Cerebras Trillion Transistor AI Wafer-Chip Crushes GPU Supercomputer by 200 Times

Researchers at the National Energy Technology Laboratory (NETL), Cerebras showed that a single wafer-scale Cerebras CS-1 can outperform one of the fastest supercomputers in the US by more than 200 X. They had 0.86 PetaFLOPS of performance on the single wafer system. The problem was to solve a large, sparse, structured system of linear equations …

Read more

World’s Fastest Supercomputer is Japan Fugaku at 415 Petaflops

The Fugaku supercomputer is the new fastest supercomputer and is almost 3 times faster than the Summit Supercomputer. Fugaku High Performance Linpack (HPL) was 415.5 petaflops. Fugaku uses Fujitsu’s 48-core A64FX SoC. It is the first world fastest computer to use ARM processors. In single or further reduced precision, which are often used in machine …

Read more

Dielectric Supercooled Supercomputers at 50 Petaflops and Soon 250 Petaflops

DownUnder GeoSolutions (DUG) has four data centres have been recently upgraded and DUG’s global compute power now exceeds 50 PF (single-precision petaflops). Petaflop supercomputers generate a lot of heat and need efficient cooling technology. Installations such as these are traditionally air-cooled, which is inefficient, expensive, and environmentally unfriendly. DUG has developed (Patent Publication WA 2017/091862 …

Read more

Graphcore Chips Could Speed Up AI by 100 Times

Graphcore’s IPU (Intelligence Processing Unit) accelerators and Poplar software together make the fastest and most flexible platform for current and future machine intelligence applications, lowering the cost of AI in the cloud and datacenter, improving performance and efficiency by between 10x to 100x. Graphcore recently closed a new $200 million funding round which values the …

Read more

Will 1000 ExaFlop Supercomputers Come from Brute Force Scaling or New Technology?

Next year the world will have its first ExaFlop Supercomputer. This supercomputer will likely be built in China. The USA will follow about 12 months after with the A21 ExaFlop supercomputer in 2021. Ten researchers from China’s National University of Defense Technology (NUDT) recently published a paper that describes pushing current technologies to 1000 times …

Read more

IBM Investing $2 Billion in an AI Center and Targets 1000 Times Improvement by 2029

IBM will invest $2 billion to develop artificial intelligence hardware and boost AI performance by 1000 times over the next ten years. IBM is partnering with State University of New York to develop an AI Hardware Center at SUNY Polytechnic Institute in Albany. New York will also provide a subsidy of $300 million. The IBM …

Read more

Cray Shasta supercomputer in 2020 with about 100 petaflops

Cray is building new Shasta Supercomputers. The first will be built for $146 million at Berkeley National Labs. It will use AMD Epyc processor and the next-generation Einstein Nvidia Tesla GPU accelerator. The “Perlmutter” NERSC-9 supercomputer will be installed in late 2020. Cray has built a new interconnect, called “Slingshot”. The current Cori NERSC-8 machine …

Read more

US now has first and third most powerful supercomputers

Lawrence Livermore National Laboratory (LLNL) unveiled Sierra the world’s third fastest supercomputer. Sierra has a peak performance of 125 petaFLOPS — 125 quadrillion floating-point operations per second. Early indications using existing codes and benchmark tests are promising, demonstrating as predicted that Sierra can perform most required calculations far more efficiently in terms of cost and …

Read more

IBM has most powerful 200 Petaflop supercomputer and will sell same architecture commercially

Oak Ridge National Labs has unveiled the 200 petaflop Summit supercomputer. The USA has retaken the world’s most powerful computer. They have smartest artificial intelligence (AI) supercomputer and reached a major computing milestone. The USA has the best supercomputer for the first time since 2013, when China introduced the 34 PetaFLOP NUDT Tianhe-2 and then …

Read more

German university getting multi-petaflop Cray CS500 supercomputer

Paderborn University (Germany) has selected a Cray CS500 cluster accelerated by FPGAs as the first phase of its Noctua multi-petaflop supercomputer. The new supercomputer will be comprised of 272 compute nodes, connected by Intel’s 100 Gbps Omni-Path fabric. Each node will be equipped with two 20-core Xeon Gold (Skylake) processors, yielding a system core count …

Read more