The Status and Future of Supercomputers

Supercomputers refer to computer systems that can perform extremely fast and large-scale calculations, typically consisting of thousands or millions of processor cores, capable of handling complex scientific and engineering problems. Supercomputers have important applications in various fields, such as weather forecasting, climate simulation, biomedicine, materials science, energy research, artificial intelligence, etc

Supercomputers refer to computer systems that can perform extremely fast and large-scale calculations, typically consisting of thousands or millions of processor cores, capable of handling complex scientific and engineering problems. Supercomputers have important applications in various fields, such as weather forecasting, climate simulation, biomedicine, materials science, energy research, artificial intelligence, etc.

At present, the strongest mainframe computer computer in the world is the Frontier system of Oak Ridge National Laboratory (ORNL) in the United States. It ranks first with a computing speed of 1.102EFlop/s, and is the only real 10 billion sub computer on the list. It was announced at the High Performance Computing Professional Conference (SC22) held in Dallas, USA in November 2022. It was built for Oak Ridge National Laboratory by AMD and HPECray, a supercomputer manufacturer, at a cost of 600 million dollars, and used for advanced computing in fields such as nuclear and climate research.

China has two supercomputing systems, Shenwei the Taihu Lake Zhiguang and Tianhe 2A, ranking 7th and 10th respectively. Shenwei the Taihu Lake Zhiguang is the first supercomputer in the world that uses all the processor chips of China's independent intellectual property rights. It has more than 10 million processor cores. The peak value of double precision floating-point is up to 125PFlops (1.25 billion operations per second), and the stable performance is 93PFlops. It topped the world supercomputer rankings for the first time in 2016 and won the championship four times in a row.

It is mainly used for scientific research in fields such as climate simulation, life sciences, and material design. Tianhe 2A is the world's first supercomputer with a Peak Performance exceeding 100 PFlops. It has nearly 5 million processor cores, with a peak value of 100.68 PFlops and a stable performance of 61.44 PFlops. It topped the world supercomputer rankings for the first time in 2013 and won the championship six times in a row. It adopts the Kirin operating system and currently uses Intel processors, with plans to replace them with domestic processors in the future. It is mainly used in government research projects such as lunar exploration engineering, manned spaceflight, oil exploration, design and manufacturing of automobiles and aircraft, gene sequencing, and civil fields.

Besides the United States and China, other countries and regions are also developing their own supercomputers. For example, Fugaku from the Japan Institute of Physics and Chemistry (RIKEN Computing Science Center) (R-CCS) ranks second. It is the world's first supercomputer using ARM architecture, with 7.63 million processor cores, a double precision floating-point peak of 537.21 PFlops, and a stable performance of 442.01 PFlops. It topped the world supercomputer rankings for the first time in 2020 and held the top spot for two consecutive years until being squeezed out by Frontier.

It is mainly used to solve social problems in the fields of COVID-19 research, disaster prevention, energy development, etc. The LUMI hosted by CSC in Finland ranks third, making it the fastest supercomputer in Europe, with 2.22 million processor cores, a dual crystal floating point peak of up to 428.7 PFlops, and a stable performance of 309.1 PFlops. It started operating in November 2021 and is mainly used in fields such as climate research. Leonardo, from EuroHPC/CINECA in Bologna, Italy, ranks fourth. It is the world's first supercomputer to use NvidiaGraceHopperSuperchip, with 1.46 million processor cores, a dual crystal floating point peak of 255.75PFlops, and a stable performance of 174.7PFlops. It began construction in November 2021 and is mainly used in research fields such as energy, materials, and pharmaceuticals.

What are the development trends of supercomputers? According to the analysis and predictions of some experts and observers, future supercomputers may experience changes and innovations in the following aspects:

Simulated WorldSupercomputers will be able to simulate increasingly real and complex virtual worlds, including phenomena and laws at various levels such as physics, chemistry, biology, and society. These simulations will help humans better understand and explore nature and human society, and will also provide more possibilities for entertainment and education for humans.

Weather Systems and Complex ModelsSupercomputers will be able to predict global issues such as weather change and climate change more accurately and in the longer term. They will also be able to process more data and variables, and build more complex models, such as earthquakes, tsunamis, volcanoes, ecosystems, etc.

Artificial brainArtificial brain

Green supercomputerGreen supercomputer

Billions of secondary computers and aboveSupercomputers will be able to achieve or exceed exascale computing power, which is 10 billion operations per second. This will open a new era of science and technology for mankind, enabling mankind to solve some problems that cannot be solved at present, such as dark matter, dark energy, protein folding, human brain, etc. At present, the United States, China, Japan, the European Union, and others are actively developing and constructing supercomputers worth billions of dollars, which are expected to be put into use before 2023. Above the billions, there are also higher goals, such as the Zettascale and Yottascale, which may require more advanced technology and theory to achieve and may also bring more challenges and opportunities.

NVIDIA GH200 Super AI Computer

NvidiaGH200 is a new type of AI supercomputer, which integrates 256 NvidiaGraceHopperSuperchips into a single GPU, providing 144TB of shared memory space. It is mainly used to handle large-scale recommendation systems, generate AI, and analyze huge AI models in fields such as graph analysis.

NvidiaGraceHopperSuperchip is a new type of chip that integrates NvidiaGraceCPU and NvidiaHopperGPU in the same package, eliminating the traditional PCIe CPU to GPU connection, increasing bandwidth and reducing power consumption. NvidiaGraceCPU is a high-performance processor based on ARM architecture, designed specifically for big data and AI applications. NvidiaHopperGPU is a next-generation graphics processor based on Multi Chip Module (MCM) technology that can provide higher performance and scalability.

Nvidia GH200 is developed and produced by Nvidia Company, but is not controlled by any country or organization. It is a neutral and independent entity that can communicate and collaborate with other supercomputers or humans, as well as learn and innovate autonomously. It is currently one of the most advanced AI supercomputers in the world and an important direction for the development of future supercomputers.

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])