

Towards a Strategy for Future Research Infrastructures
Thursday, June 12, 2025 4:00 PM to 5:00 PM · 1 hr. (Europe/Berlin)
Hall 4 - Ground floor
Birds of a Feather
Application Workflows for DiscoveryCommunity EngagementComputational Physics
Information
The amount of data gathered, shared and analysed in frontier research is set to increase dramatically in the coming decade, leading to unprecedented data processing, simulation/prediction and analysis needs. As prime examples, the High Energy Physics and Radio Astronomy communities are gearing up to operate groundbreaking instruments such as the High-Luminosity Large Hadron Collider (LHC) and the Square Kilometer Array (SKA) , which will need data and compute capabilities many times larger than the currently available resources. Given the data volumes produced by these instruments, the size of the associated scientific communities and the scale of the analysis and computation problems, it is clear that distributed infrastructures integrating Edge, Cloud and large HPC/AI centres into a data and compute continuum will be required.
Such digital infrastructures take a long time to establish and require significant capital investment; in addition, operational costs and the environmental footprint need to be considered. On the user side, the specific requirements of the target scientific communities have to be assessed, and they must be supported in an effective and efficient way. Hence, there is the urgent need to start designing and planning these ExaByte-scale research infrastructures today, starting with a comprehensive study of the scientific community needs and plans, a reflection on how the current and short-term planned infrastructures support these, and a strategy to close the gaps to become able to tackle the challenges from most relevant data-intensive scientific collaborations.
This BoF will bring together top-level domain expert representatives from the High Energy Physics and Radio Astronomy domains and top-tier High Performance Computing infrastructure representatives across Europe and the US. Feedback from ISC community will be fed into the technical blueprint of the capabilities of the future infrastructure together with its roadmap for research, innovation and deployment of the future infrastructure
Organizers:
Such digital infrastructures take a long time to establish and require significant capital investment; in addition, operational costs and the environmental footprint need to be considered. On the user side, the specific requirements of the target scientific communities have to be assessed, and they must be supported in an effective and efficient way. Hence, there is the urgent need to start designing and planning these ExaByte-scale research infrastructures today, starting with a comprehensive study of the scientific community needs and plans, a reflection on how the current and short-term planned infrastructures support these, and a strategy to close the gaps to become able to tackle the challenges from most relevant data-intensive scientific collaborations.
This BoF will bring together top-level domain expert representatives from the High Energy Physics and Radio Astronomy domains and top-tier High Performance Computing infrastructure representatives across Europe and the US. Feedback from ISC community will be fed into the technical blueprint of the capabilities of the future infrastructure together with its roadmap for research, innovation and deployment of the future infrastructure
Organizers:
Format
On Site
Targeted Audience
Domain scientists using digital infrastructures, HPC/AI at large including users and infrastructure operators, and decision makers from HPC/AI centers and funding agencies.
Speakers

Sandra Diaz Pier
Scientific lead of the Simulation and Data Lab NeuroscienceForschungszentrum Jülich
Sergio Andreozzi
Head of Strategy, Innovation and CommunicationsEGI Foundation
Maria Girone
Head of CERN OpenlabCERN
Christian Engelmann
Senior Computer ScientistORNL
Jeff Wagg
Radio Astronomy Data ScientistOCA
Hans-Christian Hoppe
Senior Project LeadParTec AG