// Prism GPU Cluster

As the world of artificial intelligence and machine learning (AI/ML) evolves into tools that can be used for scientific discovery, the NCCS has built a 22-node GPU cluster, with NVIDIA v100 GPUs, to facilitate this new research paradigm. We chose to call this new system “Prism” because just as a regular prism splits light into individual wavelengths, allowing scientists to look at light differently and increase their knowledge of the electromagnetic spectrum, Prism will allow scientists to look at their scientific data differently, resulting in new discoveries.

Prism has access to over 10 petabytes of curated NASA data products as well as data produced on both the Discover supercomputer and the ADAPT Science Cloud. The system allows users to examine datasets in the climate model, Earth observation, heliophysics, and astronomical fields to generate training datasets for identifying patterns and making predictions. Co-locating these datasets provides an ideal opportunity for cross-disciplinary science to break through barriers between traditional discipline boundaries.

For those who enjoy having system names be acronyms, PRISM stands for Promoting Research In Science using Machine learning.

// System Specifications


22 Supermicro Compute Nodes

  • 4x NVIDIA V100 GPUs with 32 GB of VRAM and NVLink
  • Dual Intel Xeon Cascade Lake Gold 6248 CPUs;
    20 cores each at 2.50GHz
  • 768 GB RAM
  • Dual 25Gb Ethernet network interfaces
  • Dual 100Gb HDR100 Infiniband high speed network interfaces
  • 3.8 TB RAID protected NVMe drives, mounted as /lscratch


One NVIDIA DGX Node – currently for pilot users

  • 8x NVIDIA A100 GPUs with 40 GB of VRAM and NVLink
  • Dual AMD EPYC Rome 7742 CPUs;
    64 cores each at 2.25GHz
  • 1 TB RAM
  • Dual 25Gb Ethernet network interfaces
  • Dual 100Gb HDR100 Infiniband high speed network interfaces
  • 14 TB RAID protected NVMe drives, mounted as /lscratch


// Using Prism 

Everything you need to know about how to work with ADAPT:

  • Environment
  • Access
  • Tools
  • Data Collections
  • Storage

LEARN MORE

Check Usage on Prism (Linux)

GANGLIA




// Data Collections

Available datasets include atmosphere, ocean, land, and flood data, both current and historical, as well as operational Global Modeling and Assimilation Office (GMAO) weather analysis data and forecasts that are updated four times daily

DATASETS

// Case Studies 

Discover CPUs and ADAPT GPUs Help Accelerate Findings in Astronomical Data

This innovative data science project applies artificial intelligence, machine learning, and...
READ MORE

NCCS GPU CLUSTER AVAILABLE NOW!

The NCCS announces a new resource for NASA-funded scientists: an NVIDIA GPU cluster running on the ADAPT platform.
READ MORE


// Account Information 


new users icon

See information about account eligibility, set-up, and maintenance, or reset password.

GET STARTED