News
IndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value ChainIndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value Chain
EveryChip

Supercomputers

Chips built for or used in the TOP500 machines. 91 chips on EveryChip so far.

Supercomputers join thousands of chips to solve science and engineering problems: weather and climate, nuclear physics, drug discovery and, more and more, AI. The TOP500 list, published twice a year since 1993, ranks them by speed on a standard maths benchmark. Most top machines now pair server CPUs with GPU accelerators. Frontier, in 2022, was the first to pass one exaflop, a billion billion calculations a second.

From special chips to GPUs

Early supercomputers used custom processors. In 2008 IBM's Roadrunner, built with Cell-based chips and AMD Opterons, was the first to pass one petaflop. China's Tianhe-2 used Intel Xeon Phi accelerators, and the Sunway TaihuLight its own ShenWei chips. Japan's Fugaku, number one in 2020, used Fujitsu's Arm-based A64FX without separate accelerators.

The exascale machines

Frontier, at Oak Ridge National Laboratory in the United States, uses AMD Epyc CPUs and Instinct MI250X GPUs and reached first place in June 2022. El Capitan, at Lawrence Livermore, took first place in November 2024 with AMD's MI300A, which combines CPU and GPU chiplets in one package. India's AIRAWAT and PARAM machines use Nvidia GPUs and Intel or AMD CPUs.

Famous supercomputers chips

Further reading

🇺🇸 United StatesAMD Instinct AMD's data centre GPU line for deep learning and HPCGPU · GPU architecture · 2017 🇺🇸 United StatesXeon Phi Intel's x86 manycore coprocessor family for supercomputers and HPCCPU · x86 · 2010 CDNA 🇺🇸 United StatesCDNA AMD microarchitecture for data centre GPU acceleratorsGPU · GPU architecture · 2020 🇺🇸 United StatesHopper Nvidia's datacentre GPU architecture, first shipping in the H100 acceleratorGPU · GPU architecture · 2022 Volta 🇺🇸 United StatesVolta Nvidia's 2017 GPU architecture, first with Tensor Cores for deep learning.GPU · GPU architecture · 2017 Nvidia H100 🇺🇸 United StatesNvidia H100 AI and HPC accelerator based on the Hopper architectureGPU · GPU architecture GB200 Grace Blackwell 🇺🇸 United StatesGB200 Grace Blackwell Superchip Nvidia AI superchip linking Grace CPUs and Blackwell GPUsSoC · Arm Nvidia H200 🇺🇸 United StatesNvidia H200 AI GPU with 141 GB of HBM3e memoryGPU · GPU architecture · 2023 Tesla V100 🇺🇸 United StatesTesla V100 Nvidia Volta GPU for AI and HPC data centresGPU · GPU architecture · 2017 Intel Data Center GPU 🇺🇸 United StatesIntel Data Center GPU Max Data centre GPU for HPC and AI workloadsGPU · GPU architecture AMD Instinct MI250X 🇺🇸 United StatesAMD Instinct MI250X Dual-GPU accelerator for supercomputersGPU · GPU architecture AMD Instinct MI100 🇺🇸 United States and 1 moreAMD Instinct MI100 CDNA architecture GPU for high-performance computing and AI workloads.GPU · GPU architecture · 2020 AMD Instinct MI300X 🇺🇸 United StatesAMD Instinct MI300X AI accelerator GPU with 192GB HBM3 memoryGPU · GPU architecture · 2023 AMD Instinct MI300A 🇺🇸 United StatesAMD Instinct MI300A Modular APU combining CPU and GPU tiles for high-performance computing.SoC · x86 · 2023 A100 🇺🇸 United StatesA100 Data centre GPU for AI and high-performance computing.GPU · GPU architecture · 2020 Geometric Arithmetic P WorldGeometric Arithmetic Parallel Processor A mesh-connected array of single-bit SIMD processing elements for parallel computing.AI accelerator · Other · 1981 🇺🇸 United StatesHopper Nvidia datacentre GPU microarchitecture for AI and high-performance computing.GPU · GPU architecture · 2022 CSX700 🇬🇧 United KingdomCSX700 192-core SIMD coprocessor for high-performance computingAI accelerator · Other · 2001

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the chip of the day, new pages on EveryChip, and the conferences and anniversaries of the week ahead.