Coding with AI

NVIDIA Has Expanded Slurm to Integrate Quantum Computing Into Supercomputing Environments

Slurm 26.11 adds quantum processor support for supercomputing centers, enabling hybrid jobs via familiar workflows and NVIDIA's platform.

NVIDIA Has Expanded Slurm to Integrate Quantum Computing Into Supercomputing Environments

Key Takeaways

  • Slurm 26.11 adds quantum processor support, enabling hybrid quantum-classical workloads at nearly 30 supercomputing centers.
  • Users can keep existing Slurm job submission and resource management workflows instead of learning a new scheduler.
  • The update uses common interfaces like QRMI and QDMI, simplifying integration for quantum hardware vendors and supercomputing operators.
  • NVIDIA CUDA-Q expands with new tools for quantum error correction, fault-tolerant primitives, algorithms, pulse control, and ecosystem integrations.
  • Large-scale GPU simulations demonstrate how CUDA-Q can support chemistry and quantum design testing before full quantum hardware deployment.

Meta description: Slurm 26.11 adds quantum processor support for nearly 30 supercomputing centers, letting scientists run hybrid quantum-classical workloads with familiar tools.

NVIDIA Has Expanded Slurm to Integrate Quantum Computing Into Supercomputing Environments

Tags: Slurm 26.11, quantum processors, supercomputing centers, NVIDIA CUDA-Q, hybrid quantum-classical computing

Featured image: A friendly abstract illustration of a supercomputer scheduler connecting classical processors, GPUs and quantum processors as simple shapes; no real people, no logos, no text.

Slurm 26.11 is expanding its support for quantum processors at supercomputing centers, giving research teams a more familiar way to mix classical scientific workloads with quantum hardware. The change arrives at a time when quantum processor installations at supercomputing centers have risen by almost seven times in the last five years, with nearly 30 centers worldwide now moving toward deployments involving more than 50 quantum processors.

A familiar path into quantum workloads

Many of these centers already rely on Slurm to manage scientific supercomputing jobs, so the update is designed to avoid forcing users to learn a completely new scheduling system. The new release draws on earlier collaboration with IBM, Pasqal and the wider quantum computing community. It gives quantum processor makers a way to expose their hardware through one simple file that can serve as the same interface across supercomputing centers.

The expansion is already being adopted by Alice & Bob, IBM, IQM, Pasqal, Quantinuum and MQSC. It also supports the widely used QRMI and QDMI interfaces. For researchers in computational chemistry, nuclear physics and materials science, the result is a lower-friction route into quantum computing: they can keep working inside established job submission, queueing and resource-management patterns while adding quantum hardware to the mix.

NVIDIA Has Expanded Slurm to Integrate Quantum Computing Into Supercomputing Environments

Why the bottleneck mattered

Before this change, adding a quantum processor to a supercomputing center often meant a new software challenge for every new quantum processor and every new center. Each hardware maker and each operator had to solve the same connection problem again. The update leans on established quantum-supercomputing connections such as QRMI and QDMI, letting quantum processor builders reuse their integrations across deployments.

The benefit is symmetrical. Supercomputing centers can add new quantum processors to their existing scheduling and resource-management workflows without developing bespoke quantum-processor-specific solutions. Hardware builders get one simpler way to serve many centers. Scientists get a practical path to run quantum-classical applications developed with platforms such as NVIDIA CUDA-Q, with Slurm coordinating access to the processing they need.

Slurm 26.11 will be available later this month.

The wider NVIDIA CUDA-Q platform at SC26

The Slurm expansion also adds a community-developed orchestration layer to NVIDIA’s open platform for quantum-GPU supercomputing. That platform centers on CUDA-Q, an open platform for hybrid quantum-classical computing. At SC26, new releases expand several parts of the stack.

CUDA-Q QEC supports researchers developing quantum error correction, allowing them to rapidly prototype and test new ideas, including through a new decoder server. A new TCBP-JOBS decoder is 20,000 times faster than the industry standard.

NVIDIA Has Expanded Slurm to Integrate Quantum Computing Into Supercomputing Environments

CUDA-Q also gains an expanded layer of fault-tolerant primitives that application builders can use to compile ideas more realistically. Fermilab has already used CUDA-Q to perform studies in three weeks that would normally have taken five months.

CUDA-Q Algorithms provides reusable components for builders of quantum applications. New capabilities help design the loading of data into quantum circuits, and tools are included for chemistry and materials applications. CUDA-Q Pulse gives quantum hardware builders a layer for shaping the pulses needed to control qubits on a quantum processor, including tools that can use GPUs to accelerate their generation.

Ecosystem links and supercomputing-scale tests

The quantum ecosystem is continuing to build around this platform. Quantum Motion integrated its QUICHE chemistry framework with CUDA-Q. Altera and Oxford Quantum Circuits designed a controller based on field-programmable gate arrays, connecting qubit control and readout to GPUs through NVQLink.

At supercomputing scale, QunaSys used cuQuantum and 1,024 NVIDIA Hopper GPUs on ABCI-Q to simulate more than 300 million quantum operations, reaching chemical accuracy for a four-hydrogen-atom model. NVIDIA researchers used JUPITER to simulate 104 qubits under realistic noise, testing fault-tolerant designs before they are built.

The next step for many research centers will be simple to measure: whether Slurm 26.11 makes it easier to put quantum processors into the same queues as classical scientific workloads. If it does, the value will be less about a new tool and more about the many existing Slurm users who can start experimenting with quantum-classical computing without rebuilding their workflow from scratch.

Comments (0)

  1. No comments yet. Be the first to share what worked for you.

Leave a comment

Comments are reviewed before they appear. Your email address is not published.