The data continuum refers to the extension of the traditional data lifecycle to encompass the flow of data across organisational, sectoral and technological boundaries. It requires integrated data infrastructures capable of supporting seamless data flows across heterogeneous computing environments and throughout the AI lifecycle. It therefore brings together complementary dimensions, including the data lifecycle, the computing continuum, and the Data and AI continuum.
Key priorities for this direction include:
- Developing the data continuum, from data generation and ingestion through processing, analytics, insights and decision-making
- Aligning data and computing continuums across edge, cloud, HPC and emerging computing architectures, including required orchestration mechanisms. This includes the post exascale scenario that implies the convergence between Data/AI/HPC/quantum, new computing paradigms (further develop in Section 6.2), and data and AI centric workflows.
- Designing advanced data architectures that support distributed, heterogeneous and multimodal data, real-time processing and scalability, while enabling interoperability with data spaces and meeting the evolving requirements of AI-driven applications
- Leveraging and connecting strategic European infrastructures, including AI Factories, HPC infrastructures and future Gigafactories