This project augments the University of Alabama’s (UA) computing infrastructure to support the increase in computational science and engineering needed to study diverse, interesting problems including: reliable computational chemistry predictions, properties of engineered materials, applied mathematics for image analysis and signal processing, bioinformatics of complex cellular systems, and hydrological simulations. The new high-performance computing (HPC) infrastructure removes bottlenecks in local UA resources caused by an increasing number of users and increasingly larger computational solutions to more realistic problems. This infrastructure enables UA researchers to make scientific and engineering advances not possible with previous UA HPC machines. To provide broader impacts, the infrastructure is also available to regional institutions of higher education, including HBCUs and private institutions who lack adequate HPC access, and to similar institutions across the nation through the Open Science Grid.
This project augments the UA HPC system by (1) doubling computing capacity, (2) adding two large-memory nodes for large-scale data analysis and mining, (3) adding ten GPUs for massively data-parallel computations, (4) significantly increasing storage node bandwidth, and (5) shifting from a “condo†model to a general use, shared model. This infrastructure provides compelling new research and educational opportunities for students, staff, and faculty at UA and other regional and national institutions (including HBCUs and private institutions). In terms of broader impacts, the infrastructure allows undergraduate students to perform state-of-the-art computational research, thereby attracting more diverse STEM participants. The infrastructure provides a platform for educating the next generation of computational and computer scientists in cutting-edge HPC techniques.
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.
