Fugaku Tops Supercomputer Rankings Again as Arm HPC Ecosystem Expands
By: Brent Gorda, senior director of HPC, Arm
- Arm technology-based Fugaku supercomputer, jointly developed by RIKEN and Fujitsu, was again awarded the number one spot on the Top500 list
- Achievement underscores the rapidly evolving demands of HPC that Arm technology uniquely addresses
- The Arm HPC ecosystem continues to expand with key players announcing plans to leverage the next-generation Neoverse V1 platform and technologies, including SVE
Fugaku—the Arm technology-based supercomputer jointly developed by RIKEN and Fujitsu—was awarded the number one spot on the Top500 list for the second time in a row. This achievement further highlights the rapidly evolving demands of high-performance computing (HPC) that Arm technology uniquely addresses through the unmatched combination of power efficiency, performance, and scalability.
In addition to the great work RIKEN and Fujitsu have done, we’re seeing more adoption for Arm-based solutions across our ecosystem. ETRI, the national computing institute of the Republic of Korea, recently announced plans to adopt the upcoming Neoverse V1 (formerly code-named Zeus) CPU design, which feature Arm Scalable Vector Extensions (SVE), for its K-AB21 system. ETRI has set a goal of 16 teraflops per CPU and 1600 teraflops per rack for AB 21 (which stands for ‘Artificial Brain 21’) while reducing power consumption by 60% compared to its target.
Meanwhile, Ampere Computing, Sandia National Laboratories, University of Bristol, and Amazon Web Services (AWS) provided details on their Arm-based HPC plans at the Arm High-Performance Computing User Group (AHUG) meeting earlier this month. At this month’s SC20 event, SiPearl, a company formed by the European Processor Initiative, also reiterated that it will adopt Neoverse V1 for its Rhea processor, which is part of SiPearl’s exascale compute project. And, NVIDIA also announced researchers are achieving a nearly 26x performance boost for Arm in HPC with NVIDIA A100 GPUs and Arm Neoverse-based Ampere Altra CPUs.
While it’s still early days for seeing the impact of many of these new Arm-based solutions, we are proud of the positive impact Fugaku has already had in COVID-19 related research efforts. Although formally launched only a few months ago, Fugaku has been deployed for five different COVID-19 research projects, including one examining how the virus circulates in the air, and another to determine the efficacy of more than 2,000 existing drugs.
Continuing to the push the boundaries of what’s possible
This burst of activity is the culmination of design initiatives that have been underway for years and are ultimately aimed at solving one of HPC’s recurring challenges: how to build a system that can scale new heights of performance while remaining within reasonable power and budget envelopes. One of the central ideas behind the Arm processor architecture is that single-threaded, multicore processors will become the new norm for supercomputer CPUs as a way to increase performance without increasing relative power, and we’re seeing greater adoption of this shared vision across our expanding HPC ecosystem.
The hardware advances, of course, parallel the burgeoning activity in the software community to expand the options for Arm in HPC. Whether it’s the work we’re doing with NVIDIA to bring CUDA support to Arm-based processors or contributing to the open-source efforts to drive standards for HPC software packages, we are working hard to ensure we’re delivering the benefits of flexible, scalable, and efficient compute. SVE, meanwhile, continues to grow in importance as it enables processors that can accommodate vectors spanning 128 bits in length to 2048 bits (in 128-bit increments). Now, software developers can write once and have their work run on multiple processors. Likewise, processor developers and HPC users can take advantage of a wider spectrum of applications.
This year has been a watershed year for Arm in HPC. From our humble beginnings in server SoC design nearly a decade ago to being awarded the top supercomputer in the world (twice!), Arm’s journey has been significant, and we aren’t planning to slow down any time soon. Across the globe we’re seeing continued investment in the Arm ecosystem and increasing demand for more choice, more customization, and more innovation. It’s a challenge we are a ready to take on and we look forward to the next ten years of enabling our partners with technologies that are truly changing the world.
Arm technology is at the heart of a computing and data revolution that is transforming the way people live and businesses operate. Our energy-efficient processor designs and software platforms have enabled advanced computing in more than 190 billion chips and our technologies securely power products from the sensor to the smartphone and the supercomputer. Together with 1,000+ technology partners we are at the forefront of designing, securing and managing all areas of AI-enhanced connected compute from the chip to the cloud.
All information is provided "as is" and without warranty or representation. This document may be shared freely, attributed and unmodified. Arm is a registered trademark of Arm Limited (or its subsidiaries). All brands or product names are the property of their respective holders. © 1995-2021 Arm Group.