Integration of Quantum Accelerators with High Performance Computing -- A Review of Quantum Programming Tools

TL;DR

The paper investigates how to integrate quantum accelerators into HPC ecosystems, arguing that quantum computing complements classical HPC rather than replaces it. It proposes an analysis blueprint to assess quantum programming tools from an HPC perspective and applies this framework to six notable QPTs to evaluate their fusion with classical workflows. Key findings show that no current toolchain fully satisfies integration requirements across hardware, software, and data management dimensions, revealing gaps in toolchain maturity and cross-technology support. The authors outline a roadmap toward a unified hybrid quantum-classical toolchain and advocate for standardized benchmarks to enable scalable, real-world quantum-HPC workloads.

Abstract

Quantum computing (QC) introduces a novel mode of computation with the possibility of greater computational power that remains to be exploited - presenting exciting opportunities for high performance computing (HPC) applications. However, recent advancements in the field have made clear that QC does not supplant conventional HPC, but can rather be incorporated into current heterogeneous HPC infrastructures as an additional accelerator, thereby enabling the optimal utilization of both paradigms. The desire for such integration significantly affects the development of software for quantum computers, which in turn influences the necessary software infrastructure. To date, previous review papers have investigated various quantum programming tools (QPTs) (such as languages, libraries, frameworks) in their ability to program, compile, and execute quantum circuits. However, the integration effort with classical HPC frameworks or systems has not been addressed. This study aims to characterize existing QPTs from an HPC perspective, investigating if existing QPTs have the potential to be efficiently integrated with classical computing models and determining where work is still required. This work structures a set of criteria into an analysis blueprint that enables HPC scientists to assess whether a QPT is suitable for the quantum-accelerated classical application at hand.

Figures (8)

• Figure 1: Abstracted workflow. Users are responsible for specifying their build dependencies and compilation toolchain, while service providers are responsible for scheduling and hardware execution. The workflow starts from the Login Node followed by the Code Analysis and Synthesis included in the Compilation. The binaries are distributed among computing resources, i. e., online, and executed on those.
• Figure 2: Bell state $\Phi^+$ preparation and subsequent measurement in quantum circuit notation.
• Figure 3: Trade-off between different - integration setups. From left to right, the demands of integration schemes change in different dimensions accordingly.
• Figure 4: Software-level integration. Starting from the left, the first diagram illustrates integration in the application layer where algorithmic modules of classical computing and quantum computing are integrated by hybrid programming languages or workflow systems. The second diagram illustrates integration in the compilation layer where hybrid source code is compiled. The final diagram illustrates integration in the hardware layer where binaries and pulse-level instructions are scheduled for execution on their target devices.
• Figure 5: Schematic representation of the three execution models referred to in this work. The arrows denote interaction between the two parties. Dashed arrows indicate interaction where hardware details are abstracted away. Thin arrows represent slow interaction and thick arrows fast interaction (based on Figure 2 in fu2021quingo).