Don’t Stop Me Now: Accelerating AI, HPC at the Speed of Light at ISC

by Emily Tobar Arango
HPC & AI Simulations

AI and high performance computing are tackling the world’s greatest challenges.

Medical imaging, climate science, genomic and proteomic sequencing and many other complex scientific areas have seen groundbreaking discoveries thanks to the development and application of HPC and AI technologies.

Next week, thousands of people will be tuning in to ISC High Performance 2021 Digital, the annual international supercomputing conference, running from June 28-July 2, for talks, demos and workshops to discover the latest breakthroughs in HPC and AI.

Hear from NVIDIA Experts at ISC

Nearly 70 percent of the TOP500 list of the world’s fastest supercomputers are accelerated by NVIDIA GPUs and networking. That’s why you’ll see NVIDIA technology and experts featured in a number of sessions and workshops throughout ISC.

Here’s what not to miss:

Discover Cutting-Edge Technology in Action

Throughout the week of ISC, industry leaders, scientists and researchers will present a variety of talks and demos that explain their groundbreaking work using GPU and HPC technologies.

In the ISC virtual theater, deep dive into the latest HPC software advancements through talks covering all things AI and HPC. Explore the possibilities of NVIDIA Omniverse for rich, impactful scientific visualizations. Discover how using GPU-accelerated adaptive sampling can reduce simulation time — from years to days — to deliver high-resolution modeling of proteomes like SARS-CoV2. Go deep with graph neural networks to learn how graph learning can enable new research opportunities.

In a series of demos, explore how NVIDIA technologies are transforming industries with new capabilities. Discover how NVIDIA GPUs are being used to view, measure and track plant growth, how galactic winds are being visualised interactively in Omniverse and how the NVIDIA UFM Cyber-AI platform is transforming data center management and security.

Watch Live: Accelerating with DPUs

HPC and AI communication frameworks and libraries are bandwidth-sensitive, and they play a critical role in determining application performance. Using the NVIDIA BlueField data processing unit (DPU) to offload libraries reduces operating system jitter, while dramatically increasing application performance, which is key to enabling a cloud-native supercomputing architecture.

Register for the live demo to interact with library developers and learn how to accelerate scientific application with DPUs.

Roll Up Your Sleeves: It’s Time to Get Hands-On

Developers, data scientists, researchers and students can join self-paced online courses and instructor-led workshops through the NVIDIA Deep Learning Institute. Get access to a fully configured, GPU-accelerated server in the cloud, gain practical skills for your work, and earn a certificate of subject-matter competency.

Students will be battling it out in the Virtual Student Cluster Competition, which lets teams show off their HPC knowledge.

A total of 13 teams will have the chance to showcase systems of their own design and compete to achieve the highest performance across a series of standard HPC benchmarks and applications.

The cluster competition award ceremony begins on July 1 at 5 p.m. CET | 7 a.m. PT.

Keep up to date on all things HPC and AI by heading to our ISC event page and following our social handles @NVIDIAEU and #ISC21.