Analytics and Visualization

Eureka (Visualizations and Data Analytics)

Most science applications that run on large-scale systems like Intrepid at the Argonne Leadership Computing Facility (ALCF) generate huge volumes of data that represent science results. Researchers analyze and explore this data output, and often convert the data to visual representations. To facilitate data analytics and visualization at the ALCF, researchers employ Eureka, one of the world’s largest installations of NVIDIA Quadro Plex S4 external graphics processing units (GPUs). By using the NVIDIA visual computing system as the base graphics building block, Eureka enables breakthrough levels of productivity and capability in visualization and data analysis.

Configuration

  • 100 dual quad-core servers
  • 200 Quadro FX5600 GPUs in the S4s
  • Memory: 3.2 TB+ RAM
  • Peak Performance: More than 111 TF

Gadzooks (Visualization and Data Analytics):

Gadzooks is the ALCF's test and development system for visualization.

Configuration

  • 4 compute nodes, each with two 2.0 GHz quad-core Xeon servers
  • 8 NVIDIA Quadro FX5600 GPUs in two S4s
  • Memory: 32 GB RAM