CC* storage-region: GP-stor: great plains scalable tiered object repository

The National Science Foundation (NSF) awarded the Board of Regents of the University of Nebraska $1.4 million on October 1, 2025, under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070) to develop and deploy the Great Plains Scalable Tiered Object Repository (GP-STOR). This two-year project, concluding September 30, 2027, addresses critical cyberinfrastructure gaps in the Great Plains region by establishing a distributed storage infrastructure accessible to research institutions across the region. The primary deliverable consists of three identical storage clusters physically located at the University of Nebraska-Lincoln, University of Missouri, and University of South Dakota, collectively providing 14 petabytes (PB) of high-capacity storage and 384 terabytes (TB) of high-speed, high-input/output operations per second (IOPS) storage. Each site features 11 high-capacity Ceph nodes and one non-volatile memory express (NVMe) node connected via 100 gigabit per second (GBPS) networks to the National Research Platform (NRP), facilitating access through block device interfaces and a user-friendly web portal.

GP-STOR directly supports NSF's cyberinfrastructure mission by enabling data-intensive research requiring substantial storage capacity, particularly benefiting smaller institutions with limited infrastructure. The infrastructure accommodates diverse research applications including space-borne computer vision, dark matter detection experiments, proteomics research, and explainable artificial intelligence applications in biomedical data analysis. Sub-awardees include the University of Missouri System (Curators of the University of Missouri) and the University of South Dakota, which serve as co-deployment sites and collaborative partners in the project's implementation and operation.

Abstract

The great plains region faces significant challenges in accessing advanced computing resources required for data-intensive research, particularly at smaller institutions with limited infrastructure. The growing use of AI and machine learning in both research and education has created a growing need for both high capacity and high performance storage. The great plains scalable tiered object repository (GP-stor) addresses these challenges by deploying a distributed storage infrastructure accessible to institutions throughout the entire great plains region and physically located at the university of nebraska-lincoln, university of missouri, and university of south dakota. The augmented storage and computational capabilities directly address NSF?s mission by expanding access to essential cyberinfrastructure resources, fostering greater collaboration, improving scientific reproducibility, and accelerating data-driven research that could significantly impact fields such as public health, agriculture, and energy. GP-stor consists of three identical storage clusters strategically distributed across different geographic locations contributing a combined 14PB of high capacity storage and 384TB of high-speed, high-iops storage. Each site features 11 high-capacity ceph nodes and one nvme node. These clusters are connected to high-speed 100 gigabit per second (gbps) science dmzs to resources in the national research platform (NRP), through the great plains network and national optical research networks. Resource access is streamlined via the NRP, which provides robust block device interfaces for performance-intensive tasks and a user-friendly web interface. The infrastructure is designed to support numerous research areas including space-borne computer vision, dark matter detection experiments, proteomics research, and explainable artificial intelligence applications in biomedical data analysis. 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.- subawards are planned for this award.