Quantum Computing for HPC Experts

Quantum computing is increasingly becoming an exploratory platform for tackling complex optimization problems, especially those that require navigating large combinatorial search spaces. Among these, task mapping and scheduling problems in HPC environments are of particular interest due to their inherent graph structure and computational difficulty. While practical quantum advantage has not yet been demonstrated, hybrid quantum-classical variational methods provide a promising framework for studying such problems on current noisy intermediate-scale quantum (NISQ) devices.

This presentation accompanies the half-day tutorial Quantum Computing for HPC Experts at ISC 2026. It introduces high-performance computing users to the core ideas of gate-based quantum computing and shows how hybrid quantum-classical algorithms can be applied to simplified HPC-inspired optimization problems.

The tutorial begins with the basic concepts needed to understand quantum computation, including qubits, quantum states, quantum gates, circuits, superposition, entanglement, measurement, and the current state of noisy intermediate-scale quantum hardware. These foundations are then connected to a concrete scheduling-inspired use case: assigning workflow tasks to two machines while accounting for communication costs between tasks.

The central computational workflow is based on the Variational Quantum Eigensolver (VQE). Participants learn how a small task-dependency graph can be simplified into a weighted graph partitioning problem, formulated as a balanced cut problem, rewritten using spin variables, and mapped directly to an Ising Hamiltonian. The VQE loop is then used to search for low-energy states of this Hamiltonian, where good bitstring solutions correspond to good task assignments.

Through a combination of lecture material, visual explanations, and hands-on notebook exercises, participants build the full pipeline step by step: constructing the graph, computing classical baselines, building the Hamiltonian, choosing an ansatz, running VQE with a classical optimizer, interpreting the resulting bitstrings, and benchmarking the solution quality.

The presentation concludes with a discussion of the practical prospects and limitations of quantum optimization for HPC, including noise, ansatz choice, optimizer behavior, scalability, and the role of quantum algorithms as exploratory tools rather than immediate replacements for classical solvers.

---

Contributors:

• Dhiraj Kumar
• Samira Altpeter
• Sadaf Ehtesabi
• Manoar Hossain
• Xin Wu