Nvidia vs Google: Who Wins the MLPerf Training Race?
MLPerf training benchmark scores reveal there are two contenders for world’s fastest AI supercomputer. Nvidia’s Selene uses 2048 A100 chips,…
Academia’s Fastest Supercomputer Delivers 700 PFLOPS of AI Performance
The University of Florida is building a supercomputer built around Nvidia GPUs using an architecture that helped vault Nvidia's own…
The World’s Combined Compute Power Takes on COVID-19
What tools can we deploy in the fight against Covid-19? Taken together, the world has enough sheer compute power to…
Europe Taps Supercomputers to Fight Pandemic
Some of the top supercomputers in Europe have been pledged to help tackle Covid-19 by joining the Covid-19 high-performance computing…
Supercomputer 2019 Exhibits the Race to Exascale
Lacking new top systems, the focus shifted to the next big systems for exascale computing. Intel, Nvidia, Arm, Cerebras and…
Nvidia Fulfils Promise to Support Arm
The GPU specialist introduced a reference design platform for GPU-accelerated Arm-based servers. Arm, Ampere, Cray, Fujitsu, HPE, and Marvell all…
Google Are Making Quantum Computing a Reality
A new Google computing experiment published in Nature has ushered in a new era of computing.
Cray Secures Another Exascale Deal
Cray won a $600 million deal to build the El Capitan exascale system, sweeping all three next-gen supercomputers for U.S.…
Intel Shows First Glimpse of Packaging Technologies on Roadmap
Intel's exascale supercomputer project may be the first to use one of three chip-stacking techniques Intel disclosed on its roadmap.
U.S. Enters Exascale Race
Race to build the worlds first exascale systems heats up as U.S. Department of Energy awards Intel and Cray a…