Minnesota Supercomputing Institute

Minnesota Supercomputing Institute
Name	Minnesota Supercomputing Institute
Formation	1984
Headquarters	University of Minnesota, Twin Cities
Leader title	Director
Leader name	Chancellor-appointed

Minnesota Supercomputing Institute is a high-performance computing center located at the University of Minnesota, Twin Cities, providing computational resources, data storage, and consulting to researchers across science and engineering. The institute supports interdisciplinary work spanning fields such as bioinformatics, climate science, computational chemistry, astrophysics, and engineering through clusters, software, and training. It integrates infrastructure, partnerships, and educational programs to enable research involving modelling, simulation, and data analysis for faculty, students, and external collaborators.

History

The institute was established amid a period of expanding computational capacity at American research universities, contemporaneous with initiatives at Lawrence Berkeley National Laboratory, Argonne National Laboratory, Oak Ridge National Laboratory, Los Alamos National Laboratory, and Sandia National Laboratories. Early collaborations involved procurement cycles similar to those at National Center for Supercomputing Applications and San Diego Supercomputer Center, and the institute participated in consortiums echoing efforts by Internet2 and National Science Foundation-funded facilities. Over successive decades it adopted architectures aligned with work at IBM, Cray Research, Dell Technologies, and Hewlett-Packard Enterprise, and benefited from funding models used by institutions like Massachusetts Institute of Technology and Stanford University. Major milestones paralleled facility upgrades at NERSC and technology transitions associated with MPI and CUDA support.

Facilities and Systems

The institute operates compute clusters, storage arrays, and visualization nodes comparable to systems deployed at Princeton University, Columbia University, University of California, Berkeley, and University of Illinois Urbana-Champaign. Hardware refresh cycles have involved components from NVIDIA, Intel Corporation, AMD, HPE, and Lenovo. Networking underpinnings have used designs consistent with Infiniband fabrics and campus connectivity practices shared with Internet2 and regional research networks such as Minnesota Supercomputing Institute's peers at University of Michigan and Purdue University. Persistent storage and archive strategies reflect standards adopted at Zettabyte File System, Lustre file system, and enterprise arrays used by Google research groups. The facility includes secure data enclaves and authentication systems interoperable with solutions from Shibboleth, GLUE schema implementations, and identity providers similar to those used by ORCID.

Research and Applications

Researchers leverage the institute for projects in genomics, neuroscience, materials science, seismology, oceanography, and remote sensing. Studies funded by agencies such as National Institutes of Health, Department of Energy, National Aeronautics and Space Administration, and National Science Foundation have used the institute's resources for simulations related to molecular dynamics, quantum chemistry, finite element analysis, and computational fluid dynamics. Cross-disciplinary teams including investigators from Mayo Clinic, Medtronic, 3M, and Cummins have executed workflows for translational research, drug discovery, and industrial design. Publications have appeared in venues like Nature, Science, Proceedings of the National Academy of Sciences, Physical Review Letters, and IEEE Transactions on Parallel and Distributed Systems.

Education and Outreach

The institute provides workshops, bootcamps, and course support modeled after programs at Carnegie Mellon University and University of California, San Diego. Training topics align with curricula from Software Carpentry, Data Carpentry, The Carpentries, and graduate programs at University of Wisconsin–Madison and Ohio State University. Outreach includes partnerships with regional high schools, community colleges such as Normandale Community College, and statewide initiatives comparable to Minnesota State Colleges and Universities systems. Student engagement has included internships linked to summer programs mirroring REU structures and collaborative projects with industry partners like IBM Research and Microsoft Research.

Governance and Funding

Administrative oversight is provided through university structures akin to those at Office of the Vice President for Research and integrates policy practices from Federal Acquisition Regulation-informed procurements. Funding sources have historically included competitively awarded grants from National Science Foundation, programmatic support from National Institutes of Health, user fees patterned after models at Purdue University, and capital investments similar to those from state appropriations seen at University of Minnesota System. Governance bodies include advisory committees with representatives from colleges such as College of Science and Engineering, School of Public Health, and College of Biological Sciences, and they adopt compliance frameworks aligned with HIPAA and FISMA where applicable.

Collaborations and Partnerships

The institute partners with federal laboratories including Argonne National Laboratory and Oak Ridge National Laboratory, university centers such as Minnesota Supercomputing Institute's counterparts at University of Minnesota Duluth and University of St. Thomas, and industry collaborators including NVIDIA, Intel, AMD, HPE, and local firms like Ecolab and Best Buy for applied projects. It engages consortia and regional networks similar to Big Ten Academic Alliance, Great Lakes Consortium for Petascale Computation, and Minnesota Supercomputing Institute-style collaborations with state research initiatives. Joint efforts extend to repositories and open science platforms such as GitHub, Zenodo, and community software projects under Apache Software Foundation governance.

User Access and Services

Access policies accommodate faculty, graduate students, postdoctoral researchers, and external affiliates using account provisioning workflows resembling those at University of Colorado Boulder and University of Texas at Austin. Services include batch scheduling with schedulers like Slurm, job orchestration influenced by Kubernetes practices, software stacks including Python (programming language), R (programming language), MATLAB, TensorFlow, and domain packages such as Gaussian (software), LAMMPS, and GROMACS. Consulting, code optimization, and data management support are delivered by staff trained in best practices from REDCap deployments and research data management guidance promoted by DuraSpace and DataONE.

Category:University of Minnesota Category:Supercomputing centers