qiskit integration

Addressing the challenges of heterogeneous resource scheduling, cross-paradigm data movement, and unified access to both NISQ and fault-tolerant quantum devices in deep quantum–HPC integration, this paper proposes the first hardware-agnostic quantum–supercomputing fusion software stack. Methodologically, it introduces (1) a quantum–classical unified resource management framework; (2) a hardware-abstracted programming interface and a cross-platform Quantum Platform Manager API; and (3) a quantum gateway supporting REST/gRPC, a hybrid scheduler, and a quantum-circuit co-optimization toolchain. The stack seamlessly integrates with mainstream HPC job schedulers (e.g., Slurm, PBS). Empirical evaluation on hybrid algorithms—including the variational quantum linear solver—on real supercomputing systems demonstrates significant improvements in quantum–classical resource utilization and task throughput. The design is both practically deployable and inherently scalable, establishing a foundational infrastructure for large-scale quantum–HPC convergence.

Large-scale quantum circuits pose significant challenges for efficient execution on noisy intermediate-scale quantum (NISQ) devices and classical simulators in quantum-classical hybrid computing. Method: This paper proposes a high-performance quantum-classical fusion framework featuring: (i) a modular, device-agnostic quantum programming interface; (ii) adaptive circuit stitching virtualization for dynamic decomposition and co-scheduling of large circuits across small-scale hardware and simulators; and (iii) a cross-platform compilation stack built upon QIR/LLVM, extending support to C/C++, Fortran, and Python, with heterogeneous backend code generation enabled via the Cray compilation framework. Contribution/Results: Evaluated on the HPE EX supercomputing platform, the framework demonstrates functional completeness and scalability to thousands of nodes for mixed workloads—including linear system solving, quantum optimization, and phase transition simulation—establishing the first unified programming environment supporting multi-language, multi-backend, and multi-architecture integration of HPC and quantum computing.

To address low resource scheduling efficiency and poor cross-platform compatibility in hybrid quantum-classical computing environments, this paper proposes Q-Middleware—the first quantum middleware built upon the HPC Pilot abstraction model—establishing a resource-task co-scheduling paradigm. Q-Middleware unifies management of heterogeneous resources, including quantum processing units (QPUs), CPUs, and GPUs, and supports parallel quantum workflows such as variational algorithms and circuit cutting. It enables seamless orchestration across quantum software frameworks (Qiskit, PennyLane) and diverse backends: real hardware (e.g., IBM Eagle) and high-performance simulators—including a 256-GPU distributed state-vector simulator. Innovatively integrating multi-backend adaptation, high-level framework bridging, and circuit-cutting-aware scheduling, Q-Middleware achieves 41-qubit distributed simulation on 256 GPUs. Experimental results demonstrate 15× and 3.5× speedups for quantum machine learning and circuit cutting, respectively, significantly advancing scalable hybrid quantum computing systems.

To address critical challenges in quantum computing—including difficult application development, poor portability, low reproducibility, and underutilized QPU resources—this paper proposes and implements a user-centric HPC-quantum computing (HPC-QC) integrated runtime environment. Methodologically, we design a vendor-neutral Quantum Resource Management Interface (QRMI) enabling unified integration of multiple quantum SDKs (e.g., Qiskit, Cirq, PennyLane); introduce a two-tier scheduling mechanism that coordinates HPC job schedulers with quantum middleware to improve QPU allocation efficiency; and integrate an observability monitoring stack for end-to-end operational visibility across hybrid quantum-classical workflows. Key contributions include: the first implementation of modular, cross-SDK quantum program deployment and portable hybrid runtime; empirical validation of a 37% improvement in QPU utilization under multi-SDK coexistence; and the establishment of the first production-grade observability framework for HPC-QC environments.

Quantum software development faces significant challenges, including scarce quantum hardware access, high computational overhead in classical simulation, and complex integration with heterogeneous accelerators. To address these issues, this paper proposes a quantum software toolchain designed for efficient iterative development. Our approach introduces three key innovations: (1) a lightweight remote computation scheduling mechanism enabling seamless switching between local development environments and remote quantum or simulation resources; (2) a plugin-based Jupyter kernel architecture supporting plug-and-play extensibility; and (3) integration of a high-performance distributed quantum simulator, extending scalable simulation capacity to 21–29 qubits. Experimental evaluation demonstrates up to 5× speedup on representative quantum circuit workloads, while maintaining robust support for complex circuit design and high-frequency iterative development. The toolchain significantly enhances quantum software engineering productivity and scalability.

This work addresses the inefficiency in developing hybrid quantum-classical algorithms caused by the current isolation of quantum processing units (QPUs) from classical high-performance computing (HPC) systems. To bridge this gap, the paper proposes a Quantum-Centric Supercomputing (QCSC) reference architecture that deeply integrates QPUs, GPUs, and CPUs through co-design across hardware, middleware, and application layers, enabling end-to-end workflows for domains such as quantum chemistry and materials science. The study further establishes, for the first time, a systematic three-stage evolutionary roadmap for QCSC: beginning with QPUs as specialized accelerators, progressing to heterogeneous cooperative scheduling, and ultimately achieving full-stack co-design. This architecture provides a standardized technical pathway for quantum-HPC convergence, substantially enhancing algorithmic exploration efficiency and system scalability.

Integrating quantum computers into high-performance computing (HPC) centers is often hindered by reliance on vendor-specific adapters, impeding production-level deployment. This work proposes and implements a standardized hardware-software boundary based on the Quantum Device Management Interface (QDMI), using an IQM superconducting quantum system as a case study to seamlessly integrate with the Slurm job scheduler and Qiskit user workflows. The implementation demonstrates that QDMI provides a unified interface for coordinating HPC resources and quantum hardware across different vendors and deployment models, thereby eliminating redundant custom development. The authors have open-sourced their QDMI-on-IQM integration, significantly simplifying the deployment of quantum-HPC hybrid systems in today’s heterogeneous computing environments.

This work addresses the critical challenge of efficiently integrating quantum processing units (QPUs) into existing high-performance computing (HPC) systems to enable unified scheduling and协同 execution of classical and quantum resources. The paper proposes a Quantum-integrated High-Performance Computing (QHPC) architecture—the first HPC framework explicitly designed for quantum-classical convergence—featuring a layered design that holistically manages CPUs, GPUs, FPGAs, and QPUs. Key innovations include a user request abstraction layer offering a Slurm-like unified interface, a quantum-aware scheduling algorithm, a hybrid workflow engine, and a hierarchical execution model supported by high-speed interconnects. This architecture provides scalable heterogeneous computing support for emerging applications such as quantum chemistry, materials discovery, combinatorial optimization, and climate modeling, thereby laying the foundation for a future quantum-classical HPC ecosystem.

This work addresses the lack of a unified mechanism for efficiently integrating and scheduling quantum computing resources within heterogeneous high-performance computing (HPC) environments. To bridge this gap, the authors propose QRMI, a lightweight, vendor-agnostic middleware that abstracts quantum resources as schedulable units and enables their co-management alongside classical CPU and GPU resources. QRMI seamlessly interoperates with diverse job schedulers—including PBS, LSF, Grid Engine, Kubernetes, and Flux—through a standardized API and modular architecture. Without requiring deep modifications to existing schedulers, QRMI achieves consistent, cross-platform access to and efficient scheduling of quantum resources for the first time. Experimental validation demonstrates its generality and practicality across both on-premises and cloud-based heterogeneous infrastructures.

This work addresses the resource scheduling challenges faced by hybrid quantum-classical applications in heterogeneous and dynamic computing environments, where existing high-performance computing (HPC) schedulers lack application semantics awareness and runtime adaptability. To overcome these limitations, the authors propose a four-layer middleware architecture that integrates an abstract execution model for hybrid applications, a Pilot-Quantum dynamic scheduling framework, and the Q-Dreamer performance modeling toolkit. This integrated approach enables application-aware, adaptive resource management and optimized quantum circuit partitioning. The system supports coordinated scheduling across CPU, GPU, and quantum processing unit (QPU) backends and has been validated on the Perlmutter and NVIDIA DGX platforms. Experimental results demonstrate that Q-Dreamer achieves an 82% accuracy rate in predicting optimal circuit-cut configurations.