Speaker
Description
Research infrastructures (RI) provide access to state-of-the-art instrumentation for the scientific community for example, The Extremely Brilliant Source at the European Synchrotron Radiation Facility, the European Free Electron Laser and the European Spallation Source.
The business case for these facilities is centred on a service model that provides visiting researchers access to cutting edge scientific instrumentation and methods, enabling accelerated delivery of high impact science.
Neutron and photon RIs share a number of key commonalities which present certain challenges that must be addressed to ensure that the RI business model is efficient.
• Data rates and experimental data volumes are challenging for users and facilities.
• High data rates enable complex experimental studies in real (or near real) time. Kinetic studies of reactions or phase transitions result in data collected at rate that challenges experimental teams’ ability to steer the experiment in a considered.
• Neutron and photon methods provide unparalleled insight into physics at the atomic scale. Analysis of data requires the use of modelling and simulation tools that are from a theoretical standpoint state of the art, requiring dedicated HPC environments and expertise.
• The research community leverage the complimentary of photon and neutron research methods, creating a need for interoperable data and data services for facility users.
Leveraging HPC is an essential ingredient for enabling scientific impact, these challenges will be discussed in the context the current status and strategy for high performance computing.
Machine learning and artificial intelligence technologies are seen as a future key enabler for experimental methods and scientific computing that could provide specific solutions for our key challenges of data rate and real time processing these will be discussed within the context of the important scientific consideration of trust and reproducibility.
Title
Considerations and Applications for HPC, ML/AI at next generation research infrastructures
| authors | Jonathan Taylor |
|---|---|
| affiliation | European Spallation Source ERIC |
