AMD-Powered Frontier Supercomputer Breaks the Exascale Barrier, Now Fastest in the World

AMD-Powered Frontier Supercomputer top500 us amd powered japan fugaku alcorn tom's hardware

That’s great news! The achievement of breaking the exascale barrier and becoming the fastest supercomputer in the world is a significant milestone for the field of high-performance computing. The AMD-powered Frontier supercomputer, which is based at the Oak Ridge National Laboratory, has been designed to deliver groundbreaking research in a range of fields, including energy, climate, and human health.

The Frontier supercomputer is powered by over 700,000 AMD EPYC CPU cores and over 1000 AMD Radeon Instinct GPUs, delivering a peak performance of 1.5 exaflops (or one quintillion calculations per second). This level of computing power will enable researchers to tackle some of the world’s most complex problems, from understanding the mechanisms of disease to developing more efficient and sustainable energy sources.

The AMD EPYC CPUs and Radeon Instinct GPUs are designed to work together seamlessly, delivering a powerful and flexible platform for scientific computing. With the ability to process massive amounts of data quickly and accurately, top500 us amdpowered japan fugakualcorn tomhardware Frontier will be a game-changer in many areas of research.

Overall, the breakthrough achieved by the Frontier supercomputer is a significant step forward in the field of high-performance computing, and it is exciting to see what new discoveries and advancements will be made possible by this cutting-edge technology.

  • Processors: Over 700,000 AMD EPYC CPU cores
  • GPUs: Over 1000 AMD Radeon Instinct GPUs
  • Peak Performance: 1.5 exaflops (one quintillion calculations per second)
  • Memory: 2.7 petabytes of total system memory (both CPU and GPU memory combined)
  • Storage: Over 100 petabytes of usable high-performance storage
  • Interconnect: Cray Slingshot interconnect with up to 1.6 terabytes per second of bandwidth per node
  • Power: 30 megawatts of power capacity
  • Operating System: A custom version of Red Hat Enterprise Linux

The combination of AMD EPYC CPUs and Radeon Instinct GPUs provides a powerful computing platform that can handle a wide range of workloads, from machine learning and artificial intelligence to modeling and simulation. The Cray Slingshot interconnect ensures high-speed data transfer between the CPUs and GPUs, while the large amounts of memory and storage make it possible to work with massive datasets.

Overall, the Frontier supercomputer is a highly advanced system that represents a significant leap forward in high-performance computing. It is designed to deliver groundbreaking research in a variety of fields, from energy and climate to human health and beyond.

Frontier Supercomputer Processor.

The AMD-powered Frontier supercomputer uses over 700,000 AMD EPYC CPU cores as its main processor. The specific model of EPYC CPU used in the Frontier supercomputer is not publicly disclosed, but it is likely to be one of the latest generations of AMD’s EPYC processors, which are designed for high-performance computing workloads.

AMD’s EPYC processors are based on the Zen architecture and are known for their high core count, high memory bandwidth, and strong performance in a variety of workloads. They are particularly well-suited for scientific computing, machine learning, and other data-intensive tasks that require a lot of computational power. The EPYC processors used in the Frontier supercomputer are likely to be customized to meet the specific requirements of the system, such as high memory bandwidth and low latency.

Overall, the use of AMD EPYC CPUs in the Frontier supercomputer is a key factor in the system’s ability to deliver groundbreaking research in a wide range of fields. With over 700,000 CPU cores working together, the Frontier supercomputer is capable of processing massive amounts of data and running complex simulations at unprecedented speeds.

Frontier Supercomputer Ram

The Frontier supercomputer has a total system memory of 2.7 petabytes, which is a combination of both CPU and GPU memory. The specific amount of RAM per node is not publicly disclosed, but it is likely to vary depending on the specific configuration of each node.

It’s worth noting that the AMD EPYC processors used in the Frontier supercomputer support high memory bandwidth and can address large amounts of memory, which is critical for scientific computing workloads that involve processing and analyzing large datasets.

The AMD Radeon Instinct GPUs used in the Frontier supercomputer also have high-bandwidth memory (HBM2) directly attached to the GPU, which provides high-speed access to data and helps to reduce data transfer bottlenecks between the CPU and GPU.

Overall, the large amount of memory in the Frontier supercomputer, combined with its high-performance CPU and GPU processors, makes it possible to work with massive datasets and run complex simulations at unprecedented speeds.

Frontier Supercomputer Storage

The Frontier supercomputer has over 100 petabytes of usable high-performance storage. The storage system is based on a high-speed parallel file system designed to provide high throughput and low latency for data-intensive workloads. The specific storage system used in the Frontier supercomputer is not publicly disclosed, but it is likely to be a custom solution designed to meet the specific requirements of the system.

The high-performance storage in the Frontier supercomputer is critical for scientific computing workloads that involve processing and analyzing large datasets. With over 100 petabytes of storage, researchers can store and analyze massive amounts of data generated by simulations, experiments, and other scientific activities.

Overall, the combination of high-performance storage, high-memory bandwidth, and high-speed CPU and GPU processors makes the Frontier supercomputer a powerful platform for scientific computing that can handle the most demanding workloads.

AMD El Capitan.

AMD El Capitan is a high-performance computing (HPC) system that is currently under development for the Lawrence Livermore National Laboratory (LLNL) in the United States. The system is being designed in collaboration with AMD, Cray (now part of Hewlett Packard Enterprise), and the U.S. Department of Energy.

Once completed, AMD El Capitan is expected to be one of the world’s fastest supercomputers, with a target performance of over 2 exaflops (two quintillion calculations per second). The system will be powered by the latest generation of AMD EPYC CPUs and Radeon Instinct GPUs, and will be built using Cray’s Shasta architecture and Slingshot interconnect technology.

The AMD EPYC CPUs and Radeon Instinct GPUs in the El Capitan system are expected to deliver industry-leading performance in a range of workloads, including scientific computing, machine learning, and data analytics. The system will be used by researchers and scientists at LLNL to perform simulations and modeling across a wide range of fields, including energy, climate, and national security.

Overall, AMD El Capitan represents a major step forward in high-performance computing and is expected to deliver groundbreaking research and insights into some of the world’s most complex problems. The system is scheduled to be delivered in 2023, and its performance is expected to continue pushing the boundaries of what is possible in scientific computing.

Frontier Supercomputer Release Date.

The Frontier supercomputer is currently under development and is expected to be delivered to the Oak Ridge National Laboratory in Tennessee, USA in 2021. The system is then expected to undergo acceptance testing and tuning before being made available for use by researchers and scientists.

The Frontier supercomputer is a collaborative project between the US Department of Energy, Oak Ridge National Laboratory, AMD, and Cray (now part of Hewlett Packard Enterprise), and is expected to be one of the world’s most powerful supercomputers when it becomes operational.

The system is being designed to support a wide range of scientific computing workloads, including simulations and modeling in areas such as materials science, chemistry, and astrophysics. The large amount of computing power and memory in the system is expected to enable researchers to tackle some of the world’s most complex problems and make breakthroughs in a variety of fields.

Overall, the Frontier supercomputer is an exciting development in the world of high-performance computing and is expected to deliver groundbreaking research and insights into a wide range of scientific fields.

Pros And Cons.

Pros:
  • The Frontier supercomputer is expected to be one of the fastest supercomputers in the world, providing researchers and scientists with unprecedented computing power to tackle complex problems.
  • The system is designed to support a wide range of data-intensive workloads in areas such as materials science, chemistry, and astrophysics.
  • With a total system memory of 2.7 petabytes and over 100 petabytes of high-performance storage, the Frontier supercomputer is well-suited to handle large datasets and complex simulations.
  • The collaboration between AMD, Cray, and the US Department of Energy has resulted in a powerful and innovative system that will push the boundaries of what is possible in scientific computing.

Cons:

  • The development and operation of supercomputers such as Frontier is very expensive and requires a significant investment of resources.
  • The high energy consumption and cooling requirements of such systems can have a significant impact on the environment and contribute to climate change.
  • The complexity of the system and the software required to run on it can make it challenging for researchers and scientists to use effectively, requiring specialized expertise and training.
  • The high-performance computing market is highly competitive, and other companies and countries are also investing in the development of powerful supercomputers, which could diminish the competitive advantage of the Frontier system in the future.

Conclusion.

The Frontier supercomputer, powered by over 700,000 AMD EPYC CPU cores and AMD Radeon Instinct GPUs, is a major milestone in high-performance computing. With a total system memory of 2.7 petabytes and over 100 petabytes of high-performance storage, the Frontier supercomputer is designed to support a wide range of data-intensive workloads in areas such as materials science, chemistry, and astrophysics.

The Frontier supercomputer is currently under development and is expected to be delivered to the Oak Ridge National Laboratory in 2021. Once operational, the system is expected to be one of the fastest supercomputers in the world and will be used by researchers and scientists to perform simulations and modeling across a wide range of fields.

In addition to the Frontier supercomputer, AMD is also working on the development of the El Capitan supercomputer, which is expected to be one of the world’s fastest when it becomes operational. These systems represent a major step forward in high-performance computing and are expected to deliver groundbreaking research and insights into some of the world’s most complex problems.