Programming tools for Analogue Quantum Computing in the High-Performance Computing Context -- A Review

TL;DR

The paper surveys analogue quantum programming tools and their readiness for integration with high-performance computing (HPC) systems. It introduces a taxonomy and readiness criteria across language design, performance, tooling, applications, learning curve, and portability, then analyzes a set of tools (SimuQ, Pulser, Bloqade, JaqalPaw, D-Wave Ocean SDK, QHDOPT, Amazon Braket AHS, C-to-D-Wave, XACC) for their suitability in HPC contexts. The findings reveal that analogue programming ecosystems remain largely experimental with limited HPC integration capabilities, lacking scalable data/resource management and cross-platform portability. The authors argue for new programming models and open, benchmarkable toolchains to enable production-grade analogue quantum software within HPC workflows and offer concrete recommendations for future research and development.

Abstract

Recent advances in quantum computing have brought us closer to realizing the potential of this transformative technology. While significant strides have been made in quantum error correction, many challenges persist, particularly in the realm of noise and scalability. Analogue quantum computing schemes, such as Analogue Hamiltonian Simulation and Quantum Annealing, offer a promising approach to address these limitations. By operating at a higher level of abstraction, these schemes can simplify the development of large-scale quantum algorithms. To fully harness the power of quantum computers, they must be seamlessly integrated with traditional high-performance computing (HPC) systems. While substantial research has focused on the integration of circuit-based quantum computers with HPC, the integration of analogue quantum computers remains relatively unexplored. This paper aims to bridge this gap by contributing in the following way: Comprehensive Survey: We conduct a comprehensive survey of existing quantum software tools with analogue capabilities. Readiness Assessment: We introduce a classification and rating system to assess the readiness of these tools for HPC integration. Gap Identification and Recommendations: We identify critical gaps in the landscape of analogue quantum programming models and propose actionable recommendations for future research and development.

Figures (6)

• Figure 1: Quantum circuit preparing Bell state from zero state. Lines represent wires or qubits; square with H denotes Hadamard gate which puts qubit in superposition; followed by two-qubit controlled NOT gate. A quantum circuit diagram is a temporal description, unlike a classical electronic circuit diagram, which is spatial.
• Figure 2: Proposed forms of HPC-Quantum integration. From left to right subfigures represent tighter integration with lower latency at the cost of technical challenges.
• Figure 3: Layers of a quantum software stack. From top to bottom, layers proceed from the higher abstraction levels to lower, starting with quantum application layer and ending with physical qubits of the target platform.
• Figure 4: Example HPC architecture with integrated QPU. Stack of GPU, CPU and QPU represents a computing node; Yellow lines connecting nodes are classical interconnects; Purple lines connecting nodes are quantum interconnects used for quantum communication.
• Figure 5: Example of a more specialised heterogeneous HPC architecture with different types of QPUs. All connections retain meaning from figure \ref{['fig:hpc-qc1']}. QA refers to Quantum Annealer; AHS to Analogue Hamiltonian Simulator; QPU to gate-based quantum machine.