The NVIDIA cuQuantum Appliance is a highly performant multi-GPU, multi-node solution for quantum circuit simulation. It contains NVIDIA’s cuStateVec, cuTensorNet, and cuDensityMat libraries, which optimize state vector, tensor network, and analog quantum dynamics simulation, respectively. The cuTensorNet library functionality is accessible through Python for Tensor Network operations. With the cuStateVec libraries, NVIDIA provides the following simulators: IBM’s Qiskit Aer frontend via cusvaer, NVIDIA’s distributed state vector backend solver, and an optimized multi-GPU Google Cirq frontend via qsim, Google’s state vector simulator.

Features

• cuStateVec Library: Optimizes state vector simulation.
• cuTensorNet Library: Optimizes tensor network simulation and is accessible through Python.
• cuDensityMat Library: Optimizes analog quantum dynamics simulation.
• Multi-GPU and Multi-Node Support: Enables high-performance simulations across multiple GPUs and nodes.
• Integration with Qiskit and Cirq: Provides optimized backends for IBM’s Qiskit Aer and Google’s Cirq.
• Docker Integration: Simplifies deployment with Docker containers.

Benefits

• High Performance: Leverages multiple GPUs and nodes for efficient quantum circuit simulation.
• Versatility: Supports various quantum simulation libraries and frameworks.
• Ease of Use: Simplifies deployment and usage with Docker containers and Python integration.
• Scalability: Suitable for large-scale quantum simulations with multi-GPU and multi-node support.