Iowa Supercomputer Network

Iowa Supercomputer Network
Name	Iowa Supercomputer Network
Formation	1990s
Headquarters	Iowa City, Iowa
Region served	Iowa
Leader title	Director
Parent organization	University of Iowa

Iowa Supercomputer Network is a statewide high-performance computing consortium serving research, industry, and education across Iowa. It connects academic institutions, state laboratories, and private partners to computational clusters, storage arrays, and high-speed networking for simulation, data analysis, and modeling. The consortium coordinates resources similar to regional initiatives such as XSEDE, Compute Canada, PRACE, National Center for Supercomputing Applications, and Argonne National Laboratory.

Overview

The network aggregates compute cycles, parallel filesystems, and visualization facilities drawing participation from institutions like University of Iowa, Iowa State University, University of Northern Iowa, Drake University, and Luther College. It interoperates with national backbones such as Internet2, ESnet, National Science Foundation, Department of Energy, National Institutes of Health, and National Aeronautics and Space Administration for federated identity and data exchange. User communities include centers of excellence in areas associated with Centers for Disease Control and Prevention, United States Geological Survey, United States Department of Agriculture, Iowa Department of Natural Resources, and private partners comparable to John Deere, Rockwell Collins, Principal Financial Group, and Pella Corporation.

History

The consortium traces roots to campus computing initiatives at institutions like University of Iowa and Iowa State University that sought regional HPC coordination in parallel with developments at Lawrence Livermore National Laboratory, Los Alamos National Laboratory, Oak Ridge National Laboratory, and Sandia National Laboratories. Early milestones included procurement cycles influenced by system deployments such as Cray Research systems, IBM supercomputers, and clusters modeled on Beowulf architectures. Funding and collaboration aligned with programs administered by National Science Foundation grants, state appropriations from the Iowa Legislature, and competitive awards from foundations like the Gordon and Betty Moore Foundation and Howard Hughes Medical Institute.

Architecture and Facilities

Compute infrastructure comprises mixed CPU/GPU clusters using architectures influenced by vendors including Intel, AMD, NVIDIA, IBM, and HPE. Storage layers deploy parallel filesystems inspired by Lustre, GPFS, and object storage approaches similar to Ceph; high-performance interconnects echo technologies from InfiniBand, Omni-Path, and Ethernet. Data centers are housed at campus locations in Iowa City, Ames, Iowa, and Cedar Falls with redundant power and cooling modeled on best practices from Uptime Institute tiers and standards from ASHRAE. Visualization and immersion labs incorporate displays and software ecosystems akin to offerings from NVIDIA Omniverse, ParaView, VisIt, Unity Technologies, and Unreal Engine.

Services and Resources

Core services include batch scheduling managed by systems reminiscent of Slurm Workload Manager, PBS Professional, and LSF, along with container orchestration leveraging technologies like Docker, Singularity, and Kubernetes. Authentication and authorization integrate federated systems comparable to InCommon, Shibboleth, and LDAP directories. Science gateways and web portals are inspired by frameworks used by Galaxy Project, Jupyter Project, and the TACC portal experience. Data management support references practices from FAIR principles, research data stewardship models promoted by DataONE, and digital preservation approaches similar to LOCKSS.

Research and Applications

Research domains span computational fluid dynamics for partners such as John Deere and aerospace projects connected to NASA Glenn Research Center, bioinformatics workflows akin to those at Broad Institute and National Center for Biotechnology Information, climate modeling consistent with NOAA and IPCC workflows, and hydrological modeling paralleling studies from USGS and Iowa Flood Center. Workloads include molecular dynamics simulations like those performed with GROMACS and NAMD, machine learning training similar to efforts at Google DeepMind and OpenAI, and finite element analysis comparable to projects using ANSYS and Abaqus. Collaborations have involved research groups associated with Eli Lilly and Company-style pharmaceutical partnerships, agricultural research aligned with USDA Agricultural Research Service, and urban informatics efforts similar to Smart Cities pilots.

Governance and Funding

The consortium governance is a cooperative model with oversight resembling boards at Big Ten Academic Alliance research cores, with advisory input from institutional representatives such as provosts and research vice presidents from member institutions like Iowa State University and University of Iowa. Funding is a mix of state appropriations through the Iowa Legislature, competitive research grants from National Science Foundation and NIH, industry contracts modeled on partnerships with firms like John Deere and Rockwell Collins, and philanthropic gifts similar to major endowments from foundations such as Simons Foundation.

Education and Outreach

Training programs mirror curricula from Carnegie Mellon University and Massachusetts Institute of Technology on parallel programming, high-performance Python workshops inspired by NumPy and SciPy communities, and summer internships comparable to REU programs funded by National Science Foundation. Outreach includes K–12 engagement resembling initiatives by FIRST Robotics Competition, public seminars like those organized by TEDx, and professional development aligned with continuing education at institutions such as Iowa Public Radio-hosted forums and cooperative extension services akin to those of Iowa State University Extension and Outreach.

Category:Supercomputer centers in the United States