AI Supercomputer
The AI Supercomputer is designed to run intensive machine learning and deep learning workloads at scale. It consists of on-premise NVIDIA DGX systems (referred to as the DGX On-Prem system) and supports GPU-accelerated computation for research and teaching use.
By requesting access, users can run Jupyter notebooks, JupyterLab, and Python scripts optimized for GPU workloads. Additionally, any of the available containers that are provided by NVIDIA on the NGC Cloud can be pulled and used with the system.
Features
24 NVIDIA DGX A100 systems (each with 8 NVIDIA A100 GPUs)
3 Lambda scalers (each with 8 NVIDIA L40s)
Over 6 petabytes of NetApp storage
Paid, prioritized GPU access
Available to
UAlbany Researchers
Startups & Industry Affiliate Partners (via UAlbany Innovation Center / OEDEIP)
SUNY Users (faculty and staff from other SUNY campuses)
Access Tiers & Pricing
UAlbany Researchers
UAlbany faculty and researchers can access the AI Supercomputer at no charge through the free tier, which provides first-come, first-served GPU resources that may be preempted by prioritized jobs. For projects requiring guaranteed resources and prioritized scheduling, access is available at $1,200 per GPU annually. To get started, complete the Research Storage Request Form and the DGX On-Prem Computation Request Form.
Innovation Center Startups
Innovation center startups interested in using the AI Supercomputer must first join the UAlbany Innovation Center, managed by the Office of Economic Development, Entrepreneurship, and Industry Partnerships (OEDEIP). Once affiliated, members can take advantage of discounted pricing for the AI Supercomputer.
Industry Affiliate Partners
Industry Affiliate Partners interested in using the AI Supercomputer should work with the Office of Economic Development, Entrepreneurship, and Industry Partnerships (OEDEIP) to learn more about the process for gaining supercomputer access.
SUNY Users
Faculty and staff from other SUNY campuses are also eligible to use the AI Supercomputer. For pricing information, SUNY users should contact the ITS Service Desk at askIT@albany.edu for more information about pricing information and onboarding instructions.
More Information
Documentation about how to use the NVIDIA on-premise system is present on the DGX On-Prem How-To page
Tutorials for running AI workloads are available here.
For more information on AI Plus Supercomputing Cluster, contact the ITS Service Desk