Arm Community
Arm Community
  • Site
  • User
  • Site
  • Search
  • User
Arm Community blogs
Arm Community blogs
Servers and Cloud Computing blog ISC 2019 in review
  • Blogs
  • Mentions
  • Sub-Groups
  • Tags
  • Jump...
  • Cancel
More blogs in Arm Community blogs
  • AI blog

  • Announcements

  • Architectures and Processors blog

  • Automotive blog

  • Embedded and Microcontrollers blog

  • Internet of Things (IoT) blog

  • Laptops and Desktops blog

  • Mobile, Graphics, and Gaming blog

  • Operating Systems blog

  • Servers and Cloud Computing blog

  • SoC Design and Simulation blog

  • Tools, Software and IDEs blog

Tags
  • High Performance Computing (HPC)
  • Petascale
  • Neoverse
Actions
  • RSS
  • More
  • Cancel
Related blog posts
Related forum threads

ISC 2019 in review

Brent Gorda
Brent Gorda
July 15, 2019
3 minute read time.

The recent International Supercomputing Conference (ISC) was again held in Frankfurt, Germany mid-June.  As is typical, the event was well-attended and each day’s schedule was full. In my new role as head of the worldwide HPC business for Arm, I spent much of my time catching up with colleagues from Europe and around the world.  The excitement for Arm was obvious and the Arm team was welcomed with open arms and warm smiles.  As has been my experience, the patio at the Marriott was the place to be after the evening events had died down and the discussion trended towards solving the world's HPC problems via lively debate.  The impending arrival of Exascale is on everyone’s mind.

From the perspective of Arm in HPC, it was an excellent event with several high-profile announcements that caught everyone’s attention. The Arm ecosystem was well represented with our partners visible on the show floor and around town. 

Leading up to ISC, our friends at RIKEN and Fujitsu announced more details on the post-K system. The supercomputer is now officially called “Fugaku”, a nickname for the famous Mt. Fuji in Japan.  The name is an appropriate choice as the new computer is reported to be up to 100 times faster than the existing K computer (recently turned off to facilitate site modifications for the new system).  Lots of interest in seeing more benchmark results on this system.  The A64FX custom SoC offers per node memory bandwidth of 1 TB/s, and the first implementation of the Arm SVE IP with 2x512 vector engines.  It is indeed a system to keep an eye on.  

Next in news, the 53rd instance of the Top500 list was announced.  The good news for Arm is that the Sandia National Labs Astra system did quite well, moving up in ranks to #156.  The HPE system is the first major Arm-based computer to make the list.  It had been in the #203 spot as it was being installed at the time the previous list was published at SC18 and with a bit of further triage and tuning jumped ahead almost 50 places for ISC19.  Arm continues to collaborate with HPE both at Sandia as well as under the umbrella of the Catalyst UK project at three major universities (Edinburgh, Leicester, and Bristol) in the UK.

A major ecosystem step was also highlighted, on Monday of ISC19, as NVIDIA announced that they will bringing full CUDA stack support to Arm-based platforms for their market-leading GPU products.  The promise of their complete stack of AI and HPC software generated quite a buzz.  Several key sites I met with suggested this was causing them to rethink plans for upcoming systems as their pro-Arm opinions had been hindered by the lack of support for the current market-leading performance delivered by GPUs.  We look forward to working with NVIDIA in support of this effort.  

Taking a walk around the exhibition floor we saw vendors large and small touting Arm-based platforms, from Mellanox, HPE, Fujitsu and Gigabyte to storage centric partners such as Ambedded and Softiron.  Furthermore, we saw a strong HPC cloud presence at ISC, with mainstays such as Amazon AWS and Google holding court alongside up and coming players such as Verne Global.  Cloud based services are a key vertical for Arm and its partners and we look forward to an increasing presence for HPC in that realm.  With the launch of the Amazon AWS Graviton CPU and their new A1 instances, we now have an
immediate developer platform for porting codes and tire-kicking toolchains.   We also have bare-metal HPC cluster offerings from Verne Global with their HPCDirect now offering Marvell ThunderX2 platforms.

The public announcements are the visible tip of the iceberg with respect to activity across the Arm ecosystem.  Our meeting schedule was quite full, engaging with major SC centers on future deployments, syncing with ISV’s on porting efforts, and tying out with our hardware platform partners on their latest schedules and roadmaps.  Numerous teams (both internal and external) are collaborating and driving hard to deliver the next generation of technology that the HPC community is seeking.  As we see real production results and continued momentum, things are getting exciting.

I must add that the energy and enthusiasm on display by the Arm HPC team at ISC19 was infectious and noticeable - I really enjoyed being there as part of this team.

  

Please join us in our efforts to make Arm a preferred platform architecture of choice for supercomputing world-wide.  Find out more at the Arm HPC Ecosystem page.

Anonymous
Servers and Cloud Computing blog
  • How SiteMana scaled real-time visitor ingestion and ML inference by migrating to Arm-based AWS Graviton3

    Peter Ma
    Peter Ma
    Migrating to Arm-based AWS Graviton3 improved SiteMana’s scalability, latency, and costs while enabling real-time ML inference at scale.
    • July 4, 2025
  • Arm Performance Libraries 25.04 and Arm Toolchain for Linux 20.1 Release

    Chris Goodyer
    Chris Goodyer
    In this blog post, we announce the releases of Arm Performance Libraries 25.04 and Arm Toolchain for Linux 20.1. Explore the new product features, performance highlights and how to get started.
    • June 17, 2025
  • Harness the Power of Retrieval-Augmented Generation with Arm Neoverse-powered Google Axion Processors

    Na Li
    Na Li
    This blog explores the performance benefits of RAG and provides pointers for building a RAG application on Arm®︎ Neoverse-based Google Axion Processors for optimized AI workloads.
    • April 7, 2025