NVIDIA, Rolls-Royce and Classiq Announce Quantum Computing Breakthrough in Jet Engine Fluid Dynamics

NVIDIA, Rolls-Royce and Classiq Announce Quantum Computing Breakthrough in Jet Engine Computational Fluid Dynamics

The world's largest industrial analog quantum circuit will promote the development of quantum computing in the aviation field

HAMBURG, Germany – International Supercomputing Conference (ISC) – May 21, 2023 – NVIDIA, Rolls-Royce and quantum software company Classiq today announced a quantum computing breakthrough that aims to continually improve jet engine efficiency.

Using NVIDIA's quantum computing platform, the two companies designed and simulated the world's largest computational fluid dynamics (CFD) quantum computing circuit. The circuit measures 10 million layers deep and has 39 qubits. Rolls-Royce is using GPUs to prepare for the quantum future, even though today's quantum computers can only support circuits with depths of only a few layers.

Rolls-Royce plans to use the new circuit to exploit the advantages of quantum in CFD, using both classical and quantum computing methods to simulate the performance of jet engine designs.

The breakthrough is vital for Rolls-Royce, a global leader in aviation, as it builds state-of-the-art jet engines and drives the energy transition towards more sustainable aviation.

“Jet engines are among the most complex devices on Earth, extremely expensive to design and computationally challenging to perform,” said Ian Buck, vice president of hyperscale and high-performance computing at NVIDIA . “NVIDIA’s quantum computing platform not only provides Rolls-Royce with a potential path to solving these problems, it also accelerates the company’s research and development of more efficient jet engines in the future.”

Leigh Lapworth, a computational science researcher at Rolls-Royce, said: “Applying classical and quantum computing methods directly to solving jet engine design challenges will help us speed up our processes and perform more complex calculations.”

Rolls-Royce and its partner, Israeli company Classiq, designed the circuit using Classiq’s synthesis engine and then simulated it using NVIDIA® A100 Tensor Core GPUs, with speed and scale enabled by NVIDIA cuQuantum, a software development tool suite of optimized libraries and tools to accelerate quantum computing processes.

NVIDIA Grace Hopper Accelerates Quantum Computing

NVIDIA provides a unified computing platform to accelerate quantum research and development breakthroughs across disciplines. The Grace Hopper superchip combines the incredible performance of NVIDIA Hopper™ architecture GPUs with the versatility of NVIDIA Grace CPUs, making it ideal for ultra-large-scale quantum simulation workloads.

In addition, its high-speed, low-latency NVIDIA NVLink®-C2C interconnect technology perfectly optimizes the connection between classical systems built using this superchip and quantum processors, or QPUs. Grace Hopper has a total of 600GB of fast-access memory per node, enabling the quantum ecosystem to further scale these simulations.

As a "strategic bridge" to future quantum computing, Grace Hopper drives the world's first GPU-accelerated quantum computing system, DGX™ Quantum, which combines quantum computing with the most advanced classical computing. NVIDIA also provides developers with a powerful open source programming model that connects GPUs and QPUs - NVIDIA CUDA® Quantum.

NVIDIA’s Quantum Ecosystem Continues to Expand

Today, much of the world’s quantum computing research runs on NVIDIA GPUs.

The Jülich Supercomputing Center, one of Europe's largest quantum computing facilities, also announced at ISC that it plans to build a quantum computing lab with NVIDIA . This highlights the growing importance of quantum-classical hybrid computing systems. The lab will also use tools such as CUDA Quantum to help developers advance the field of quantum computing.

In addition, ORCA Computing, the manufacturer of the latest QPUs that integrate CUDA Quantum, is combining its photonic quantum computers with GPUs for machine learning. Two popular quantum machine learning frameworks, TensorFlow Quantum and TorchQuantum, now also integrate cuQuantum. Today, most quantum computing software in the world supports the GPU acceleration capabilities of the NVIDIA quantum platform.

Learn more about the NVIDIA quantum computing platform at ISC.