The US currently has the world's most powerful supercomputer and the country could hold this claim for some time thanks to the announcement that Cray Computing and AMD are building an exascale machine together with the Oak Ridge National Laboratory.
The system is on schedule to debut in 2021 when Cray and Intel will release their own Aurora exascale supercomputer to the Argonne National Laboratory.
Secretary of Energy Rick Perry praised the announcement in a statement, saying:
"Frontier's record-breaking performance will ensure our country's ability to lead the world in science that improves the lives and economic prosperity of all Americans and the entire world. Frontier will accelerate innovation in AI by giving American researchers world-class data and computing resources to ensure the next great inventions are made in the United States."
- 10 of the world's most impressive supercomputers
- HPE unveils world's largest supercomputer
- Mercedes F1: How HPE supercomputing is helping push us to success
Frontier supercomputer
According to Cray's CEO Pete Ungaro, Frontier will have the combined power of the current top 160 supercomputers worldwide at 1.5 exaflops.
Oak Ridge National Laboratory director Thomas Zacharia pointed out that Frontier will be the “most powerful AI machine anyone has seen up until that point in time”. He also revealed that it will be the most expensive machine ever produced with the contract for Frotnier's system and technology coming in at over $600m.
Frontier will utilize Cray's Shasta supercomputer system and Slingshot interconnect in addition to GPUs and custom EPYC processors from AMD. The system will occupy a space almost as large as two full-sized basketball courts and will include 90 miles of cabling.
Scientist and engineers will use Frontier for a wide variety of projects including climate change forecasting, predicting the path of hurricanes and developing new pharmaceuticals. The supercomputer will also be used to model fusion energy sources, simulate combustion engine systems and conduct cosmology simulations for the distribution of billions of galaxies.
Via Engadget
