Equipment: MRI: Track 2 Acquisition of a HPC Cluster for Fostering Interdisciplinary Collaboration on AI-driven and Data-intensive Research and Education in West Tennessee

University Of Memphis, Memphis TN

Investigators

Abstract

This project will acquire, deploy, and operate a high-performance computing (HPC) cluster at the University of Memphis (UofM). By sharing the proposed equipment across the West Tennessee region and throughout nearby states (e.g., Mississippi and Arkansas), this cluster will significantly increase research quality and productivity, enable new areas of research efforts, promote training HPC workforce, and enhance the educational commitment of at least 10 participating institutions in this region. The cluster encompasses multiple types of over 100 state-of-the-art GPUs (NVIDIA 4090, A6000, and A100) that fit the diverse needs of regional institutions for research development. The cluster will support teaching and research for undergraduate and graduate students in the region across interdisciplinary research fields, helping prepare the next generation of AI scientists and HPC operation workforces. Education opportunities in the cluster for female students, underrepresented racial groups, and those from other marginalized communities will broadly promote the diversification of the AI and HPC workforce in the region. The cluster will also be available for regional colleges, such as LeMoyne-Owen College and Christian Brothers University, which lack a computing facility. The rapid development of deep learning across research computing communities has facilitated increasingly massive needs for computing capacity to develop AI approaches and big data applications (e.g., ChatGPT). The GPU cluster will support AI research and applications across five research areas: intelligent health, cybersecurity, education technology, smart transportation, and computational biology. Each thread will enable fundamental and applied research in many areas, such as 1) pediatric cancer diagnosis; 2) securing real-time systems; 3) AI-driven tutoring systems; 4) auto-driving navigation systems; and 5) computational simulations for the evolution of insect-plant interactions. The research will put West TN at the forefront of integrating AI and data science into addressing the timely research of scientific and societal issues and providing training opportunities for the next generation of AI scientists and HPC administrators. This cluster will involve more than 37 faculty members with 193 student advisees from a diverse population from over 10 research institutions at West TN, particularly benefiting the underrepresented minorities in STEM (e.g., first-gen and women). Finally, the proposed cluster will enable teaching and training programs, which will empower to better equip students with AI and HPC operations skills; in turn, this will benefit and diversify the future workforce of AI research, HPC management, and regional industry. 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.