CUDA-Q Logical adds an open orchestration layer for designing, testing and optimising fault-tolerant quantum computing systems.
NVIDIA has expanded its open-source CUDA-Q platform with CUDA-Q Logical, an orchestration layer designed to help researchers develop applications for fault-tolerant quantum computers. The addition provides a programmable way to represent and test the different components of a fault-tolerant system, allowing researchers to explore combinations of quantum hardware, error-correction methods and system resources.
CUDA-Q Logical is intended to address the complexity of designing systems around logical qubits, where quantum algorithms, error correction, hardware architecture and classical computing resources need to be considered together. Researchers can switch between different implementation choices and evaluate their impact on performance and resource requirements, rather than building separate specialised infrastructure for each configuration.
Fermilab has used the platform to evaluate fault-tolerant architectures and compare physical-qubit requirements, runtimes and error-correction strategies. According to NVIDIA, this reduced a development process that would normally take about five months to three weeks, providing a sevenfold reduction in the time required to explore the configurations.
The expanded platform also incorporates QUOPS, a benchmark developed by Sandia National Laboratories for evaluating progress towards fault-tolerant quantum computing. Rather than focusing only on physical-qubit counts, the benchmark is intended to provide a broader way of assessing the capabilities of quantum hardware for useful applications. Initial QUOPS measurements have been performed on quantum processors from Google, IBM and Quantinuum.
CUDA-Q remains an open, QPU-agnostic platform for combining quantum processors with accelerated classical computing. NVIDIA says CUDA-Q Logical is available through GitHub, while the expanded ecosystem is being used by quantum computing companies and research organisations to explore applications including drug discovery, financial modelling and materials development.
For more information, click here.
