pith. sign in

arxiv: 2508.04217 · v1 · submitted 2025-08-06 · 🪐 quant-ph · cs.DC

Dynamic Solutions for Hybrid Quantum-HPC Resource Allocation

Roberto Rocco , Simone Rizzo , Matteo Barbieri , Gabriella Bettonte , Elisabetta Boella , Fulvio Ganz , Sergio Iserte , Antonio J. Pe\~na
show 11 more authors
Petter Sand\r{a}s Alberto Scionti Olivier Terzo Chiara Vercellino Giacomo Vitali Paolo Viviani Jonathan Frassineti Sara Marzella Daniele Ottaviani Iacopo Colonnelli Daniele Gregori
This is my paper

Pith reviewed 2026-05-19 00:51 UTC · model grok-4.3

classification 🪐 quant-ph cs.DC
keywords hybrid quantum-HPCresource allocationdynamic allocationmalleabilityworkflow-based strategyquantum acceleratorsresource utilizationHPC integration
0
0 comments X

The pith

Malleability and workflow strategies let hybrid HPC-quantum systems release classical resources during quantum offloads and reallocate them afterward.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows how two approaches, one based on malleability and one on workflows, manage resources when quantum computers act as accelerators inside classical high-performance computing setups. These methods free classical processors and memory while the quantum device handles its part of the calculation, then reclaim them once that part ends. A reader would care because quantum hardware is expected to speed up only certain tasks, leaving the rest of a large job on classical machines that currently stay occupied even during the quantum phase. Experiments on a concrete hybrid workload confirm that this switching reduces wasted capacity.

Core claim

The paper claims that a malleability-based approach together with a workflow-based strategy allows hybrid HPC-quantum workloads to release classical resources when computations are offloaded to the quantum computer and to reallocate those resources once quantum processing finishes, with experiments on a hybrid use case demonstrating the resulting gains in utilization.

What carries the argument

The malleability-based approach and workflow-based strategy that support releasing classical resources during quantum offloads and reallocating them afterward.

If this is right

  • Classical resources stay available for other jobs while the quantum computer runs its portion of the workload.
  • Overall utilization rises because the classical side no longer idles during the quantum phase.
  • Both the malleability method and the workflow method produce measurable gains on a tested hybrid application.
  • Resource managers can treat quantum offloads as explicit points where classical capacity can be freed and later reclaimed.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar release-and-reallocate logic could apply to other accelerators such as GPUs inside the same HPC environment.
  • Job schedulers would need new interfaces that recognize quantum-phase boundaries to make the approach routine.
  • The techniques might extend to multi-user queues where one user's quantum task frees nodes for another's classical work.

Load-bearing premise

Hybrid HPC-quantum workloads can be structured such that offloading to the quantum computer allows seamless release and later reallocation of classical resources without prohibitive overheads or compatibility barriers.

What would settle it

An experiment on the same hybrid use case that measures either no net reduction in idle classical resources or an overhead from release and reallocation larger than the time saved by quantum acceleration would undermine the practical benefit.

Figures

Figures reproduced from arXiv: 2508.04217 by Alberto Scionti, Antonio J. Pe\~na, Chiara Vercellino, Daniele Gregori, Daniele Ottaviani, Elisabetta Boella, Fulvio Ganz, Gabriella Bettonte, Giacomo Vitali, Iacopo Colonnelli, Jonathan Frassineti, Matteo Barbieri, Olivier Terzo, Paolo Viviani, Petter Sand\r{a}s, Roberto Rocco, Sara Marzella, Sergio Iserte, Simone Rizzo.

Figure 1
Figure 1. Figure 1: Data and processes of the target application. Yellow shapes are data [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Timeline of cumulative usage of classical nodes on our SLURM [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
read the original abstract

The integration of quantum computers within classical High-Performance Computing (HPC) infrastructures is receiving increasing attention, with the former expected to serve as accelerators for specific computational tasks. However, combining HPC and quantum computers presents significant technical challenges, including resource allocation. This paper presents a novel malleability-based approach, alongside a workflow-based strategy, to optimize resource utilization in hybrid HPC-quantum workloads. With both these approaches, we can release classical resources when computations are offloaded to the quantum computer and reallocate them once quantum processing is complete. Our experiments with a hybrid HPC-quantum use case show the benefits of dynamic allocation, highlighting the potential of those solutions.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a malleability-based approach together with a workflow-based strategy for dynamic resource allocation in hybrid HPC-quantum systems. The central claim is that these methods permit classical resources to be released when work is offloaded to a quantum device and reallocated upon completion, with experiments on a hybrid use case demonstrating the resulting benefits in resource utilization.

Significance. The problem of efficient resource management at the HPC-quantum boundary is timely. If the proposed techniques can be shown to operate with low overhead on production schedulers and real quantum hardware, the work would provide a practical contribution to hybrid-system design. At present the absence of quantitative evidence prevents a clear assessment of whether the claimed gains are realized.

major comments (2)
  1. Abstract and experimental description: the manuscript asserts that 'experiments with a hybrid HPC-quantum use case show the benefits of dynamic allocation,' yet supplies no methodology, quantitative metrics (e.g., utilization improvement, reallocation latency), baselines, or error analysis. This omission is load-bearing for the central claim and must be remedied before the result can be evaluated.
  2. Malleability-based approach section: the strategy presupposes that the underlying HPC runtime (e.g., Slurm or equivalent) supports malleable job resizing with negligible reconfiguration cost and that quantum turnaround is short enough for freed nodes to remain useful. No scheduler details, measured overheads, or compatibility discussion are provided; these assumptions are central to the feasibility argument.
minor comments (1)
  1. Clarify whether quantum access in the reported experiments used real hardware, emulation, or simulation, and state the specific scheduler and quantum backend employed.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and timely comments. We address each major point below and describe the revisions that will be incorporated to strengthen the manuscript.

read point-by-point responses

  1. Referee: Abstract and experimental description: the manuscript asserts that 'experiments with a hybrid HPC-quantum use case show the benefits of dynamic allocation,' yet supplies no methodology, quantitative metrics (e.g., utilization improvement, reallocation latency), baselines, or error analysis. This omission is load-bearing for the central claim and must be remedied before the result can be evaluated.

    Authors: We agree that the current description of the experiments is insufficient to allow evaluation of the claimed benefits. In the revised manuscript we will add a dedicated experimental section that details the hybrid use-case workflow, the precise quantitative metrics (including measured utilization gains and reallocation latencies), the chosen baselines, and the associated error analysis. These additions will directly support the central claim. revision: yes

  2. Referee: Malleability-based approach section: the strategy presupposes that the underlying HPC runtime (e.g., Slurm or equivalent) supports malleable job resizing with negligible reconfiguration cost and that quantum turnaround is short enough for freed nodes to remain useful. No scheduler details, measured overheads, or compatibility discussion are provided; these assumptions are central to the feasibility argument.

    Authors: We acknowledge that the feasibility discussion must be grounded in concrete scheduler considerations. The revised section will include an explicit compatibility analysis with production schedulers such as Slurm, any measured or reported reconfiguration overheads, and the range of quantum turnaround times for which node reallocation remains advantageous. This will clarify the practical scope of the approach. revision: yes

Circularity Check

0 steps flagged

No circularity: descriptive engineering paper with no derivations or fitted predictions

full rationale

The manuscript presents two practical strategies (malleability-based and workflow-based) for releasing and reallocating classical HPC resources around quantum offloads. No equations, first-principles derivations, parameter fits, or predictions appear anywhere in the text. Claims rest on experimental demonstration of a hybrid use case rather than any self-referential construction, self-citation chain, or renaming of known results. The work is therefore self-contained as an applied systems contribution with no load-bearing circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only access prevents identification of any free parameters, axioms, or invented entities; none are explicitly described or implied in the provided summary.

pith-pipeline@v0.9.0 · 5710 in / 1141 out tokens · 40780 ms · 2026-05-19T00:51:41.928670+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Test Taxonomy and Continuous Integration Ecosystem for Dynamic Resource Management in HPC

    cs.DC 2026-04 unverdicted novelty 6.0

    Introduces a test taxonomy and HPC CI ecosystem to improve validation of dynamic resource management frameworks, evaluated via the DMR case study.

  2. Hybrid Quantum-HPC Middleware Systems for Adaptive Resource, Workload and Task Management

    quant-ph 2026-04 unverdicted novelty 5.0

    The authors present Pilot-Quantum, a middleware for adaptive resource management in hybrid quantum-HPC systems, along with execution motifs and a performance modeling toolkit called Q-Dreamer.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages · cited by 2 Pith papers

  1. [1]

    Np-hard but no longer hard to solve? using quantum computing to tackle optimization problems,

    R. Au-Yeung, N. Chancellor, and P. Halffmann, “Np-hard but no longer hard to solve? using quantum computing to tackle optimization problems,” Frontiers in Quantum Science and Technology , vol. 2, Feb. 2023

    work page 2023

  2. [2]

    Bringing quantum acceleration to supercomputers,

    M. Ruefenacht, B. G. Taketani et al. , “Bringing quantum acceleration to supercomputers,” IQM/LRZ Technical Report, https://www. quantu m. lrz. de/fileadmin/QIC/Downloads/IQM HPC-QC-Integration-White paper. pdf, 2022

    work page 2022

  3. [3]

    Integrating quantum computing resources into scientific hpc ecosystems,

    T. Beck, A. Baroni et al. , “Integrating quantum computing resources into scientific hpc ecosystems,” Future Generation Computer Systems , vol. 161, pp. 11–25, 2024

    work page 2024

  4. [4]

    Integration of quantum accelerators into hpc: Toward a unified quantum platform,

    A. Elsharkawy, X. Guo, and M. Schulz, “Integration of quantum accelerators into hpc: Toward a unified quantum platform,” in2024 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 01, 2024, pp. 774–783

    work page 2024

  5. [5]

    Pilot-quantum: A quantum-hpc middleware for resource, workload and task management,

    P. Mantha, F. J. Kiwit et al., “Pilot-quantum: A quantum-hpc middleware for resource, workload and task management,” 2024

    work page 2024

  6. [6]

    Bbq-mis: A parallel quantum algorithm for graph coloring problems,

    C. Vercellino, G. Vitali et al., “Bbq-mis: A parallel quantum algorithm for graph coloring problems,” in 2023 IEEE International Conference on Quantum Computing and Engineering (QCE) , vol. 02, 2023, pp. 141–147

    work page 2023

  7. [7]

    Performance analysis of an optimization al- gorithm for metamaterial design on the integrated high-performance computing and quantum systems,

    S. Kim and I.-S. Suh, “Performance analysis of an optimization al- gorithm for metamaterial design on the integrated high-performance computing and quantum systems,” 2024

    work page 2024

  8. [8]

    Challenges in hpcqc integration,

    A. Elsharkawy, X.-T. M. To et al. , “Challenges in hpcqc integration,” in 2023 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 02, 2023, pp. 405–406

    work page 2023

  9. [9]

    Building a software stack for quantum-hpc integration,

    A. Shehata, P. Groszkowski et al. , “Building a software stack for quantum-hpc integration,” 2025

    work page 2025

  10. [10]

    A survey on integrating quantum computers into high performance computing systems,

    P. D ¨obler and M. S. Jattana, “A survey on integrating quantum computers into high performance computing systems,” 2025

    work page 2025

  11. [11]

    Accelerating HPC With Quantum Computing: It Is a Software Challenge Too,

    M. Schulz, M. Ruefenacht et al. , “Accelerating HPC With Quantum Computing: It Is a Software Challenge Too,” Computing in Science & Engineering, vol. 24, no. 4, pp. 60–64, Jul. 2022

    work page 2022

  12. [12]

    High-performance computing with quantum processing units,

    K. A. Britt and T. S. Humble, “High-performance computing with quantum processing units,” ACM Journal on Emerging Technologies in Computing Systems (JETC) , vol. 13, no. 3, pp. 1–13, 2017

    work page 2017

  13. [13]

    A Conceptual Architecture for a Quantum-HPC Middleware,

    N. Saurabh, S. Jha, and A. Luckow, “A Conceptual Architecture for a Quantum-HPC Middleware,” Aug. 2023

    work page 2023

  14. [14]

    Assessing the elephant in the room in scheduling for current hybrid hpc-qc clusters,

    P. Viviani, R. Rocco et al. , “Assessing the elephant in the room in scheduling for current hybrid hpc-qc clusters,” 2025

    work page 2025

  15. [15]

    Workflows community summit 2022: A roadmap revolution,

    R. Ferreira Da Silva, R. Badia et al. , “Workflows community summit 2022: A roadmap revolution,” Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States), Tech. Rep., 03 2023

    work page 2022

  16. [16]

    Paving the way to hybrid quantum–classical scientific workflows,

    S. S. Cranganore, V . De Maio et al. , “Paving the way to hybrid quantum–classical scientific workflows,” Future Generation Computer Systems, vol. 158, pp. 346–366, 2024

    work page 2024

  17. [17]

    Rigoletto: A workflow definition language for hybrid quantum-classical scientific applications,

    V . De Maio, D. Bork, and I. Brandic, “Rigoletto: A workflow definition language for hybrid quantum-classical scientific applications,” in 2024 26th International Conference on Business Informatics (CBI) , 2024, pp. 40–49

    work page 2024

  18. [18]

    Streamflow: Cross-breeding cloud with HPC,

    I. Colonnelli, B. Cantalupo et al. , “Streamflow: Cross-breeding cloud with HPC,” IEEE Trans. Emerg. Top. Comput., vol. 9, no. 4, pp. 1723– 1737, 2021

    work page 2021

  19. [19]

    Clustering aggregation as maximum-weight independent set,

    N. Li and L. J. Latecki, “Clustering aggregation as maximum-weight independent set,” in Neural Information Processing Systems , 2012

    work page 2012

  20. [20]

    A clustering aggregation algorithm on neutral-atoms and annealing quantum processors,

    R. Scotti, G. Bettonte et al. , “A clustering aggregation algorithm on neutral-atoms and annealing quantum processors,” 2024

    work page 2024

  21. [21]

    DMRlib: Easy-coding and Efficient Resource Management for Job Malleability,

    S. Iserte, R. Mayo et al., “DMRlib: Easy-coding and Efficient Resource Management for Job Malleability,” IEEE Transactions on Computers , vol. 70, pp. 1443–1457, Sep. 2020

    work page 2020

  22. [22]

    Hybrid classical-quantum clustering aggregation,

    E. C. Engineering, “Hybrid classical-quantum clustering aggregation,” https://github.com/E4-Computer-Engineering/clustering-mis, 2025

    work page 2025

  23. [23]

    Multivariate observations,

    J. MacQueen, “Multivariate observations,” in Proceedings ofthe 5th Berkeley Symposium on Mathematical Statisticsand Probability , vol. 1, 1967, pp. 281–297

    work page 1967

  24. [24]

    A density-based algorithm for discovering clusters in large spatial databases with noise,

    M. Ester, H.-P. Kriegel et al., “A density-based algorithm for discovering clusters in large spatial databases with noise,” in Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, ser. KDD’96. AAAI Press, 1996, p. 226–231

    work page 1996

  25. [25]

    fastcluster: Fast hierarchical, agglomerative clustering routines forrandpython,

    D. M ¨ullner, “fastcluster: Fast hierarchical, agglomerative clustering routines forrandpython,” Journal of Statistical Software , vol. 53, no. 9, 2013

    work page 2013

  26. [26]

    Optimization by simulated annealing,

    S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, “Optimization by simulated annealing,” Science, vol. 220, no. 4598, pp. 671–680, 1983

    work page 1983

  27. [27]

    A terminology for scientific workflow systems,

    F. Suter, T. Coleman et al. , “A terminology for scientific workflow systems,” Future Generation Computer Systems , vol. 174, p. 107974, 2026

    work page 2026

  28. [28]

    Nextflow enables reproducible computational workflows,

    P. Di Tommaso, M. Chatzou et al. , “Nextflow enables reproducible computational workflows,” Nature biotechnology , vol. 35, no. 4, pp. 316–319, 2017

    work page 2017

  29. [29]

    Pycompss: Parallel computational work- flows in python,

    E. Tejedor, Y . Becerra et al., “Pycompss: Parallel computational work- flows in python,” The International Journal of High Performance Computing Applications, vol. 31, no. 1, pp. 66–82, 2017

    work page 2017

  30. [30]

    Introducing SWIRL: an intermediate representation language for scientific workflows,

    I. Colonnelli, D. Medic et al. , “Introducing SWIRL: an intermediate representation language for scientific workflows,” in Formal Methods - 26th International Symposium, FM 2024, Milan, Italy, September 9-13, 2024, Proceedings, Part I, ser. Lecture Notes in Computer Science, vol. 14933. Springer, 2024, pp. 226–244

    work page 2024

  31. [31]

    Methods included: Standardizing com- putational reuse and portability with the common workflow language,

    M. R. Crusoe, S. Abeln et al., “Methods included: Standardizing com- putational reuse and portability with the common workflow language,” Communication of the ACM , 2022

    work page 2022

  32. [32]

    Federated learning meets HPC and cloud,

    I. Colonnelli, B. Casella et al. , “Federated learning meets HPC and cloud,” in Astrophysics and Space Science Proceedings , vol. 60. Springer, 2023, p. 193–199

    work page 2023

  33. [33]

    Packing Schemes for Gang Scheduling,

    D. G. Feitelson, “Packing Schemes for Gang Scheduling,” in Pro- ceedings of the Workshop on Job Scheduling Strategies for Parallel Processing, ser. IPPS ’96. Berlin, Heidelberg: Springer-Verlag, Apr. 1996, pp. 89–110

    work page 1996

  34. [34]

    Towards the democratization and standardization of dynamic resources with MPI spawning,

    S. Iserte, I. Mart ´ın-´Alvarez et al. , “Towards the democratization and standardization of dynamic resources with MPI spawning,” in in Pro- ceedings of the International Conference on Parallel Processing & Applied Mathematics (PPAM). Best Paper Award , 2024

    work page 2024

  35. [35]

    A Survey on Malleability Solutions for High-Performance Distributed Computing,

    J. I. Aliaga, M. Castillo et al., “A Survey on Malleability Solutions for High-Performance Distributed Computing,”Applied Science, vol. 12, pp. 1–32, May 2022

    work page 2022

  36. [36]

    Malleability in Modern HPC Sys- tems: Current Experiences, Challenges, and Future Opportunities,

    A. Tarraf, M. Schreiber et al. , “Malleability in Modern HPC Sys- tems: Current Experiences, Challenges, and Future Opportunities,” IEEE Transactions on Parallel and Distributed Systems , pp. 1–14, Jun. 2024, conference Name: IEEE Transactions on Parallel and Distributed Sys- tems

    work page 2024

  37. [37]

    A Study of the Effect of Process Malleability in the Energy Efficiency on GPU-based Clusters,

    S. Iserte and K. Rojek, “A Study of the Effect of Process Malleability in the Energy Efficiency on GPU-based Clusters,” Journal of Supercom- puting, vol. 76, pp. 255–274, Oct. 2020

    work page 2020

  38. [38]

    Dynamic Reconfiguration of Non-iterative Scientific Applications: A Case Study with HPG-aligner,

    S. Iserte, H. Mart ´ınezet al., “Dynamic Reconfiguration of Non-iterative Scientific Applications: A Case Study with HPG-aligner,” International Journal of High Performance Computing Application, vol. 33, pp. 1–10, Aug. 2018

    work page 2018

  39. [39]

    DMR API: Improving Cluster Productivity by Turning Applications into Malleable,

    S. Iserte, R. Mayo et al. , “DMR API: Improving Cluster Productivity by Turning Applications into Malleable,” Parallel Computing, vol. 78, pp. 54–66, Jul. 2018

    work page 2018

  40. [40]

    Bridging the Gap Between Genericity and Programmability of Dynamic Resources in HPC,

    D. Huber, S. Iserte et al. , “Bridging the Gap Between Genericity and Programmability of Dynamic Resources in HPC,” in ISC High Performance 2025 Research Paper Proceedings (40th International Conference), Jun. 2025, pp. 1–11

    work page 2025

  41. [41]

    Resource optimization with MPI process malleability for dynamic workloads in HPC clusters,

    S. Iserte, I. Mart ´ın-´Alvarez et al. , “Resource optimization with MPI process malleability for dynamic workloads in HPC clusters,” Future Generation Computer Systems , p. 107949, Jun. 2025

    work page 2025

  42. [42]

    Hyperqueue: Efficient and ergonomic task graphs on hpc clusters,

    J. Ber ´anek, A. B ¨ohm et al., “Hyperqueue: Efficient and ergonomic task graphs on hpc clusters,” SoftwareX, vol. 27, p. 101814, 2024

    work page 2024

  43. [43]

    Qdmi - quantum device management interface: Hardware-software interface for the munich quantum software stack,

    R. Wille, L. Schmid et al. , “Qdmi - quantum device management interface: Hardware-software interface for the munich quantum software stack,” in 2024 IEEE International Conference on Quantum Computing and Engineering (QCE) , vol. 02, 2024, pp. 573–574

    work page 2024

  44. [44]

    Slurm heterogeneous jobs for hybrid classical-quantum workflows,

    A. Esposito and U.-U. Haus, “Slurm heterogeneous jobs for hybrid classical-quantum workflows,” 2025

    work page 2025

  45. [45]

    Distributed quantum circuit cutting for hybrid quantum-classical high-performance computing,

    M. Tejedor, B. Casas et al. , “Distributed quantum circuit cutting for hybrid quantum-classical high-performance computing,” 2025

    work page 2025