Equipment: MRI: Track 2 Acquisition of a Transformational High-Performance Computing Cluster for AI Research, Innovation, and Capacity Building for Vermont

This project funds construction of IceCore, a graphics processing unit (GPU)-focused, high-performance computing (HPC), enhancing critically needed research computing capacity in Northern New England (NNE). IceCore will enable research of national significance conducted by scientists across NNE, including projects enhancing the explainability of machine learning, infectious disease simulation, hi-resolution mapping of neural circuitry, and the computational design of living robots. Managed by the Vermont Advanced Computing Center (VACC) at the University of Vermont (UVM), IceCore fulfills the unique needs of smaller institutions while catalyzing continued growth in research infrastructure. Access to IceCore will improve the ability of NNE institutions to recruit and retain new faculty, staff, researchers, and students, building their institutional capacity. IceCore will also serve as a powerful educational platform that will train the next generation science and technology workforce for a region whose growth has been critically limited by a shortage of workers with the necessary skills for a digital and computational economy. IceCore will increase the computational capacity of UVM’s GPU cluster by nearly two orders of magnitude to over 100-petaflops, accelerating large-scale computing research and facilitating a broad set of important scientific investigations. Crucially, it will also offer improved HPC access for rural health network research spanning Vermont and Upstate New York (UVM Health Network), as well as enhanced compute capacity for State Agencies that harness the VACC. The presence of a local cluster powered by 64 NVIDIA GPUs provides the UVM research community and the region at large with locally supported advanced computing capabilities, high-speed access, flexible scheduling policy, and customized training. The scientific and technical team at the VACC is committed to training the next generation of data scientists in the region by broadening the number of researchers and students able to access and work with IceCore. This commitment is exemplified by a holistic cyberinfrastructure ecosystem as well as new training workshops focused on GPU computing and machine learning frameworks, courses, and partnerships with NSF-funded undergraduate training initiatives. IceCore will be leveraged to develop cross-sector partnerships with industry and community organizations across NNE to provide computational resources and establish workforce development initiatives that are urgently needed in this rural area. This project is jointly funded by Office of Integrative Activities (OIA) and Office of Advanced Cyberinfrastructure (OAC). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.