Enhancing fusion AI/ML research with the Fusion Data Platform
POSTER
Abstract
The Fusion Data Platform (FDP) is an initiative designed to accelerate advancements in fusion research and artificial intelligence/machine learning (AI/ML) by providing seamless access to curated fusion data, state-of-the-art data science tools, and high-performance computational resources.
The FDP aims to bridge the gap between fusion scientists and AI/ML researchers by offering a unified environment that integrates experimental and simulation data from multiple sources. This comprehensive platform includes advanced features such as federated data sharing via the Open Science Data Federation, fusion-specific data curation tools, and integrated provenance tracking for code, data, and generated artifacts.
This presentation will serve as the announcement of the initial release of the platform, and will detail the FDP design, highlight its capabilities, and discuss the collaborative efforts that have shaped its creation. Beta users will be invited to participate and will gain access to FDP tools, along with federated usage of DIII-D data, including both experimental and gyrokinetic simulation datasets, at the San Diego Supercomputer Center.
The FDP aims to bridge the gap between fusion scientists and AI/ML researchers by offering a unified environment that integrates experimental and simulation data from multiple sources. This comprehensive platform includes advanced features such as federated data sharing via the Open Science Data Federation, fusion-specific data curation tools, and integrated provenance tracking for code, data, and generated artifacts.
This presentation will serve as the announcement of the initial release of the platform, and will detail the FDP design, highlight its capabilities, and discuss the collaborative efforts that have shaped its creation. Beta users will be invited to participate and will gain access to FDP tools, along with federated usage of DIII-D data, including both experimental and gyrokinetic simulation datasets, at the San Diego Supercomputer Center.
*Work supported by the U.S. DOE, Office of Fusion Energy Sciences, using the DIII-D National Fusion Facility, a DOE Office of Science user facility, under Award No. DE-FC02-04ER54698, along with Office of Fusion Energy Sciences Awards No. DE-SC0024426, DE-SC0024499, DE-SC0024409, and DE-SC0024571.
Presenters
-
Brian Sammuli
- General Atomics