Rathaus in Hamburg

AHUG ISC 22 Workshop and Birds of a Feather

Please continue to check this page for the latest information on the AHUG ISC 2022 non-proceedings workshop and Birds of a Feather at this page and on our Twitter, Slack and mailing lists. We anticipate both of these events to be held in-person as ISC22 travel restrictions allow. 

ISC 2022 AHUG Workshop

Date: June 2nd, 2022. 9 AM-6 PM CEST (GMT+2)

Location: Hall Y5

Please note that this event will be held in person! Please see the ISC agenda for more information on this event.

AHUG ISC 2022 Workshop Agenda

Time (CEST)

Description

Presenter / Moderator

Affiliation

09:00-

9:10

Welcome remarks Jeffrey Young AHUG, Georgia Tech
Arm HPC Applications and Initiatives Roxana Rusitoru Arm

09:10-

9:40

Keynote: Life With and After Fugaku Satoshi Matsuoka RIKEN CCS
09:40-10:10 AWS Graviton3: The first cloud native SVE-enabled Arm-based processor Olly Perks AWS
10:10-10:40 How Arm platforms can be efficiently used to execute lifescience workflows Emma Fine Atos
10:40 – 11:00 UK Catalyst Programme Review Ben Bennett HPE
11:00-11:30 Break
Student Lightning Talks Jeffrey Young
11:30 – 11:50 Top-down model of a A64FX to study data layout optimization of a CDF code Fabio Banchelli

Barcelona Supercomputing

Center

11:50 – 12:10 Atomics in Arm: Are they ruining your performance? Ricardo Jesus Edinburgh Parallel Computing Centre
12:10 – 12:30 Deploying an HPC Software Stack with EasyBuild on the ARM-based HAICGU cluster Stepan Nassyr Jülich Supercomputing Centre
12:30 – 12:50 GenarchBench: Porting a Genomics Suite to A64FX Lorién Lapez Villellas Barcelona Supercomputing Center
13:00 – 14:00 Lunch
Arm HPC – Government, Academic, and Industry Collaborations Michèle Weiland
14:00 – 14:30 Co-design exploration with vectorized CNNs Miquel Pericas Chalmers University of Technology
14:30 – 15:00 Tensor Processing Primitives on Arm Processors Alex Breuer Friedrich Schiller University Jena
15:00 – 15:30 SiPearl Updates And Developments Craig Prunty SiPearl
15:30 – 16:00 Commercial vs Open ISAs: a High Performance Data Analytics Perspective Antonino Tumeo Arm and RISC-V Evaluations
16:00 – 16:30 Break
Upcoming Silicon Efforts, Extensions for Arm Scientific Computing Jeffrey Young
16:30 – 17:00 Accelerating HPC & AI with NVIDIA Grace and BlueField-3 Han Vanholder NVIDIA
17:00-17:30 Performance analysis of a quantum simulator on A64FX processor (Virtual Talk) Miwako Tsuji RIKEN CCS
17:30-18:00 Open Discussion, Closing Remarks
18:00 Workshop End

 

ISC 2022 Birds of Feather

Accelerating Arm for Zettascale

Date: May 30, 2022. 1 PM to 2 PM CEST (GMT+2)

Location: Hall E

Please note that this event will be held in person! Please see the ISC agenda for more information on this event.

This Birds of Feather session will allow for attendees to discuss topics they feel are relevant for growing Arm HPC around the world. The discussion will be led by a moderator, and we will address topics related to the growth of Arm HPC with an eye focused on (the possibly far off) requirements for future Arm “Zettascale” platforms.

Birds of a Feather Expert Panelists: 

  • Brendan Bouffler, AWS
  • John Linford, NVIDIA
  • Satoshi Matsuoka, RIKEN-CCS 
  • Roxana Rusitoru, Arm
  • Michèle Weiland, EPCC, AHUG 

Moderators/Organizers:

  • Sato Mitsuhisa, RIKEN-CCS, AHUG
  • Steve Poole, LANL, AHUG
  • Jeffrey Young, Georgia Tech, AHUG

BoF Panelist Bios

Brendan Bouffler, Head of Developer Relations, HPC Engineering, AWS
Brendan Bouffler has over 25 years of experience in the global tech industry creating very large systems in high performance environments. He has been responsible for designing and building hundreds of HPC systems for commercial enterprises as well as research and defence sectors all around the world and has quite a number of his efforts listed in the Top 500, including some that have placed in the top 5. Brendan previously lead the HPC Organization in Asia for a hardware maker but joined Amazon in 2014 to accelerate the adoption of cloud computing in the scientific community globally.
John Linford, Principal Technical Product Manager for Datacenter CPU Software, NVIDIA
Dr. John Linford enjoys weird and wonderful computing systems and the weird and wonderful people who build and use them. The last four years at Arm gave John frontline experience with every Arm-based HPC-class CPU you’ve ever heard of, as well as some you likely haven’t. Before Arm, John spent six years at ParaTools porting and tuning extreme-scale applications. John has recently joined NVIDIA to help with the Grace Hopper project.
Satoshi Matsuoka, Director of RIKEN-CCS
Dr. Satoshi Matsuoka is the current head of the Riken Center for Computational Science (R-CCS) at RIKEN, the largest Supercomputing center in Japan. Dr. Matsuoka’s research is principally in system software for large scale supercomputers and similar infrastructures such as Clouds for HPC, and more recently, convergence of Big Data / AI with HPC, as well as investigating the Post-Moore Technologies towards 2025. Over the years he has been involved and lead a number of large collaborative projects that worked on basic elements that are now significant for the current and more importantly future exascale systems, such as fault tolerance, low power, strong scalability, programmability, as well as large-scale I/O.
Roxana Rusitoru, Datacenter System Architect, Arm

Roxana Rusitoru is the Datacenter and HPC Systems Architect within Arm’s Infrastructure Line of Business. She joined Arm in 2012 and, as part of the effort to bring Arm into HPC, she has worked on most software stacks, from Linux kernel optimizations aimed at HPC up to application analysis and optimization for Arm’s Scalable Vector Extension (SVE), HPC power management and deep learning training for SVE. From a hardware perspective, she contributed to next-generation system architecture and memory systems. Currently, Roxana is covering all  technology areas for next-generation datacenters and HPC and is helping to drive requirements and technology strategies.

Michèle Weiland, Senior Research Fellow, EPCC, AHUG Associate Director

Dr. Michèle Weiland is a Senior Research Fellow  t EPCC and an associate director for the Arm HPC User Group (AHUG). She specializes in novel technologies for extreme scale parallel computing, leading EPCC’s technical involvements in the ASiMoV Strategic Prosperity Partnership with Rolls-Royce and the NEXTGenIO project, which focuses on innovations for I/O at the Exascale. She also leads EPCC’s involvement in the Catalyst UK rogram, a partnership with HPE and Arm to accelerate to adoption of the Arm ecosystem. She is the EPCC PI on a number of research grants, including the EC Horizon 2020 projects ExaFLOW, HPC-WE and SAGE2, as well as Co-PI on the ExCALIBUR project ELEMENT and for the Cirrus2 Tier-2 service. She is a member of the EPSRC Strategic Advisory Team for e-Infrastructure.