In this episode of the @HPCpodcast, join us for a rare, behind-the-scenes glimpse at the Frontier exascale supercomputer, how it was built in the middle of a pandemic and how it’s being prepared for full user-readiness.
Frontier is a $600 million, 30 MW system comprised of 50-60 million parts in more than 100 cabinets, deployed at the Oak Ridge National Laboratory by HPE using AMD CPUs and GPUs. It is slated to be the first U.S. exascale computing resource with a target performance of about 1.5 exaFLOPS in double-precision (64-bit) arithmetic.
It’s a milestone computer – but not the first such milestone project in the career of our special guest, Dr. Jeff Nichols, who oversees the Department of Energy’s National Center for Computational Sciences. Nichols, a witness and participant in recent HPC history, has been a key figure in the installation of breakthrough supercomputers, such as Titan, Summit and now Frontier, along with one of the first systems to combine CPUs and GPUs. Since 2009, he’s been associate laboratory director of the lab’s Computing and Computational Sciences organization.
His appearance with us is something of a valedictory because Nichols plans to retire this year after 20 years at Oak Ridge. We discuss the past and future of supercomputing as well as the current state of Frontier.
You can find our podcasts at insideHPC’s @HPCpodcast page, on Twitter and at the OrionX.net blog. Here’s the RSS feed.
We welcome your ideas for special topics and guest commentators. Feel free to contact Doug Black or Shahin Khan with your suggestions.