Nvidia is partnering with Google Quantum AI to improve the design of next-generation quantum computing devices through advanced simulations.
Using the hybrid quantum-classical computing platform and the Nvidia Eos supercomputer, Google is simulating the physics of its quantum processors. This initiative aims to address the limitations of current quantum computing hardware, which struggles with noise interference during operations.
Guifre Vidal, a research scientist at Google Quantum AI, stated, "The development of commercially useful quantum computers is only possible if we can scale up quantum hardware while keeping noise in check." The collaboration focuses on understanding noise implications in larger quantum chip designs.
Traditionally, the complex simulations needed to analyze noise have been computationally expensive. However, with the CUDA-Q platform, Google can utilize 1,024 Nvidia H100 Tensor Core GPUs at the Eos supercomputer to conduct rapid and cost-effective simulations.
Tim Costa, director of quantum and HPC at Nvidia, emphasized the importance of AI supercomputing in the advancement of quantum computing, noting that Google's use of the CUDA-Q platform exemplifies the role of GPU-accelerated simulations in solving real-world problems.
The collaboration has achieved significant milestones, enabling comprehensive simulations of devices with 40 qubits—the largest of its kind. Techniques from CUDA-Q allow simulations that once took a week to complete to now be executed in mere minutes.
The software enabling these accelerated simulations will be made publicly available on the CUDA-Q platform, facilitating faster scaling of quantum hardware designs for engineers.