Nvidia announces two large substrategies

On May 21, Nvidia detailed three programs in which researchers will use its chips to support scientific discovery.

One of the plans is for the University of Bristol to deploy a supercomputer powered by Nvidia's Grace CPU superchip processor. The other two projects focus on quantum computing. The chipmaker detailed all three projects at this week's ISC 2023 conference in Hamburg.

In the United Kingdom, the University of Bristol is leading an initiative to build a new supercomputer optimized for research workloads: The supercomputer will be called Isambard 3, and it will be built by Hewlett Packard Enterprise using Nvidia chips.

According to the chipmaker, Isambard 3 will be equipped with 384 Grace CPU Superchips processors. Each such processor includes 144 central processing unit cores based on Arm Ltd. designs; these cores support SVE -- an Arm instruction set technology that allows CPUs to run artificial intelligence workloads more efficiently.

NVIDIA expects Isambard 3 to deliver a top speed of 2.7 petaflops (one petaflop equals 1 trillion computational operations per second) when performing calculations involving FP64 data, a data format commonly used by supercomputers to organize the information they process.

Scientists plan to use Isambard 3 for a range of tasks. According to NVIDIA, the system will support research programs in areas such as artificial intelligence, life sciences, medicine, biotechnology and astrophysics. It will be able to simulate large, complex structures such as wind turbines and molecular-level biological mechanisms.

Isambard 3 is scheduled to go live in 2024. Once operational, the system is expected to consume less than 270 kilowatts of power, which will make it one of the most energy-efficient supercomputers in Europe. NVIDIA also estimates that Isambard 3 is six times more energy efficient than the previous generation of supercomputers at the University of Bristol.

In response, Ian Buck, NVIDIA's vice president for exascale and high-performance computing, said, "As climate change becomes an increasingly present issue, it is critical that computing technologies embrace energy-efficient technologies. NVIDIA is working with the Arm Neoverse ecosystem to provide a path forward for creating more energy-efficient supercomputing centers, driving important breakthroughs in scientific and industrial research."

At ISC 2023, NVIDIA also detailed how its technology is helping to support two quantum computing initiatives in the UK and Germany.

The first initiative is being led by Rolls-Royce Holdings Ltd. in partnership with quantum computing startup Classiq Technologies Inc. The goal of the project is to advance Rolls-Royce's development of more efficient jet engines. The company hopes to use quantum computers to ease the work of its engineers.

Working with Classiq, Rolls-Royce used a set of Nvidia A100 graphics cards to develop and simulate a quantum circuit: a specialized algorithm that could be run on a quantum computer in the future. Rolls-Royce says its algorithm circuit has 10 million layers and 39 quantum bits of depth. By using GPUs, Rolls-Royce is preparing for a quantum future; despite the limitations of today's quantum computers, which support circuits only a few layers deep.

According to NVIDIA, the quantum circuit is designed to perform calculations related to computational fluid dynamics, a field of science focused on simulating complex physical phenomena. Rolls-Royce's quantum circuit is considered to be the largest of its kind developed specifically for computational fluid dynamics.

Of the collaboration, Buck said, "Jet engines are among the most complex devices on the planet, and their design is costly and computationally challenging. NVIDIA's quantum computing platform provides Rolls-Royce with a potential path to address these issues head-on while accelerating its research and future development of more efficient jet engines."

The second quantum computing initiative detailed by NVIDIA is a collaboration with the Jülich Supercomputer Center (FZJ), a German research institute and one of the largest quantum computing facilities in Europe. Ullrich is working with NVIDIA to establish a new lab that will have what both companies describe as a classical-quantum supercomputer; the system will include both classical and quantum computing elements.

NVIDIA details that the machine will be developed by Munich-based ParTec AG. And FZJ researchers will use NVIDIA's cuQuantum and CUDA Quantum software tools to simulate quantum processors on the system.

FZJ is one of the largest interdisciplinary research centers in Europe, and the lab will run high-performance, low-latency quantum-classical computing workloads as part of the Jülich UNified Infrastructure for Quantum Computing (JUNIQ), which is currently using the JUWELS booster system. FZJ is already running a classical supercomputer with 3744 NVIDIA A100 graphics cards.

Unifying quantum computing and GPU supercomputing is a key part of enabling future scientific breakthroughs," said Timothy Costa, NVIDIA's head of HPC and quantum. NVIDIA's collaboration with innovators such as the Ullrich Supercomputing Center and ParTec is an important milestone for quantum-classical computing, enabling countless new researchers to use it and bringing us one step closer to the first quantum-accelerated supercomputer."

About NVIDIA

Since its founding in 1993, NVIDIA (NASDAQ: NVDA) has been a pioneer in accelerated computing. The company's invention of the GPU in 1999 sparked the growth of the PC gaming market, redefined computer graphics, ignited the modern era of artificial intelligence, and is driving the creation of the industrial metaspace. NVIDIA is now a full-stack computing company with data center-scale products that are reshaping industry.

Reference links:

[1] https://siliconangle.com/2023/05/22/nvidia-chips-drive-new-supercomputing-quantum-initiatives/

[2]https://www.rolls-royce.com/

[3]https://nvidianews.nvidia.com/news/nvidia-rolls-royce-and-classiq-announce-quantum-computing-breakthrough-for-computational- fluid-dynamics-in-jet-engines

[4]https://nvidianews.nvidia.com/news/nvidia-and-julich-supercomputing-centre-to-build-quantum-computing-lab