Blueprinting AI for Science at Exascale (BASE-II)

BASE-II is the UK research project with the main goal of developing a software suite and designs to serve as blueprints for using AI for scientific discoveries at exascale. This is phase II of BASE-I (Benchmarking for AI for Science at Exascale) project, started in December 2022 and will run for two years, facilitating faster scientific discovery and cross-scientific knowledge exchange.

BASE-II will deliver software:

to support exascale computing from the AI point of view by developing a software suite and designs
to address challenges in critical areas of AI:
- AI in benchmarking
- AI in HPC
- Hardware software co-design
- Learning from large-scale datasets, tools, frameworks, and workflows.

Blueprinting AI for Science at Exascale:

Supports collaboration and knowledge exchange between academia, national laboratories, and industry.
Encourages cooperation between various scientific disciplines.

BASE-II for exascale community:

Actively engages and supports knowledge exchange with ExCALIBUR funded projects in the UK, and relevant initiatives in Europe and USA.
Shares experience with evolving UKRI e-Infrastructures, such as Science Clouds, the IRIS science platform, and exascale-cloud middleware.

About

The project

An AI Benchmark Toolbox consisting of benchmarks, datasets and frameworks for an extensive number of scientific, climate and fusion research cases
Methodologies and solutions for different classes of surrogate cases for complex simulations, including an open-source model zoo for surrogate models (SMs)
Hardware-Software codesign for exascale systems, supporting both communities and stimulating their partnership
Knowledge exchange and community building, by organising training and communication events, between exascale, research software engineering, AI and science communities

BASE-I

Successful project preceding BASE-II, which delivered the first AI for science benchmark suite and shaped the current project to address challenges.

Current progress

BASE-II is driving change for the AI and Exascale community. Learn more about our progress in developing AI benchmarks, increasing AI/HPC convergence, input in Hardware-Software codesign, and community building.

BASE-II GitHub

We are proud to be part of an open-source science community. You can find information about our benchmarks, ML models, and links to datasets we used in this research project.