OLCF staff and users are presenting at NVIDIA’s GPU Technology Conference in March with a “Summit Supercomputer Application Readiness Projects” Session.

Oak Ridge Leadership Computing Facility (OLCF) staff members and users are presenting at the GPU Technology Conference (GTC) in March with a Summit-focused session in preparation for the launch of the OLCF’s next supercomputer later this year.

GTC, taking place March 26–29 in San Jose, California, is hosted by technology vendor NVIDIA. The annual meeting brings together around 7,000 GPU developers from industry, academia, and government, providing participants with training, insights, and direct access to experts from NVIDIA and other leading organizations.

Since its founding a decade ago, GTC has evolved into one of the largest meetings focused on artificial intelligence in the world. This development has coincided with the rise of GPUs from the tech world’s fringes to the computing engines of forward-looking technologies such as autonomous vehicles, virtual reality, and advanced robotics.

Below is information on the OLCF’s Summit session at GTC, “Summit Supercomputer Application Readiness Projects Session,” and the associated presenters and co-authors.

Monday, March 26:

  • Enabling Large-Scale Science on Summit Through the Center for Accelerated Application Readiness (S8908: Hilton Almaden 2, 1:00 p.m.)
    Jack Wells, ORNL
  • Scaling Deep Learning for Scientific Workloads on Summit (S8551: Hilton Almaden 2, 1:30 p.m.)
    Fernanda Foertter, NVIDIA
    Jack Wells, ORNL (co-author)
    Arjun Shankar, ORNL (co-author)
  • Petascale Molecular Dynamics Simulations on the Summit POWER9/Volta Supercomputer (S8747: Hilton Almaden 2, 2:00 p.m.)
    James Phillips, National Center for Supercomputing Applications (NCSA)
  • Exposing Particle Parallelism in the XGC PIC Code by Exploiting GPU Memory Hierarchy (S8909: Hilton Almaden 2:30 p.m.)
    Stephen Abbott, NVIDIA
  • GPU Acceleration of Multiphysics CFD Software for Propulsion and Power Flow Systems (S8910: Hilton Almaden 2, 3:00 p.m.)
    Joseph C. Oefelein, Georgia Institute of Technology
    Ramanan Sankaran, ORNL (co-author)
    Brian M. Rogers (co-author)
    Lixiang Luo (co-author)
    Levi Barnes (co-author)
  • Accelerated Simulations of Stellar Explosions with FLASH: Towards Exascale Capability (S8926: Hilton Almaden 2, 3:30 p.m.)
    Austin Harris, ORNL
  • GPU Accelerated Performance of QMCPACK on Leadership-Class HPC Systems using Cuda and Cublas (S8937: Hilton Almaden 2, 4:00 p.m.)
    Andreas Tillack, ORNL