Chipmaker AMD recently announced that its high-performance computing (HPC) technology is behind the Frontier system developed by the Oak Ridge National Laboratory (ORNL) which achieved a score of 1.1 exaflops.
With this performance, the Frontier system has earned the title of ‘world’s fastest supercomputer’ to date as the first ever supercomputer to go beyond the exascale ceiling, clocking higher than the next seven systems on the Top500 list combined.
“A recent development to the Frontier system has allowed the machine to surpass the one exaflop barrier. With an exact HPL score of 1.102 Exaflop/s, Frontier is not only the most powerful supercomputer to ever exist – it’s also the first true exascale machine,” Top500 said in a statement.
The Frontier system ousted the two-year top spot holder Fugaku system from the Riken Center for Computational Science in Japan, a supercomputer which has a theoretical peak above one exaflops that it has yet to demonstrate on the HPL benchmark test.
Once fully operational later this year, researchers will be able to access the Frontier system, marking a milestone for the three-year long collaboration among experts from the Department of Energy, HPE, and AMD.
“Scientists and engineers from around the world will put these extraordinary computing speeds to work to solve some of the most challenging questions of our era, and many will begin their exploration on Day One,” ORNL associate lab director for computing and computational sciences Jeff Nichols remarked.
In the biannual energy efficiency ranking of supercomputers called Green500, Frontier’s test and development system (TDS) also managed to secure the top spot by delivering 62.68 gigaflops/watt power-efficiency in a single cabinet of 3rd Gen AMD EPYC processors and AMD Instinct MI250x accelerators.
AMD said the dominant presence of its solutions in the HPC industry signifies its growth, powering a total of 94 systems from the Top500 list which is an increase to 95% year-over-year, and four efficient supercomputers on the Green500 list.
“Innovation and delivering more performance and efficiency for supercomputers is critical to addressing the world’s most complex challenges. AMD EPYC processors and AMD Instinct accelerators continue to push the envelope in high performance computing, providing the performance needed to advance scientific discoveries,” said Forrest Norrod, senior vice president and general manager for the Data Center Solutions Group at AMD.
Meanwhile, Oak Ridge Leadership Computing Facility science director Bronson Messer noted that the collaboration with AMD was a critical component in deploying what is now considered as the world’s leading platform for computational science.
“The Frontier supercomputer taps into the combined performance of enhanced AMD CPUs and AMD Instinct accelerators, along with an enhanced AMD ROCm 5 open software platform, to deliver the performance researchers need to carry out scientific research for the good of all mankind,” he explained.
AMD’s HPC business currently powers the supercomputer of the National Science and Technology Development Agency in Thailand, a new HPC cluster called Ascend which was unveiled by the Ohio Supercomputer Center, and the Microsoft Azure HBv3 virtual machines by electric supercar innovator Rimac.
