Quantum processors are increasingly used as accelerators inside high-performance computing (HPC) workflows. A workflow runtime decides which quantum circuits to run, how to group them into provider jobs, and what to repeat after a failed attempt. Many search problems split into independent parts, but the runtime can use that structure only when the workflow exposes the parts and a validity check for each one. SemQBench measures what a runtime gains from this visibility. It compares seven execution policies for the same search problem, including two controls without amplitude amplification. The policies change one runtime choice at a time: whether to amplify, whether partial results may be combined across runs, whether the parts share one circuit, and what a retry keeps. We run them on the 20-qubit IQM Garnet processor using 106 distinct circuits in three jobs, and report circuit executions, active device time, and provider jobs separately before combining them in a cost model. Amplifying the problem as one whole circuit shows no statistically clear gain over random sampling and uses 6.7-104$\times$ more active device time. Exposing the parts and collecting valid partial results across a fixed batch of runs cuts the executions needed for 95% completion confidence by 2.20-4.00$\times$, and this policy is never worse than the other amplified policies in executions, device time, or jobs; it is selected in 99 of 108 overhead settings. Fewer executions do not always mean less device time: on one workload, executions fall from thirteen to four while active device time rises from 7.44 to 9.81 $μ$s. Exposing problem structure therefore gives the runtime a strong default policy and the measurements needed to decide when the simpler unamplified option is cheaper.
Quantum processors are increasingly used as accelerators inside high-performance computing (HPC) workflows. A workflow runtime decides which quantum circuits to run, how to group them into provider jobs, and what to repeat after a failed attempt. Many search problems split into independent parts, but the runtime can use that structure only when the workflow exposes the parts and a validity check for each one. SemQBench measures what a runtime gains from this visibility. It compares seven execution policies for the same search problem, including two controls without amplitude amplification. The policies change one runtime choice at a time: whether to amplify, whether partial results may be combined across runs, whether the parts share one circuit, and what a retry keeps. We run them on the 20-qubit IQM Garnet processor using 106 distinct circuits in three jobs, and report circuit executions, active device time, and provider jobs separately before combining them in a cost model. Amplifying the problem as one whole circuit shows no statistically clear gain over random sampling and uses 6.7-104× more active device time. Exposing the parts and collecting valid partial results across a fixed batch of runs cuts the executions needed for 95% completion confidence by 2.20-4.00×, and this policy is never worse than the other amplified policies in executions, device time, or jobs; it is selected in 99 of 108 overhead settings. Fewer executions do not always mean less device time: on one workload, executions fall from thirteen to four while active device time rises from 7.44 to 9.81 μs. Exposing problem structure therefore gives the runtime a strong default policy and the measurements needed to decide when the simpler unamplified option is cheaper.
Quantum computers are the next evolution of computing hardware. Quantum devices are being exposed through the same familiar cloud platforms used for classical computers, and enabling seamless execution of hybrid applications that combine quantum and classical components. Quantum devices vary in features, e.g., number of qubits, quantum volume, CLOPS, noise profile, queuing delays and resource cost. So, it may be useful to split hybrid workloads with either large quantum circuits or a large number of quantum circuits, into smaller units. In this paper, we profile two workload splitting techniques on IBM's Quantum Cloud: (1) Circuit parallelization, to split one large circuit into multiple smaller ones, and (2) Data parallelization to split a large number of circuits that run on one device to smaller batches of circuits that run on concurrently on multiple devices. These can improve the utilization of heterogeneous quantum hardware, but involve trade-offs. We evaluate these techniques on two key algorithmic classes: Variational Quantum Eigensolver (VQE) and Quantum Support Vector Machine (QSVM), and measure the impact on quality of the result, circuit execution times, quantum resource usage, and processing overheads, relative to a baseline without such parallelization. The results are obtained on real hardware and complemented by simulations. We see that (1) VQE with circuit cutting is ~ 39% better in ground state estimation than the uncut version, and (2) QSVM that combines data parallelization with a reduced feature set yields upto 3 ×improvement in quantum workload execution time and reduces quantum resource use by 3 ×, while providing comparable accuracy. Error mitigation can improve the accuracy by ≈ 7% and reduce resource footprint by ≈4% compared to the best case among the considered scenarios.
Quantum processors have crossed the one-hundred-qubit mark, but noise continues to limit circuit performance, while full quantum error correction remains too costly for routine use. Error suppression and mitigation therefore play an important role in extracting value from current hardware, yet independent comparisons of commercial solutions on identical workloads and devices remain scarce. We benchmark IBM Qiskit Runtime, Q-CTRL Performance Management, and Qedma QESEM on IBM Pittsburgh, a 156-qubit IBM Quantum Heron r3 processor. For Sampler workloads, we run Bernstein-Vazirani, quantum phase estimation, GHZ-state preparation, and randomized mirror circuits with up to 100 measured qubits, comparing raw execution, IBM measurement twirling, and Q-CTRL. For Estimator workloads, we measure chain-averaged magnetization and correlation observables of an eight-layer transverse-field Ising circuit at 25, 50, and 75 qubits against an exact matrix-product-state reference, comparing IBM raw execution, IBM TREX plus twirling, Q-CTRL, and QESEM. Q-CTRL produced the best results on the three structured Sampler workloads while keeping reported QPU times within the same order as the IBM configurations. Across six Ising observable and system-size cases, aggregate mean absolute error was 0.0883 for IBM raw execution, 0.0807 for IBM TREX plus twirling, 0.0285 for Q-CTRL, and 0.0188 for QESEM. Relative to raw execution, Q-CTRL and QESEM reduced aggregate error by factors of 3.10 and 4.70, respectively, while QESEM used 7.5 to 11.1 times the reported QPU time of Q-CTRL. These results show that managed error suppression and mitigation can substantially improve current hardware performance, but with distinct accuracy and execution-time tradeoffs.
Hybrid quantum-classical workflows are expected to underpin practical quantum computing applications, yet the quantum and HPC communities lack a shared framework for reasoning about where and when their integration requirements matter most. Such a framework must separate two distinct levels of analysis: the application level, where communication overhead affects runtime performance, and the real-time level, where it determines feasibility. To address this, we introduce a runtime model that decomposes workflow execution into quantum compute, classical compute, and communication costs. At the application level, a communication-to-computation ratio from this decomposition quantifies whether a workflow is communication-bound or compute-bound; at the real-time level, a feasibility constraint determines whether timing requirements can be met at all, with the reaction time setting the logical clock speed of fault-tolerant computation once they are. Application of this model to representative workflows demonstrates that co-location of quantum processors with HPC infrastructure offers negligible performance benefit for compute-intensive applications today, while tight integration remains crucial for real-time tasks such as quantum error correction needed for large scale quantum computations. However, we discuss how even at the application level these assessments may shift with hardware evolution, illustrating how the model can identify specific crossover conditions, and how, under fault tolerance, the reaction time can set application-level performance.
As quantum computing (QC) technologies mature, their integration into established high-performance computing (HPC) infrastructures is becoming a central objective for next-generation computing systems. However, unlocking the potential of hybrid platforms for computationally demanding workloads remains challenging. The mismatch between quantum and classical programming models, the limited maturity of quantum software stacks, and the scarcity of quantum processing units (QPUs) above all, necessitate scheduling strategies that go beyond standard HPC mechanisms to manage such heterogeneous and constrained resources.
To address this issue, we investigate three distinct methodologies for HPC-QC resource scheduling: time-based multiplexing, dynamic resource management, and workflow decomposition. Experimental validation on production HPC clusters and real quantum hardware demonstrates the effectiveness of these approaches under different workload scenarios. Malleability and workflow strategies significantly optimize classical resource utilization, reducing consumption by up to 45.7% and 64% respectively, proving to be best fitted for hybrid jobs where quantum and classical workloads are evenly balanced. Conversely, time-multiplexing enhances QPU utilization and reduces execution time at the cluster level, making it the optimal strategy for the opposite context, which is characterized by high classical-quantum workload imbalances. These findings underscore the practical viability of tailored scheduling strategies for hybrid HPC-QC environments and highlight their complementarity in building efficient, scalable software stacks for next-generation quantum-accelerated facilities.
Distributed quantum computing (DQC) is a promising near-term approach for executing circuits that exceed the capacity of a single quantum processing unit (QPU). Prior work has made significant progress on circuit partitioning and reconstruction, but practical DQC requires more than cutting a circuit into smaller pieces. A runtime system must determine whether to execute the circuit intact, cut it for sequential execution on one QPU, or distribute it across multiple QPUs, while accounting for queueing, communication, reconstruction, and hardware constraints. We present a resource-aware scheduler for DQC that compares these execution alternatives at runtime using explicit models of QPU execution, communication, and host-side reconstruction, together with hardware-informed timing derived from simulated IBM QPUs. By treating communication and reconstruction as first-class scheduling costs, the scheduler selects execution plans using a more realistic view of end-to-end overhead than cutting-only approaches. Across our experiments, reconstruction accounts for over 83% of cutting overhead, distributed execution reduces P90 latency by up to 39% relative to uncut single-QPU execution, and scheduling overhead itself remains negligible, never exceeding 0.038% of end-to-end job latency. Together, these results show that DQC must be treated as a runtime scheduling problem over both quantum and classical resources, not just circuit partitioning.
Quantum computers are making their way into High Performance Computing (HPC) centers as next-generation accelerators. Due to their physical implementation as mostly large appliances in separate racks, their number in typical data centers is significantly lower than the number of nodes offloading work to them, unlike the case with GPU accelerators. As a consequence, they form large-scale disaggregated infrastructures that pose a number of integration challenges due to their diverse implementation technologies and their need to be used as a shared resource for optimal utilization. Running hybrid High Performance Computing-Quantum Computing (HPCQC) applications in HPC environments, where the quantum portion is offloaded to the quantum processing units (QPUs), requires sophisticated resource management strategies to optimize resource utilization and performance. In this paper, we present one aspect of the Munich Quantum Software Stack (MQSS) - a Just-In-Time (JIT) compilation and execution software stack for quantum and hybrid quantum-HPC workloads - beneficial for integrating disaggregated quantum accelerators into traditional HPC workflows. MQSS is centered around a series of novel contributions aimed at optimizing the compilation process while managing the disaggregated resources ensuring efficient utilization of all available quantum resources. The key stages of our JIT compilation stack involve hardware-agnostic optimizations, quantum circuit cutting – when necessary – and a novel hardware platform selection process. Included here is description of the use of the MQSS’s Quantum Device Management Interface (QDMI), a unified interface between the software stack and the quantum accelerators, which allows connection to several, potentially different, quantum resources. Our methodology demonstrates notable advancements in HPCQC integration workflows, offering researchers a powerful framework for leveraging quantum computational capabilities. Overall, our work represents a significant step towards the practical integration of disaggregated quantum devices within the realm of supercomputing, unlocking new avenues for computational exploration and discovery.