MRI: Acquisition of a computer system for Research and Education – Seawulf

Suny At Stony Brook, Stony Brook NY

Investigators

Abstract

This project by Stony Brook University, with broad benefit to other research institutions across the region, will purchase and operate for at least five years an advanced high-performance computer system. Focused upon enabling new frontiers in science and engineering, this will be the first deployment in an academic setting of multiple new high-performance computing technologies that directly address the major bottlenecks in high-performance computing and data-centric computing. A huge number of projects from various disciplines will benefit from this computational resource. The new computer will transform scientific and engineering computing with speedups per node approaching 4 to 8-fold for bandwidth-intensive applications. Many scientific and engineering applications important to the National Science Foundation and the campus are bandwidth limited, and with conventional memory technology simply cannot scale to use even a fraction of the cores within modern servers. After demonstrating need and benefit, access will be free to all Stony Brook faculty, students, and their external collaborators. The project will partner with Stony Brooks Women in Science and Engineering Program and Center for Inclusive Education to enable undergraduate women and minority graduate students to work with peers and faculty in computational science, and to receive training and access to advanced computational resources. Regionally, we will collaborate with minority-serving institutions and schools to motivate careers in science and engineering and provide access and training in research and classroom settings. Nationally, the project will leverage existing organizations and activities including National Science Foundation sponsored cyberinfrastructure activities, including the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support Broadening Participation program. All information like documentation, frequently asked questions, login instructions, data, code examples and much more will be gathered online at https://it.stonybrook.edu/services/high-performance-computing. The project website will be maintained during the whole project duration, enabling users to get fast and easy support for the system. 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.