced services also increasingly leverage the usage of Artificial Intelligence, including Machine Learning or Natural Language Processing techniques, raising the complexity even higher. Due to the cascade of dependencies among the different products and services, the need arose to bring more agility to the certification process of cloud-based services, e.g., using continuous monitoring and assessment, as evidenced by references to it in the certifications of the EU Cybersecurity Act (EU CSA). To transform the continuous assessment and certification concept into the complete realization of a Certification-as-a-Service (CaaS), several challenges need to be solved: 1) current proposed proofs of concepts for continuous monitoring lack interoperability at technology level, 2) the adoption of cloud and edge computing and the incorporation of regulations on specific topics or domains, such as AI, put significant strain on companies to comply with a multitude of different security schemes, 3) existing market fragmentation for continuous certification (scope, methodologies), hinder transparency and accountability in the provision of European cloud services, 4),smart tools and models need to be adopted to ease the agile application and implementation of the CaaS concept reducing complexity in the whole cloud certification value chain easing the adoption of CaaS by the different stakeholders. To overcome these challenges, the design and implementation of the EMERALD CaaS solution leverages the H2020 project MEDINA’s outcomes and advances them to TRL 7 in the EMERALD core. Two PoCs will be provided; one for composite certification and one for mapping requirements to upcoming AI certification schemes. EMERALD will pave the road towards CaaS for continuous certification of harmonized cybersecurity schemes.

One of the key enablers of 6G is undoubtedly the Native support of AI/ML at all the system levels, components, and mechanisms, from the orchestration and management levels to the low-level optimization of the infrastructure resources, including Cloud, Edge, RAN, Core Network, as well as a transport network. Despite the opportunities, there are several gaps that hinder the adoption of AI/ML in 6G, such as the lack of extensive and high-quality datasets that are required to train the models. On the other hand, AI model testing and performance evaluation in a representative staging environment (by emulation or real deployment) is also challenging without access to an end-to-end 6G testbed or representative Digital Twin environment. To this end, 6G-DALI aims to deliver an end-to-end AI framework for 6G, structured in two interdependent pillars, (1) AI experimentation as a service via MLOps and (2) Data and analytics collection and storage via DataOps. The 6G-DALI DataOps pillar provides the mechanisms for preparing clean and processed data that are stored within a 6G Dataspace and are made available for training and validating machine learning models as a service, a part of the MLOps Pillar. The end-to-end framework also delivers continuous monitoring, drift detection and retraining of models. Finally, 6G-DALI will deliver open datasets, a 6G Dataspace for dataset storage and secure sharing, and a Digital Twin testbed for data generation on demand.

The integration of edge computing, advanced 5G connectivity, and decentralized processing drives the widespread deployment of private edge ecosystems capable to reshape numerous industry sectors. However, unlocking the full potentials of edge-level intelligent management requires concerted efforts in platform development and cross-sector collaboration. COP-PILOT, develops a Collaborative Open Platform framework geared towards orchestrating end-to-end services across diverse industry domains. In crafting an open platform, COP-PILOT provides a flexible solution designed to effectively manage various industry sectors while ensuring robust security, automation, and intelligence features. Regarding interoperability, the framework seamlessly integrates with underlying technologies, ranging from IoT platforms to core infrastructure, facilitating collaboration across the compute continuum. Furthermore, COP-PILOT empowers the development of advanced cross-sector applications by offering support for cutting-edge network services, thereby enabling heightened security, resource management, and automation capabilities. The implementation strategy revolves around two primary directions: enabling platform implementation and real environment integration. For the former, COP-PILOT adopts a modular orchestration approach, simplifying the onboarding of complex applications through a user-friendly generative AI interface. This approach includes integration with multi-tiered data processing, policy-driven optimization, and dynamic reasoning capabilities, ensuring alignment with prevailing industry standards. In terms of real environment integration, the platform is deployed across four large piloting clusters, addressing a diverse array of edge paradigms. These use cases span across energy, smart city, agriculture, and industrial manufacturing sectors, fostering the development of cross-sector applications in mobility, logistics, and resource management.