GPU-Accelerated Distributed QAOA on Large-scale HPC Ecosystems

2025-06-12

AI Summary

To address the scalability limitations of high-dimensional, dense combinatorial optimization problems, this paper proposes DQAOA—a scalable, distributed implementation framework of the Quantum Approximate Optimization Algorithm (QAOA) tailored for the Frontier exascale supercomputer. Methodologically, we introduce a novel GPU-accelerated heterogeneous parallel architecture integrating MPI-based inter-node communication, the QFw quantum programming framework, and fine-grained CPU/GPU co-scheduling; we further design a problem decomposition strategy and a quantum-classical workload dynamic scheduling mechanism. Experimental evaluation on Frontier demonstrates up to 10× speedup over CPU-only simulation and enables solving problems with up to one thousand variables. To our knowledge, this work represents the first large-scale deployment of DQAOA on an exascale platform with over 100 million computing cores, marking a significant step toward the practical integration of hybrid quantum-classical algorithms into the HPC ecosystem.

Technology Category

• Search and Optimization: Distributed Search
• Machine Learning: Quantum Machine Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Quantum computing holds great potential to accelerate the process of solving complex combinatorial optimization problems. The Distributed Quantum Approximate Optimization Algorithm (DQAOA) addresses high-dimensional, dense problems using current quantum computing techniques and high-performance computing (HPC) systems. In this work, we improve the scalability and efficiency of DQAOA through advanced problem decomposition and parallel execution using message passing on the Frontier CPU/GPU supercomputer. Our approach ensures efficient quantum-classical workload management by distributing large problem instances across classical and quantum resources. Experimental results demonstrate that enhanced decomposition strategies and GPU-accelerated quantum simulations significantly improve DQAOA's performance, achieving up to 10x speedup over CPU-based simulations. This advancement enables better scalability for large problem instances, supporting the practical deployment of GPU systems for hybrid quantum-classical applications. We also highlight ongoing integration efforts using the Quantum Framework (QFw) to support future HPC-quantum computing systems.

Problem

• Enhancing scalability of Distributed QAOA on HPC systems
• Optimizing quantum-classical workload management for large problems
• Accelerating quantum simulations using GPU-based parallel execution

Innovation

• GPU-accelerated quantum simulations for speedup
• Distributed problem decomposition on HPC systems
• Hybrid quantum-classical workload management