IQM Supports NVIDIA’s Work to Make Quantum Computers Schedulable in Slurm
AI-summarised brief · reviewed before publication
IQM Quantum Computers announced at SC26 that it is backing NVIDIA’s effort to make quantum processing units (QPUs) a native, schedulable resource in the Slurm workload manager. NVIDIA’s Slurm 26.11 release adds a Hierarchical Resources (HRES) plugin that lets supercomputing centers describe a QPU as a shared cluster‑level asset and link it to external control software. IQM partner MQSC has released an open‑source orchestrator that combines Slurm HRES with the Quantum Device Management Interface (QDMI) to manage access to IQM systems such as the Emerald QPU. The orchestrator is already used at sites including ORNL, VTT, LRZ and CINECA, where researchers submit hybrid quantum‑classical jobs through the same queues as CPUs and GPUs. IQM is testing HRES integration in its hybrid setups and plans validation runs as HPC centers upgrade their Slurm installations.
💡 Why It Matters
- · Enabling QPUs to be scheduled like traditional CPUs and GPUs removes a major integration barrier, allowing quantum hardware to be woven directly into existing HPC workflows.