Nvidia has partnered with the College of Florida to up grade the university’s HiPerGator supercomputer. Established to go live in early 2021, the challenge will make HiPerGator the fastest AI supercomputer in academia with seven-hundred petaflops of performance, in accordance to Nvidia and UF.
HiPerGator three. will use a cluster comprised of one hundred forty Nvidia DGX A100 nodes, which are $two hundred,000 supercomputing AI systems that every consist of 8 A100 GPUs. The DGX A100, furnishing 320 GB of RAM for schooling significant AI datasets, is capable of 5 petaflops of performance.
The cluster of one hundred forty DGX A100 nodes, termed a DGX SuperPOD, will so strike seven-hundred petaflops, Nvidia and the college explained. HiPerGator three. will also incorporate four petabytes of high-performance storage and tap Nvidia Mellanox HDR 200Gb/s InfiniBand networking.
“The projected seven-hundred petaflops are surely up there,” explained Nick McQuire, senior vice president and head of AI and business investigate at CCS Insight. “It is part of an arms race we are viewing at a national level and in just academia to ensure universities remain competitive globally.”
UF is the initially academic institute to use the DGX A100 technological innovation, which was declared in May well.
The fastest supercomputer on the Super Computer system Top 500 list from June is a program at RIKEN Middle for Computational Science in Japan termed Supercomputer Fugaku, which hits 415 petaflops of performance. For precise responsibilities, Fugaku peaks at one,000 petaflops, or one exaflop.
If Nvidia’s promises bear out, UF’s future program will be comparable to Fugaku in terms of electricity.
Nick McQuireAnalyst, CCS Insight
Supercomputing level of competition shifts
Up until finally now, most of the high-profile level of competition in AI supercomputing has centered on program speed and been funded at the national federal government level, these kinds of as with Fugaku, McQuire explained.
The Top 500 list does incorporate quite a few supercomputers at universities, which include the College of Texas, Nagoya College in Japan and Mississippi State College. Nvidia technological innovation helps electricity quite a few of the supercomputers on the list, as do HPE and IBM. The a few tech giants dominate the supercomputing industry together with sellers these kinds of as Fujitsu, Dell, Lenovo, Atos and Penguin Computing.
The UF challenge is remaining funded via a $25 million donation from UF alumnus and Nvidia co-founder Chris Malachowsky, along with $25 million in components, program, companies and schooling from Nvidia. UF also pledged $20 million of its individual funds.