SMART WORLDJOURNALSubmit Manuscript
← All technology news

Computing

NVIDIA Enhances Slurm to Integrate Quantum Processors with Supercomputing Workflows

NVIDIA expands the Slurm workload manager to support quantum processors across supercomputing centers, simplifying integration and enabling scientists to access quantum and classical computing resources seamlessly within familiar HPC environments.

NEWS

Quantum processor deployments at supercomputing centers have increased nearly sevenfold over the past five years, with around 30 centers globally deploying over 50 quantum processors. Many utilize Slurm, a popular workload manager for scientific computing, to handle their supercomputing tasks. NVIDIA’s recent update to Slurm extends its capabilities to include quantum processing units (QPUs), fostering easier incorporation of quantum computing into established high-performance computing (HPC) workflows.

The latest Slurm release enables QPU manufacturers to implement a single, standardized interface for their hardware across multiple supercomputing centers using a simple configuration file. This design builds on previous collaborations with IBM, Pasqal, and the broader quantum community and supports existing quantum-HPC interfaces like QRMI and QDMI. As a result, centers can integrate multiple quantum processors without developing custom solutions, significantly reducing software overhead.

This integration addresses a major bottleneck in combining quantum and classical computing resources. Previously, each new QPU or center required distinct software adaptations, complicating resource management. NVIDIA’s Slurm enhancement allows centers to schedule and manage quantum and classical resources cohesively, facilitating hybrid quantum-classical applications such as those developed with NVIDIA CUDA-Q platforms.

Alongside Slurm updates, NVIDIA announced advancements in its CUDA-Q open platform, which supports hybrid quantum-classical computing. Enhancements include a rapid quantum error correction (QEC) decoder server, accelerating testing of fault-tolerant designs by up to 13,000 times. CUDA-Q also offers expanded fault-tolerant primitives and reusable algorithmic components focused on chemistry and materials science applications, accelerating scientific research timelines.

Moreover, CUDA-Q Pulse provides tools for hardware makers to design and GPU-accelerate control pulses necessary for qubit manipulation, integrating hardware control with software innovation. Various partners including Alice & Bob, IBM, IQM, Pasqal, Quantinuum, and MQSC have already begun adopting these integrations, demonstrating broad ecosystem engagement.

Notable demonstrations of the platform's capabilities include QunaSys using over a thousand NVIDIA Hopper GPUs on ABCI-Q to simulate quantum operations at chemical accuracy, and NVIDIA researchers employing the JUPITER simulator to model 104-qubit systems under realistic noise. These efforts highlight the practical impact of integrating quantum computing into supercomputing workflows.

While the expanded Slurm support and CUDA-Q platform mark significant steps toward hybrid quantum-classical supercomputing, challenges remain. The technology depends on continued hardware maturation, error correction advances, and ecosystem adoption. Nonetheless, this development streamlines access for scientists, potentially accelerating quantum advantage in research domains such as chemistry and materials science.

Source & further reading

NVIDIA Expands Slurm to Bring Quantum Computing Into Supercomputing ↗

Source published 2026-11-16. Claims and limitations should be checked against the original announcement or research.