Reproduce “Enabling Hybrid HPCQC Workflows with a Heterogeneous Software Stack” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Muhammad Nufail Farooqi, Minh Chung, Burak Mete, Eric Mansfield, Bernd Hoffmann, Teemu Mattsson, Laura Schulz, Jorge Echavarria
Researchers from Leibniz Supercomputing Centre, IQM Quantum Computers, Argonne National Laboratory, and Munich Quantum Valley developed a heterogeneous software stack (MQSS) and a two-level scheduling model to integrate quantum processors into production-scale High Performance Computing (HPC) environments. Their solution, deployed on SuperMUC-NG with an IQM Q-Exa QPU, demonstrated minimal latency overhead for hybrid workflows, efficiently coordinating classical and quantum resources.
Key insights
- Treating QPUs as scheduler-managed accelerators via standard SLURM extension mechanisms allows for their non-intrusive integration into production HPC environments.
- A two-level scheduling approach effectively separates cluster-level resource management (SLURM) from device-level quantum execution optimization (QRM&CI), enabling efficient hybrid workflows.
- The architectural design and operational model are portable and directly applicable to next-generation HPC systems, providing a practical blueprint for future integrations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.