IQM Supports NVIDIA’s Work to Make Quantum Computers Schedulable in Slurm

IQM is collaborating with NVIDIA and MQSC to make quantum processors schedulable through the Slurm workload manager.

IQM Supports NVIDIA’s Work to Make Quantum Computers Schedulable in Slurm

Integrating Quantum Processors into Slurm

IQM Quantum Computers announced support for NVIDIA’s effort to expose quantum processing units (QPUs) as first‑class resources in the Slurm workload manager. By leveraging NVIDIA’s new Slurm 26.11 plugin for Hierarchical Resources (HRES) and MQSC’s open‑source orchestrator, IQM aims to let high‑performance computing (HPC) centers schedule quantum jobs alongside traditional CPU and GPU workloads.

Why Scheduling Matters for Quantum Computing

HPC environments rely on a single scheduling system to allocate compute, storage, and network resources. Until now, quantum hardware has required bespoke integration layers, making it difficult for users to submit hybrid quantum‑classical workloads through familiar tools. By treating a QPU as a shared cluster‑level resource, Slurm can manage access, queue jobs, and enforce policies just like any other compute node.

The Technical Stack

  • Slurm 26.11 HRES – Extends Slurm’s resource model to describe non‑standard devices such as QPUs.
  • Quantum Device Management Interface (QDMI) – Provides a standard API for controlling quantum hardware, exposing status, configuration, and job execution.
  • MQSC Orchestrator – Open‑source software that bridges HRES and QDMI, translating Slurm scheduling decisions into quantum device commands.

IQM’s partner MQSC built the orchestrator, and IQM is evaluating how the HRES model maps onto its hybrid quantum‑classical setups. Validation will occur in staging environments at several HPC centers that are upgrading to the new Slurm release.

Early Adoption and Outlook

IQM already runs alongside classical systems at major HPC sites such as ORNL, VTT, LRZ, and CINECA, where researchers submit hybrid jobs through existing queues. The new integration promises to streamline that workflow, reducing the need for site‑specific glue code and lowering the barrier for scientists to incorporate quantum acceleration into their simulations and data‑analysis pipelines.

The collaboration will be showcased at the SC26 conference in Chicago, where IQM will host a booth and provide deeper technical details via blog posts from both IQM and NVIDIA.

---

This article summarizes the announced partnership and technical approach without claiming performance improvements or new quantum milestones.

Hype check

Not a research claim

The article describes a collaboration to integrate quantum hardware with Slurm, which is a partnership announcement without a measurable result to assess.

We rate research and technical results only. How we rate hype →

Source

Read original source

Why we picked this

Accepted from curated QuantumUnzipped source (relevance scoring disabled).

← Back to all articles