INT 24-2a Highlights Report
QCD at the Femtoscale in the Era of Big Data
June 10 - July 5, 2024
J. Bessac, I. Cloët, N.Sato, E. Constantinescu
Modern facilities worldwide have been built to study the quark and gluon structure of matter, such as Jefferson Lab, RHIC, and the forthcoming EIC, which will produce many exabytes of data on the visible universe at the femtometer scale. This new era of big data in nuclear physics holds the promise of solving many of the mysteries in QCD, such as the origin of the nucleon’s mass and spin. However, analyzing this data to extract the required information is an extremely challenging task since quarks and gluons are not directly accessible in experiments because of confinement in QCD.
To solve this challenge at scale, nuclear physics must develop and adopt methods in data science, AI/ML, applied mathematics, and large-scale computing from other fields and mold them to the needs faced in nuclear physics. Such a multidisciplinary approach, which allows for cross-field collaboration and the exchange of ideas, is still in its infancy in the field of QCD phenomenology.
The INT has hosted a 4-week long program that brought together QCD domain scientists with experts in computational science, data science, AI/ML, and applied mathematics to begin developing the ideas, methods, and tools to address these key challenges for nuclear physics in the era of big data. A key goal for the program was to have a significant impact on the synergy between theory and experiment in nuclear physics and drive collaboration with other fields that, over time could have a transformational impact on nuclear physics.
The INT program was successful in bringing together an interdisciplinary group of scientists and QCD experts to address the challenges of the DOE's Femtoscale physics program, particularly in relation to upcoming experiments from Jefferson Lab's 12 GeV program and the future Electron-Ion Collider (EIC). The program featured a flexible format that allowed ample time for discussions and provided space for one-on-one conversations, fostering synergistic exchanges between domain and out-of-domain scientists. A wide range of topics was discussed, from “gauge links to CPU/GPU flops.” The topics included QCD theory, QCD phenomenology, experimental simulations, data science, applied math, HPC, reproducibility, and future directions. An emerging conclusion was that 'the collective is more powerful than individuals,' emphasizing that collaboration between data scientists, HPC scientists, applied mathematicians, and QCD researchers holds great promise for the future of nuclear particle physics. |
INT 24-2a Workshop Webpage |