Covering Disruptive Technology Powering Business in The Digital Age

Home > DTA news > News > Red Hat Enterprise Linux Powers the Future of Supercomputing
Red Hat Enterprise Linux Powers the Future of Supercomputing
July 3, 2020 News

 

Red Hat, Inc., the world’s leading provider of open source solutions, announced that Red Hat Enterprise Linux provides the operating system backbone for the top three supercomputers in the world and four out of the top 10, according to the newest TOP500 ranking. These rankings also show that the world’s leading enterprise Linux platform can deliver a foundation to meet even the most demanding computing environments, already serving as a catalyst for enterprise innovation across the hybrid cloud.

In the top ten of the current TOP500 list, Red Hat Enterprise Linux serves as the operating system for:

  • Fugaku, the top-ranked supercomputer in the world based at RIKEN Centre for Computational Sciences in Kobe, Japan.
  • Summit, the number two-ranked supercomputer based at Oak Ridge National Laboratory in Oak Ridge, Tennessee.
  • Sierra, the third-ranked supercomputer globally based at Lawrence Livermore National Laboratory in Livermore, California.
  • Marconi-100, the ninth-ranked supercomputer installed at CINECA research centre in Italy.

 

High-performance computing across architectures

Red Hat Enterprise Linux is engineered to deliver a consistent, standardised and high-performance experience across nearly any certified architecture and hardware configuration. Providing a predictable and reliable interface regardless of the underlying hardware, these same exacting standards and consistency are also brought to supercomputing environments.

Fugaku is the first Arm-based system to take first place on the TOP500 list, highlighting Red Hat’s commitment to the Arm ecosystem from the data centre to the high-performance computing laboratory. Sierra, Summit and Marconi-100 all boast IBM POWER9-based infrastructure with NVIDIA GPUs; combined, these four systems produce more than 680 petaflops of processing power to fuel a broad range of scientific research applications.

According to the Green500 list, in addition to enabling this immense computation power, Red Hat Enterprise Linux also underpins six out of the top 10 most power-efficient supercomputers on the planet. Systems on the list are measured in terms of both performance results and the power consumed achieving those. When it comes to sustainable supercomputing the premium is put on finding a balanced approach for the most energy-efficient performance.

In the top ten of the Green500 list, Red Hat Enterprise Linux serves as the operating system for:

  • A64FX prototype, at number four, was created as the prototype system to test and develop the Fugaku supercomputer and is based at Fujitsu’s plant in Numazu, Japan.
  • AIMOS, the number five supercomputer on the Green500 list based at Rensselaer Polytechnic Institute in Troy, New York.
  • Satori, the seventh-ranked most power-efficient system in the world, installed at MIT Massachusetts Green High Performance Computing Center (MGHPCC) in Holyoke, Massachusetts. It serves as the home for the Mass Open Cloud (MOC) project, where Red Hat supports a number of activities.
  • Summit at number eight.
  • Fugaku at number nine.
  • Marconi-100 at number ten.

 

From the laboratory to the data centre and beyond

Modern supercomputers are no longer purpose-built monoliths constructed from expensive bespoke components. Each supercomputer deployment powered by Red Hat Enterprise Linux uses hardware that can be purchased and integrated into any data centre, making it feasible for organisations to use enterprise systems that are similar to those breaking scientific barriers. Red Hat Enterprise Linux provides the common control plane for supercomputers to be run, managed and maintained in the same manner as traditional IT systems regardless of the underlying hardware.

Red Hat Enterprise Linux also opens supercomputing applications up to advancements in enterprise IT, including Linux containers. Red Hat is helping to drive advancements to make Podman, Skopeo and Buildah, components of Red Hat’s distributed container toolkit, more accessible for building and deploying containerised supercomputing applications while working closely in open source communities with organisations like the Supercomputing Containers project.

 

Supporting Quote

Stefanie Chiras, vice president and general manager, Red Hat Enterprise Linux Business Unit, Red Hat

“Supercomputing is no longer the domain of custom-built hardware and software. With the proliferation of Linux across architectures, high-performance computing has now become about delivering scalable computational power to fuel scientific breakthroughs. Red Hat Enterprise Linux already provides the foundation for innovation to the enterprise world and, with the recent results of the TOP500 list, we’re pleased to now provide this same accessible, flexible and open platform to the world’s fastest and some of the most power-efficient computers.”

Steve Conway, senior adviser, HPC Market Dynamics, Hyperion Research

“Every one of the world’s TOP500 most powerful supercomputers runs on Linux, and a recent study we did confirmed that Red Hat is the most popular vendor-supported Linux solution in the global high performance computing market. Red Hat Enterprise Linux is designed to run seamlessly on a variety of architectures underlying leading supercomputers, playing an important part in driving HPC into new markets and use cases, including AI, enterprise computing, quantum computing and cloud computing.”

Satoshi Matsuoka, director, RIKEN Center for Computational Science (R-CCS); professor, Department of Mathematical and Computing Sciences, Tokyo Institute of Technology

“Fugaku represents a new wave of supercomputing, delivering the performance, scale and efficiency to help create new scientific breakthroughs and further drive research innovation. A key consideration of the project was to deliver an open source software stack, starting with the addition of Red Hat Enterprise Linux. With Red Hat Enterprise Linux running on Arm-based processors, we have been able to make supercomputing resources accessible and manageable by our distributed community of scientists and simplify development and deployment of a broader range of workloads and applications.”

Professor Jack Dongarra, University of Tennessee, Oak Ridge National Laboratory, and the University of Manchester

“Computing innovation and scientific advancement is not done in a vacuum – the supercomputing community, from laboratories to the vendor ecosystem, collaborates to help drive breakthroughs at both the architectural and the research level. Red Hat is a key part of this global community, helping to deliver a standards-based, open control plane that can make all of this processing power accessible and usable to an extensive range of scientists across disciplines.”

(0)(0)