EN
PL

Tag: AMD EPYC

HPE FRONTIER - The world's most powerful supercomputer.



/HPE FRONTIER - The world's most powerful supercomputer.

The Hewlett Packard Enterprise (HPE) Frontier supercomputer is one of the most powerful supercomputers in the world. It was developed in cooperation with the US Department of Energy (DOE) and is located at Oak Ridge National Laboratory in Tennessee, USA. The Frontier supercomputer was designed to help scientists solve the most complex and pressing problems in a variety of fields, including medicine, climate science and energy.

Tech specs

The HPE Frontier supercomputer is built on the HPE Cray EX supercomputer architecture, which consists of a combination of AMD EPYC processors and NVIDIA A100 GPUs. Its peak performance is 1.5 exaflops (one quintillion floating-point operations per second) and can perform more than 50,000 trillion calculations per second. The system has 100 petabytes of storage and can transfer data at up to 4.4 terabytes per second.

Applications

The HPE Frontier supercomputer is used for a wide range of applications, including climate modeling, materials science and astrophysics. It is also being used to develop new drugs and treatments for diseases such as cancer and COVID-19.

Climate modeling

The Frontier supercomputer is being used to improve our understanding of the Earth's climate system and to develop more accurate climate models. This will help scientists predict the impacts of climate change and develop mitigation strategies.

Development of materials

The supercomputer is also being used to model and simulate the behavior of materials at the atomic and molecular levels. This will help scientists develop new materials with unique properties, such as increased strength, durability and conductivity.

Astrophysics

The Frontier supercomputer is being used to simulate the behavior of the universe on a large scale, including the formation of galaxies and the evolution of black holes. This will help scientists better understand the nature of the universe and the forces that govern it.

Medical developments

The supercomputer is being used to simulate the behavior of biological molecules, such as proteins and enzymes, in order to develop new drugs and treatments for diseases. This will help scientists identify new targets for drug development and develop more effective treatments for a wide range of diseases.

Summary

The HPE Frontier supercomputer represents a major step forward in the development of high-performance computing. Its unprecedented computing power and storage capacity make it a valuable tool for researchers in many fields. Its ability to simulate complex systems at a high level of detail helps us better understand the world around us and develop solutions to some of the most pressing challenges facing humanity.


HPE FRONTIER – World’s Most Powerful Supercomputer



/HPE FRONTIER – World’s Most Powerful Supercomputer

The Hewlett Packard Enterprise (HPE) Frontier supercomputer is one of the most powerful supercomputers in the world. It was developed in collaboration with the US Department of Energy (DOE) and is located at Oak Ridge National Laboratory in Tennessee, US. The Frontier supercomputer is designed to help scientists solve the most complex and pressing problems in a variety of fields, including medicine, climate science and energy.

Technical specifications

The HPE Frontier supercomputer is built on the HPE Cray EX supercomputer architecture, which consists of a combination of AMD EPYC processors and NVIDIA A100 GPUs. It has a peak performance of 1.5 exaflops (one quintillion floating point operations per second) and can perform more than 50,000 trillion calculations per second. The system has 100 petabytes of storage and can transfer data at up to 4.4 terabytes per second.

Applications

The HPE Frontier supercomputer is used for a wide range of applications, including climate modelling, materials science and astrophysics. It is also used to develop new drugs and treatments for diseases such as cancer and COVID-19.

Climate modelling

The Frontier supercomputer is being used to improve our understanding of the Earth’s climate system and to develop more accurate climate models. This will help scientists predict the impacts of climate change and develop mitigation strategies.

Materials science

The supercomputer is also being used to model and simulate the behaviour of materials at the atomic and molecular level. This will help scientists develop new materials with unique properties such as increased strength, durability and conductivity.

Astrophysics

The Frontier supercomputer is being used to simulate the large-scale behaviour of the universe, including the formation of galaxies and the evolution of black holes. This will help scientists better understand the nature of the universe and the forces that govern it.

Drug development

The supercomputer is being used to simulate the behaviour of biological molecules, such as proteins and enzymes, in order to develop new drugs and treatments for diseases. This will help scientists identify new targets for drug development and develop more effective treatments for a wide range of diseases.

Summary

The HPE Frontier supercomputer represents a major step forward in the development of high performance computing. Its unprecedented computing power and storage capacity make it a valuable tool for researchers in many fields. Its ability to simulate complex systems at a high level of detail helps us better understand the world around us and develop solutions to some of the most pressing challenges facing humanity.


Supermicro Ultra SuperServer



/Supermicro Ultra SuperServer

Supermicro Ultra SuperServer® is Supermicro’s 11th generation high performance general purpose server. The Ultra is designed to provide the highest performance, flexibility, scalability and serviceability in demanding IT environments, as well as to power critical corporate workloads.

Unmatched performance: support for two 2nd Generation Intel® Xeon® Scalable processors with up to 28 cores per socket and up to 6TB of ECC DDR4 memory in 24 DIMM slots with Intel® Optane “¢ DCPMM support, the Ultra is designed to support demanding and complex loads. The Ultra is available in NVMe all-flash configurations where users can benefit from reduced latency and increased IOP. With NVMe, it is possible to increase storage latency up to 7x and increase throughput by up to 6x.1 The ROI benefits of NVMe deployments are immediate and significant.

Exceptional flexibility: discover the freedom to adapt to different loads with the versatile Supermicro Ultra system. Improve your server environment with the perfect combination of computing power, memory and storage performance, network flexibility and serviceability. This highly scalable system provides excellent expansion and storage options thanks to our patented vertical system. With support for multiple PCIe add-on cards, the Ultra Future protects your business against ever-changing computation and storage. This Ultra server is designed to handle any workload in any number of demanding environments.

Continuous reliability and serviceability: Achieve higher levels of high availability and data storage with the latest Intel® Xeon® Scalable processors, ECC DDR4 memory modules, NVMe-enabled disk bays, and energy-efficient redundant power supplies. Designed from the ground up as an enterprise class, the Ultra is fully equipped with energy-efficient components and built-in redundancy.

Supermicro Ultra Servers are designed to give the greatest possible power, flexibility and scalability. It is a great choice to meet the most demanding operations in Enterprise, Data Center and Cloud Computing environments.


Supermicro Ultra SuperServer



/Supermicro Ultra SuperServer

Supermicro Ultra SuperServer® is Supermicro's 11th-generation, high-performance general-purpose server. Ultra is designed to deliver superior performance, flexibility, scalability and serviceability in demanding IT environments, and to power critical enterprise workloads.

Unmatched performance: support for two scalable second-generation Intel® Xeon® processors with up to 28 cores per socket and up to 6 TB of ECC DDR4 memory in 24 DIMM slots with support for Intel® Optane Technology makes Ultra designed to handle demanding and complex workloads . Ultra is available in NVMe all-flash configurations, where users can benefit from reduced latency and increased IOP. With NVMe, it is possible to increase storage latency by up to 7 times and increase throughput by up to 6 times.1 The ROI benefits from NVMe deployments are immediate and significant.

Exceptional flexibility: Discover the freedom to adapt to different workloads with the versatile Supermicro Ultra system. Enhance your server environment with the perfect combination of computing power, memory and storage performance, network flexibility and serviceability. This highly scalable system provides excellent expansion and storage options with our patented vertical system. With support for multiple additional PCIe cards, the Ultra future-proofs your business with ever-changing computing and storage. This Ultra server is designed to handle any workload in any number of demanding environments.

Continued reliability and ease of service: Achieve higher levels of high availability and storage with the latest scalable Intel® Xeon® processors, ECC DDR4 memory modules, hot-swappable drive bays with NVMe support and energy-efficient redundant power supplies. Designed from the ground up as enterprise grade, Ultra is fully equipped with energy-efficient components and built-in redundancy.


AMD EPYC – Does Intel still have a chance?



/AMD EPYC – Does Intel still have a chance?

The AMD EPYC 7643 from the Milan family was noticed in the Geekbench benchmark database. There it showed the power of the Zen 3 microarchitecture-based cores. It turns out that one processor scored better in the test compared to the platform consisting of dual Intel Xeon.

The EPYC 7643 is one of the new AMD Milan series processors. It is a 48-core chip with support for up to 96 threads, which is based on the Zen 3 microarchitecture. It was noticed in the Geekbench benchmark database. Here we have the opportunity to learn about CPU performance.

AMD EPYC 7643 in the test of one core in Geekbench 4 dialed 5850 points. In the case of many cores, it was over 121 thousand. points. We know that the basic clock speed here is 2.3 GHz. However, in the case of increased power, it is a maximum of 3.6 GHz.

The performance of this AMD EPYC chip is really high. For comparison, the dual Intel Xeon Platinum 8276 with 56 cores and supporting 112 threads in the Single Core test sets 4913 points. However, in the Multi-Core it is 112,457 points. So the new processor from the competition is better.