The new plugins model quantum processing units as native cluster-level resources for hybrid HPC workflows.
Slurm 26.11 adds Hierarchical Resources plugins that let a site schedule a quantum processing unit as a native cluster-level resource. A compact Lua plugin checks device health, admits jobs before Slurm allocates nodes, and releases resources. The plugin leaves site-specific policies in resources.yaml and keeps device logic portable across Slurm sites.
National programs in North America, Europe, and Asia are driving QPU installations alongside high-performance computing systems. Scientists in chemistry, weather modeling, and engineering simulation already run workflows on HPC systems and can treat a QPU as another accelerator alongside CPUs and GPUs. A workload manager for quantum devices was the missing piece, and Slurm 26.11 addresses that gap with HRES.
HRES ties a QPU to the exact nodes permitted to use it, which overcomes generic resource limits that bind devices to individual nodes. The model also avoids license-based approaches that leave compute nodes idle while they wait for the QPU. Operators can keep site policies in resources.yaml and rely on portable device logic across Slurm sites.
Open source communities QRMI and QDMI have released HRES plugins that connect Slurm clusters to quantum computers from multiple providers. Teams can deploy the QRMI plugin to link a cluster to quantum backends and explore the QDMI example for provider-managed queue scheduling. HRES documentation covers the plugin API and resources.yaml configuration for site-specific deployment.
What matters
- Slurm 26.11 introduces HRES plugins that treat QPUs as schedulable cluster-level devices.
- Operators can bind each QPU to allowed nodes and admit jobs before node allocation.
- Watch QRMI and QDMI HRES plugins as they connect Slurm clusters to multiple quantum providers.
Why it matters
Watch QRMI and QDMI HRES plugins as they connect Slurm clusters to multiple quantum providers.
This GenAI News article was prepared in original wording using reporting and materials published by NVIDIA Developer Blog. Source reference: https://developer.nvidia.com/blog/integrating-qpus-into-supercomputing-centers-using-slurm-hres-plugins/.
Drafted by the GenAI News review pipeline.
