At SC18, NVIDIA GPUs Are at Every Turn

by Christina Olmsted

The brightest minds in HPC, deep learning, and machine learning from all over the world will come together at SC18, in Dallas next week, to discuss and demo their computationally intensive projects.

This year, NVIDIA will showcase its work with some of the most powerful supercomputers in the world, including the Summit system at Oak Ridge National Laboratory and the Sierra system at Lawrence Livermore National Laboratory. Summit is the world’s fastest supercomputer and Sierra is the third fastest, according to this summer’s Top500 list.

Summit, an open system for researchers worldwide, features more than 27,000 NVIDIA Volta Tensor Core GPUs for over 200 petaflops of double-precision computing for HPC and 3 exaflops of mixed-precision.

Sierra features more than 17,000 NVIDIA Volta Tensor Core GPUs delivering 7 peak teraflops of double-precision floating-point performance and 125 teraflops of deep learning performance.

SC18 is also where the winner of the Gordon Bell Prize — the top award for science, engineering and large-scale data science — will be announced. Five of the six finalists accomplished their amazing achievements on the GPU-accelerated Summit and Sierra systems.

NVIDIA SC18 preview theater talk

Don’t miss the following SC18 sessions hosted by NVIDIANs:

More NVIDIA-hosted sessions can be found here.

Additional SC18 highlights include:

  • Special address from NVIDIA CEO Jensen Huang: RSVP now, watch live from the Hyatt Regency in Dallas or stream the presentation to hear about the latest innovations in HPC.
  • HPC hot topics: Hear talks featuring cutting-edge technologies, groundbreaking research and more from a broad range of disciplines such as weather forecasting, energy exploration and molecular dynamics. We’ll be featuring these talks in our in-booth theater. See the full theater schedule to plan your week.
  • Developer Zone: Want to make your applications faster and smarter? Stop by NVIDIA’s in-booth Developer Zone to speak with an HPC or deep learning expert on Tuesday, Nov. 13, and Wednesday, Nov. 14, from 1-3 pm. Or take an NVIDIA Deep Learning Institute self-paced lab.
  • One million and counting: Thanks to the contributions of our many developers, the NVIDIA developer program has grown to over 1 million in the last decade. Swing by the booth and help us celebrate key HPC and AI GPU-accelerated milestones that are changing the world. If you’re a member of the NVIDIA developer program, be sure to ask for your special gift.
  • Interactive demos: Experience incredible demos on scientific simulation and AI.
  • NVIDIA SC18 on social: Get real-time SC18 updates from NVIDIA via Twitter, Facebook, LinkedIn, Instagram and YouTube.

NVIDIA SC18 preview show floor

Many of our partners are also showcasing NVIDIA Tesla V100 Tensor Core GPUs and other GPU-accelerated computing offerings in the exhibit hall. Here are a few of the companies presenting the latest NVIDIA GPU-accelerated systems:

  • Altos
  • ASRock Rack
  • Boston Limited
  • BOXX
  • Dell EMC
  • Cisco
  • Colfax
  • Cray
  • HPE
  • IBM
  • Inspur Electronic Information
  • Lenovo
  • Microway
  • One Stop Systems
  • Penguin Computing
  • QCT
  • RAVE Computer
  • Supermicro
  • 2CRSI

For more information, including our in-booth theater schedule, developer zone activities, recruiting and more, check NVIDIA’s Supercomputing 2018 website.