Nvidia on How Supercomputing Is Helping Drive Quantum

Key Concepts

• NVIDIA's role in quantum computing: Providing architecture, software, and services, not building quantum computers.
• Integration of quantum processors into data centers: Quantum technology as an additional element alongside CPUs, GPUs, storage, and memory.
• Accelerated computing for quantum development: Using GPUs to simulate quantum devices, develop error correction methods, and calibrate devices.
• AI for quantum: Using AI models to control and improve quantum devices.
• Quantum for AI: Using quantum computers to generate data for training and fine-tuning AI models, particularly for understanding quantum physics.
• Quantum-accelerated supercomputers: Integrating quantum devices with GPUs to create new computing architectures.
• Focus on specific problems: Narrowing the focus to define success in quantum computing.
• Potential applications in chemistry and biochemistry: Using quantum computers to model quantum physics for accurate chemistry simulations.

NVIDIA's Role in Quantum Computing

NVIDIA does not build or sell quantum computers. Instead, NVIDIA focuses on providing the necessary architecture, software, and services to support the quantum computing industry. Their vision involves integrating quantum processors into existing data centers and large-scale computing infrastructure, working alongside CPUs, GPUs, storage, and memory. NVIDIA aims to be a partner to quantum device developers, providing tools for error correction, device calibration, and overall system management.

Accelerated Computing for Quantum Development

NVIDIA's GPUs and accelerated computing technologies, including software stacks, are used by quantum computing companies in their research programs. This includes simulating quantum devices, building better versions of these devices, working with clients, and designing algorithms. NVIDIA's technology also supports fundamental research aimed at integrating quantum devices with traditional computing resources. This integration involves developing interconnects and novel AI methods for error correction, deployed at scale on supercomputers. NVIDIA is working with over 160 groups in quantum computing, accelerating their work across a wide range of application areas.

AI and Quantum Computing: A Two-Way Street

There are two key aspects to the relationship between AI and quantum computing:

1. AI for Quantum: AI models are used to control and improve quantum devices. This is crucial for scaling and enhancing the capabilities of quantum computers, accelerating their development timeline.
2. Quantum for AI: Quantum computers can generate data to train and fine-tune AI models, particularly for understanding the quantum physics phenomena within the quantum computer itself. This could potentially answer questions that humanity has been unable to address.

Quantum-Accelerated Supercomputers and the Boston Research Center

NVIDIA is actively working towards building quantum-accelerated supercomputers, which involve integrating quantum devices with GPUs. A new research center in Boston will serve as a hub for NVIDIA and its partners to collaborate on developing interconnects, error correction technologies, and the first versions of these quantum-accelerated supercomputers. This center will allow partners to physically connect their quantum devices to NVIDIA's GPUs and develop the necessary infrastructure for integrated quantum-classical computing.

Defining "Useful" Quantum Computing and Potential Applications

Jensen Huang emphasized the importance of narrowing the focus and defining specific problems to achieve success in quantum computing. There is a general consensus that chemistry and biochemistry are promising areas for early disruption. The ability of quantum devices to model quantum physics makes them well-suited for accurate chemistry simulations. While opinions vary within the quantum community, this area is considered a strong candidate for demonstrating the value of quantum computing.

Synthesis/Conclusion

NVIDIA's strategy in quantum computing is to enable and accelerate the field through its existing strengths in GPUs, software, and system architecture. By focusing on integration with classical computing resources, developing tools for error correction and device management, and exploring the interplay between AI and quantum, NVIDIA aims to play a crucial role in realizing the potential of quantum computing for solving complex problems in areas like chemistry and beyond. The establishment of the research center in Boston signifies a concrete step towards building the infrastructure needed for quantum-accelerated supercomputers.