Supercomputers have held a position of huge esteem within the technology space. They are used for many things, including climate modeling, disease research, and nuclear science.
They are also used to run complicated simulations that involve many variables. Recently, supercomputers have been used to track the spread and mutation of the coronavirus.
Supercomputers are often ranked in terms of the number of computations per second. These so-called TeraFLOPS are units of computing power equal to one trillion (10¹²) floating-point operations per second.
What are the most powerful computers of all time? Find out below.
1. Fugaku
Location: Riken Center for Computational Science, Japan
Cores: 7,630,848
Memory: 5,087,232 GB
Processor: A64FX 48C 2.2GHz
Peak performance: 537,212 TFlops
This supercomputer, developed by Japan's state-backed Riken research institute, is the world's fastest for computing speed. named Fugaku after Mt. Fuji, this computer was jointly developed with Fujitsu Ltd. at the institute's facility in Kobe. It is currently involved in researching potential medicines to fight coronavirus.
2. Summit
Location: DOE/SC/Oak Ridge National Laboratory, US
Cores: 2,414,592
Memory: 2,801,664 GB
Processor: IBM POWER9 22C 3.07GHz
Peak Performance: 200,795 TFlops
Summit, launched in 2018, uses 4,608 computing nodes, each containing multiple IBM POWER9 CPUs and Nvidia Volta GPUs connected with Nvidia’s high-speed NVLink. Each node has over half a terabyte of coherent memory. The computer solves complex tasks in energy, artificial intelligence (AI), human health, and other areas of research.
3. Sierra
Location: DOE/NNSA/LLNL, US
Cores: 1,572,480
Memory:1,382,400 GB
Processor: IBM POWER9 22C 3.1GHz
Peak performance: 125,712 TFlop/s
This IBM-built supercomputer combines two types of processor chips — IBM’s Power 9 processors and Nvidia’s Volta graphics processing units. Nuclear weapon scientists use it in nuclear weapons simulations in place of underground testing.
It helps assess the performance of nuclear weapon systems as well as nuclear weapon science and engineering calculations.
4. Sunway TaihuLight
Location: National Supercomputing Center in Wuxi, China
Cores: 10,649,600
Memory: 1,310,720 GB
Processor: Sunway SW26010 260C 1.45GHz
Peak Performance: 125,436 TFlop/s
This Chinese supercomputer features 40,960 Chinese-designed SW26010 manycore 64-bit RISC processors based on the Sunway architecture. Each processor chip contains 256 processing cores.
The name is translated as divine power, the light of Taihu Lake. From June 2016 to June 2018, it was the world's fastest supercomputer. The computer is involved in climate, weather and earth systems modeling, life science research, advanced manufacturing, and data analytics.
5. Perlmutter
Location: DOE/SC/LBNL/NERSC, US
Cores: 706,304
Memory: 390,176 GB
Processor: AMD EPYC 7763 64C 2.45GHz
Peak Performance: 89,794.5 TFlop/s
This supercomputer is based on the HPE Cray “Shasta” platform and is a heterogeneous system with both GPU-accelerated and CPU-only nodes. It is named in honor of Nobel prize winner Saul Perlmutter.
Among its use cases are nuclear fusion simulations, climate projections, and material and biological research.
6. Selene
Location: NVIDIA Corporation, US
Cores: 555,520
Memory: 1,120,000 GB
Processor: AMD EPYC 7742 64C 2.25GHz
Peak Performance: 79,215 TFlop/s
Selene is a DGX SuperPOD powered by Nvidia’s A100 GPUs and AMD’s Epyc Rome CPUs within the DGX A100 form factor. It uses Mellanox HDR InfiniBand as the system network.
Selene’s platform supports a wide community of users, such as supercomputer-scale continuous integration for software, research, “big iron AI” work (e.g., Megatron, ASR), and automotive.
7. Tianhe-2A
Location: National Supercomputer Center in Guangzhou, China
Cores: 4,981,760
Memory: 2,277,376 GB
Processor: Intel Xeon E5-2692v2 12C 2.2GHz
Peak Performance: 100,679 TFlop/s
This supercomputer was developed by China's National University of Defense Technology (NUDT) and deployed at the National Supercomputer Center in Guangzhou, China.
When first deployed, it had 16,000 computer nodes, each comprising two Intel Ivy Bridge Xeon processors and three Xeon Phi coprocessor chips, one of the world’s largest deployments of Ivy Bridge and Xeon Phi chips.
It runs on Kylin Linux, a version of the operating system developed by China’s National University of Defense Technology (NUDT).
