Extending the OmpSs-2 Programming Model for Hybrid Quantum-Classical Programming

2025-02-28

AI Summary

High-performance computing (HPC) programs lack native support for hybrid quantum-classical parallel programming. Method: This work pioneers the extension of the OmpSs-2 task-based parallel programming model—originally designed for HPC—to quantum computing. It introduces a compiler-runtime co-designed quantum offloading mechanism: a Clang-based quantum-aware frontend, an extended Nanos6 runtime supporting instruction-level quantum task scheduling, and a lightweight, distributed, networked quantum simulator. Contribution/Results: The framework enables dynamic quantum task offloading, low-overhead classical–quantum communication, and portable cross-platform execution. End-to-end evaluation across four representative application classes—quantum random number generation, parameter sweeps, variational quantum algorithms, and hybrid quantum-classical convolutional neural networks—demonstrates substantial improvements in developer productivity and scalability. To our knowledge, this is the first systematic programming model enabling seamless integration of HPC and quantum computing.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Sampling/Simulation-based Search
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The OmpSs-2 programming model is used in HPC programs to parallelize code and offload code to accelerators. In this work, we extend the offloading capability to quantum computers. We explain the necessary changes to the Clang compiler and the Nanos6 runtime, which are both part of OmpSs-2. In addition, we develop a simulator that simulates a quantum computer in the network and receives the jobs offloaded by the runtime. Four detailed examples show how our programming model can be used to write hybrid quantum-classical software. The examples are random number generation, a parameter scan using the mean-field ansatz, a variational algorithm using this ansatz, and handwritten digit recognition using a hybrid convolutional neural network.

Problem

• Extends OmpSs-2 for quantum-classical hybrid programming.
• Modifies Clang compiler and Nanos6 runtime for quantum offloading.
• Develops simulator for quantum computer job offloading.

Innovation

• Extend OmpSs-2 for quantum-classical hybrid programming
• Modify Clang compiler and Nanos6 runtime for quantum offloading
• Develop quantum computer simulator for network-based job offloading