NVIDIA has recently disclosed a groundbreaking system developed in collaboration with Quantum Machines, designed to provide an innovative architecture for researchers engaged in high-performance and low-latency quantum-classical computing. Known as the NVIDIA DGX Quantum, this system is the world’s first GPU-accelerated quantum computing solution, integrating the NVIDIA Grace Hopper Superchip and CUDA Quantum open-source programming model with Quantum Machines’ advanced quantum control platform, OPX.
Combining these technologies results in the world’s most powerful accelerated computing platform, capable of tackling complex computational challenges at an unprecedented speed. The NVIDIA DGX Quantum represents a significant breakthrough in quantum computing, promising to revolutionize various industries through its ability to harness the power of quantum and classical computing in a single, unified system.
“Quantum-accelerated supercomputing has the potential to reshape science and industry with capabilities that can serve humanity in enormous ways. NVIDIA DGX Quantum will enable researchers to push the boundaries of quantum-classical computing.”
Tim Costa, director of HPC and quantum at NVIDIA.
At the core of the DGX Quantum is an NVIDIA Grace Hopper platform linked to Quantum Machines OPX+ through PCIe, creating a seamless connection between GPUs and quantum processing units (QPUs) with sub-microsecond latency. This integration enables fast and efficient communication between the components, facilitating the execution of complex computational tasks at an unprecedented pace.
The DGX Quantum’s innovative design promises to break new ground in quantum-classical computing, enabling researchers to explore new frontiers in various fields, including finance, healthcare, and energy.
“We are heading toward a new age of quantum computing that is more accessible to more researchers than ever,” said Itamar Sivan, co-founder and CEO of Quantum Machines. “Our collaboration with NVIDIA on the DGX Quantum system will enable a new generation of innovators to solve some of the world’s greatest challenges.”
NVIDIA’s Large-scale Artificial Intelligence (AI) and High-Performance Computing (HPC) Applications
The Grace Hopper system combines the powerful NVIDIA Hopper architecture GPU with its latest Grace CPU. It is an ideal candidate for large-scale artificial intelligence (AI) and high-performance computing (HPC) applications. With the capability of delivering up to 10 times better performance for applications operating on terabytes of data, the Grace Hopper system provides quantum-classical researchers with unprecedented computing power to solve the most complex problems in the world.
On the other hand, OPX+ is a versatile quantum control system that incorporates classical computing engines into the core of the quantum control framework, allowing the optimal performance of any quantum processing unit (QPU) in real-time.
This integration opens up new possibilities for quantum algorithms. The Grace Hopper and OPX+ systems can be easily scaled to suit the system’s size, whether a few-qubit QPU or a quantum-accelerated supercomputer. The combination of these cutting-edge technologies promises to revolutionize the field of quantum computing, enabling researchers to tackle some of the most challenging computational problems in various domains.
NVIDIA’s CUDA Quantum Enables Seamless Integration of Quantum and Classical Computing for Hybrid Computing
In addition to the powerful hardware systems, DGX Quantum also provides developers with access to NVIDIA CUDA Quantum, which is an integrated software stack now available in open source. Quantum is a platform that combines quantum and classical computing, allowing for the programming and integration of QPUs, GPUs, and CPUs in a single system.
To expand the reach of CUDA Quantum, NVIDIA has partnered with various companies and institutions such as Anyon Systems, Atom Computing, IonQ, ORCA Computing, Oxford Quantum Circuits, and QuEra, as well as Agnostiq and QMware, which are quantum software companies.
Supercomputing centers like the National Institute of Advanced Industrial Science and Technology, the IT Center for Science (CSC), and the National Center for Supercomputing Applications (NCSA) have also partnered with NVIDIA to integrate CUDA Quantum into their platforms. This expansion of partners will enable more researchers and developers to leverage the power of CUDA Quantum for their quantum-classical computing applications.
Read more about it here.