pith. sign in

arxiv: 2604.24059 · v2 · submitted 2026-04-27 · 🪐 quant-ph

Beyond Monolithic Scaling: Modularity and Heterogeneity as an Architectural Imperative for Utility-Scale Quantum Computing

Bo Fan , Renzhou Fang , Yuntao Zhang , Xiaolong Yuan , Dafa Zhao This is my paper

Pith reviewed 2026-05-14 20:56 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum computingmodularityscalabilityfault toleranceLOCCquantum error correctiondistributed quantum computingcoherence time
0
0 comments X

The pith

Classical control latency grows with system size while qubit coherence stays bounded, forcing modular architectures for quantum computers beyond roughly 10^5 physical qubits.

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

The paper contends that quantum computers hit a hard limit not from qubit fabrication but from the growing time classical signals need to coordinate across larger machines, which eventually exceeds the fixed window in which qubits remain coherent. This mismatch creates a superlinear penalty that makes single-block synchronization physically impossible past a certain scale, pushing the design toward separate modules that perform local quantum operations and exchange classical messages. The authors formalize the transition with a scaling law and introduce a reserve-commit protocol that aborts out-of-time operations while turning those failures into known erasure errors that ease later error correction. Their projections place the required shift at 100,000 to one million physical qubits under near-term hardware assumptions, coinciding with the onset of useful fault-tolerant computation. A reader would care because the argument reframes the engineering roadmap from ever-larger monolithic chips to networked, time-aware modules.

Core claim

The temporal mismatch between macroscopic classical coordination latency that grows with system diameter and strictly bounded microscopic quantum coherence produces a superlinear geometric penalty that breaches the classical control light cone, rendering monolithic synchronization impossible and mandating modular decomposition together with a shift from global unitaries to Local Operations and Classical Communication managed by a layered semantic architecture and time-aware Reserve-Commit protocol.

What carries the argument

The governing scaling law 1+ε > γ, where ε is the superlinear geometric penalty arising from the latency-coherence mismatch, which enforces the structural phase transition to modular architectures and LOCC.

If this is right

  • Modular decomposition and LOCC become mandatory once the system crosses N_c ~ 10^5--10^6 physical qubits.
  • Global unitary operations must be replaced by local operations plus classical communication to stay within coherence budgets.
  • The Reserve-Commit protocol converts scheduling-induced failures into location-known erasure metadata that relaxes downstream QEC fidelity requirements.
  • Time-aware distributed orchestration aligns exactly with the scale of early fault-tolerant utility under realistic transduction efficiencies.

Where Pith is reading between the lines

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

  • Heterogeneous qubit technologies across modules could be matched to local coherence and control needs without forcing uniformity.
  • The same latency-coherence argument may apply to large-scale quantum networks or sensor arrays facing analogous timing constraints.
  • Smaller modular testbeds could be used to validate the predicted crossover scaling before full utility-scale hardware is built.
  • Quantum error correction codes would need explicit interfaces to accept the erasure metadata generated by the protocol.

Load-bearing premise

Classical coordination latency must increase with the physical diameter of the system while quantum coherence times remain fixed and independent of scale.

What would settle it

A working monolithic quantum processor containing more than 10^6 physical qubits that maintains full synchronization and low error rates without modular decomposition or distributed classical control.

Figures

Figures reproduced from arXiv: 2604.24059 by Bo Fan, Dafa Zhao, Renzhou Fang, Xiaolong Yuan, Yuntao Zhang.

Figure 1
Figure 1. Figure 1: FIG. 1 view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 view at source ↗
read the original abstract

Scalable quantum computing is fundamentally bottlenecked not by qubit count or fabrication yield, but by a rigid temporal mismatch: macroscopic classical coordination latency ($\tau_c$) inevitably grows with system diameter, while microscopic quantum coherence ($\tau_q$) remains strictly bounded. Beyond a critical scale, this mismatch breaches the classical control light cone, triggering a superlinear geometric penalty ($\epsilon > 0$) that renders monolithic synchronization physically impossible. We formalize the resulting structural phase transition through a governing scaling law, $1+\epsilon > \gamma$, which mandates modular decomposition and a shift from global unitaries to Local Operations and Classical Communication (LOCC). To manage the resulting resource contention under strict coherence budgets, we introduce a layered semantic architecture and a time-aware Reserve--Commit protocol. By embedding predictive temporal pre-validation, the protocol acts as an architectural semantic classifier: it preemptively aborts transactions that exceed the causal horizon and explicitly converts scheduling-induced failures into location-known erasure metadata, directly relaxing hardware fidelity thresholds for downstream QEC decoders. Under near-term transduction targets ($\eta_{\mathrm{trans}} \sim 0.1$), we project a crossover scale at $N_c \sim 10^5$--$10^6$ physical qubits. This threshold marks a profound architectural convergence: the footprint required for modularity aligns precisely with early fault-tolerant utility, establishing time-aware distributed orchestration, rather than monolithic expansion or centralized classical control, as the physical imperative for utility-scale quantum computing.

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 paper claims that scalability in quantum computing is limited by a fundamental mismatch between macroscopic classical coordination latency τ_c (which grows with system diameter) and bounded microscopic quantum coherence τ_q, producing a superlinear geometric penalty ε > 0 that breaches the classical control light cone. This mismatch is formalized via a governing scaling law 1 + ε > γ, which is argued to mandate a transition from monolithic designs to modular decomposition using Local Operations and Classical Communication (LOCC). The authors introduce a layered semantic architecture and a time-aware Reserve-Commit protocol to handle resource contention under coherence constraints, and project a crossover scale N_c ~ 10^5--10^6 physical qubits under near-term transduction efficiency η_trans ~ 0.1, at which point modularity becomes the physical imperative for utility-scale quantum computing.

Significance. If the scaling law and its numerical consequences can be rigorously derived, the result would provide a physically motivated argument for prioritizing modular, distributed architectures over monolithic scaling in quantum hardware design. The alignment of the modularity threshold with early fault-tolerant regimes and the proposal to convert scheduling failures into erasure metadata for QEC decoders could influence both theoretical architecture studies and experimental roadmaps for large-scale systems.

major comments (2)
  1. [Abstract] Abstract: The governing scaling law 1+ε > γ is stated without any derivation, explicit functional form for ε (e.g., dependence on system diameter or τ_c growth rate), definition of γ, or step-by-step calculation showing how η_trans ~ 0.1 yields the specific numerical range N_c ~ 10^5--10^6. This renders the central crossover projection untraceable and load-bearing for the architectural claim.
  2. [Abstract] Abstract: The weakest assumption—that τ_c inevitably grows superlinearly with diameter while τ_q remains strictly bounded, producing ε > 0 that breaches the control light cone—is introduced axiomatically without supporting analysis, references to control-latency literature, or quantitative bounds, making the phase-transition argument circular by construction.
minor comments (1)
  1. [Abstract] Abstract: The symbols ε and γ are introduced without prior definition or relation to standard quantum information quantities, which reduces clarity for readers outside the immediate subfield.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments, which highlight important areas for clarification in our presentation of the scaling arguments. We have revised the abstract and added supporting details in the main text to make the derivations and assumptions fully traceable.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The governing scaling law 1+ε > γ is stated without any derivation, explicit functional form for ε (e.g., dependence on system diameter or τ_c growth rate), definition of γ, or step-by-step calculation showing how η_trans ~ 0.1 yields the specific numerical range N_c ~ 10^5--10^6. This renders the central crossover projection untraceable and load-bearing for the architectural claim.

    Authors: We agree that the abstract, as originally written, does not include the derivation steps. The full manuscript derives the scaling law in Section II by modeling τ_c as scaling superlinearly with diameter D (τ_c ∝ D^{1+δ} for δ>0 due to classical interconnect delays), leading to ε = (τ_c / τ_q) - 1, with γ defined as the critical threshold for light-cone breach (γ = 1 + ε_crit). The crossover N_c is calculated by setting the effective error rate from the penalty equal to the fault-tolerance threshold, incorporating η_trans in the communication cost, resulting in the range 10^5-10^6 for η_trans=0.1. We have revised the abstract to include a brief outline of this derivation and explicit definitions of ε and γ, with a pointer to the detailed calculation in the main text. revision: yes

  2. Referee: [Abstract] Abstract: The weakest assumption—that τ_c inevitably grows superlinearly with diameter while τ_q remains strictly bounded, producing ε > 0 that breaches the control light cone—is introduced axiomatically without supporting analysis, references to control-latency literature, or quantitative bounds, making the phase-transition argument circular by construction.

    Authors: The assumption is not introduced axiomatically but follows from standard models of classical control in large-scale systems. We have added references to literature on control latency in quantum hardware (e.g., works on cryogenic wiring and signal propagation delays showing superlinear scaling with system size) and included quantitative bounds: τ_q is limited to ~100 μs for superconducting qubits, while τ_c grows as O(D log D) or worse in 2D layouts. A new paragraph in the introduction provides this analysis to avoid any appearance of circularity, explicitly deriving the breach condition. revision: yes

Circularity Check

1 steps flagged

Crossover N_c projection reduces to insertion of assumed η_trans into scaling law 1+ε>γ whose parameters encode the mismatch by definition

specific steps
  1. fitted input called prediction [Abstract]
    "We formalize the resulting structural phase transition through a governing scaling law, 1+ε > γ, which mandates modular decomposition and a shift from global unitaries to Local Operations and Classical Communication (LOCC). ... Under near-term transduction targets (η_trans ~ 0.1), we project a crossover scale at N_c ~ 10^5--10^6 physical qubits."

    The law 1+ε>γ is defined to capture the superlinear penalty ε from the very latency mismatch whose breach is being predicted; substituting an external assumption η_trans~0.1 then yields the concrete N_c value, so the 'projection' is obtained by construction from the law's parameters rather than from an independent derivation of how ε(N) crosses the threshold.

full rationale

The abstract introduces the scaling law 1+ε>γ to formalize the structural phase transition arising from the τ_c vs τ_q mismatch, then immediately projects the specific numerical threshold N_c~10^5-10^6 by substituting the assumed near-term value η_trans~0.1. No explicit functional dependence of ε on system diameter, growth rate of τ_c, or derivation of the inequality from first-principles parameters is supplied in the provided text; the numerical claim therefore reduces to the assumed form of the law itself rather than an independent computation.

Axiom & Free-Parameter Ledger

4 free parameters · 3 axioms · 2 invented entities

The central claim depends on standard domain assumptions about signal propagation and coherence limits, plus ad-hoc parameters ε and γ in the scaling law and newly introduced architectural constructs without independent falsifiable evidence supplied in the abstract.

free parameters (4)
  • η_trans = ~0.1
    Near-term transduction efficiency target used to compute the crossover scale N_c.
  • N_c = ~10^5--10^6
    Projected crossover qubit count at which modularity becomes mandatory.
  • ε
    Superlinear geometric penalty factor introduced in the governing scaling law.
  • γ
    Threshold parameter in the inequality 1+ε > γ that triggers the phase transition.
axioms (3)
  • domain assumption Classical coordination latency τ_c grows with system diameter
    Invoked to establish the temporal mismatch with bounded quantum coherence.
  • domain assumption Quantum coherence time τ_q remains strictly bounded
    Standard physical assumption due to environmental decoherence.
  • ad hoc to paper The latency-coherence mismatch produces a superlinear penalty ε > 0 that breaches the classical control light cone
    Introduced to formalize the structural phase transition in the abstract.
invented entities (2)
  • Layered semantic architecture no independent evidence
    purpose: Manage resource contention under strict coherence budgets
    New architectural layer proposed to handle modular quantum operations.
  • Reserve-Commit protocol no independent evidence
    purpose: Time-aware scheduling that preemptively aborts transactions and converts failures into erasure metadata
    Novel protocol introduced to relax hardware fidelity requirements for QEC.

pith-pipeline@v0.9.0 · 5581 in / 2000 out tokens · 62531 ms · 2026-05-14T20:56:48.217868+00:00 · methodology

Review history (2 revisions) →

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

  1. [1]

    The Causal Locality Bound on Real-Time Quantum Control Building on this locality requirement, we can formalize a strict causal bound that explicitly rules out remote control paradigms. By enforcing τc < τ (p) q , we derive the maximum allowable physical distance Lmax ctrl between the quantum substrate and its classical causal backbone: Lmax ctrl ≤ c 2n ( ...

    work page

  2. [2]

    Query: Requesting the Entanglement Management Layer for a valid, pre-distributed resource tuple eij

    work page

  3. [3]

    Local Entanglement : Actuating local gates between the computational data qubits and the network interface qubits within each respective module

    work page

  4. [4]

    Measurement: Projectively measuring the interface qubits. 15

    work page

  5. [5]

    Coordination: Transmitting the classical heralds (measurement outcomes) across the distributed control plane

    work page

  6. [6]

    This composite sequence is extremely fragile

    F eedforward: Executing the conditional Pauli corrections on the target data qubits. This composite sequence is extremely fragile. If any step stalls—due to network congestion or classical processing delays—the participating quantum data qubits are forced to idle. In the quantum domain, idling implies decoherence. Therefore, the architecture requires a ri...

    work page

  7. [7]

    Arute, K

    F. Arute, K. Arya, R. Babbush, D. Bacon, J. C. Bardin, R. Barends, R. Biswas, S. Boixo, F. G. Brandao, D. A. Buell, et al. , nature 574, 505 (2019)

    work page 2019

  8. [8]

    Bluvstein, S

    D. Bluvstein, S. J. Evered, A. A. Geim, S. H. Li, H. Zhou, T. Manovitz, S. Ebadi, M. Cain, M. Kalinowski, D. Hangleiter, et al. , Nature 626, 58 (2024)

    work page 2024

  9. [9]

    Monroe and J

    C. Monroe and J. Kim, Science 339, 1164 (2013)

    work page 2013

  10. [10]

    L. M. Vandersypen, H. Bluhm, J. Clarke, A. Dzurak, R. Ishihara, A. Morello, D. Reilly, L. Schreiber, and M. Veldhorst, npj Quantum Information 3, 34 (2017)

    work page 2017

  11. [11]

    E. H. Lieb and D. W. Robinson, Communications in mathematical physics 28, 251 (1972)

    work page 1972

  12. [12]

    Cheneau, P

    M. Cheneau, P. Barmettler, D. Poletti, M. Endres, P. Schauß, T. Fukuhara, C. Gross, I. Bloch, C. Kollath, and S. Kuhr, Nature 481, 484 (2012)

    work page 2012

  13. [13]

    M. B. Hastings, Quantum Theory from Small to Large Scales 95, 171 (2010)

    work page 2010

  14. [14]

    Y. Kim, A. Eddins, S. Anand, K. X. Wei, E. Van Den Berg, S. Rosenblatt, H. Nayfeh, Y. Wu, M. Zaletel, K. Temme, et al. , Nature 618, 500 (2023)

    work page 2023

  15. [15]

    Awschalom, K

    D. Awschalom, K. K. Berggren, H. Bernien, S. Bhave, L. D. Carr, P. Davids, S. E. Economou, D. Englund, A. Faraon, M. Fejer, et al. , Prx Quantum 2, 017002 (2021)

    work page 2021

  16. [16]

    Krantz, M

    P. Krantz, M. Kjaergaard, F. Yan, T. P. Orlando, S. Gustavsson, and W. D. Oliver, Applied physics reviews 6 (2019)

    work page 2019

  17. [17]

    A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, Physical Review A—Atomic, Molecular, and Optical Physics 86, 032324 (2012)

    work page 2012

  18. [18]

    Acharya, I

    R. Acharya, I. Aleiner, R. M. Allen, T. I. Andersen, M. Ansmann, F. Arute, K. Arya, A. Asfaw, J. Atalaya, R. Babbush, et al. , Nature 614, 676 (2023)

    work page 2023

  19. [19]

    Dennis, A

    E. Dennis, A. Kitaev, A. Landahl, and J. Preskill, Journal of Mathematical Physics 43, 4452 (2002)

    work page 2002

  20. [20]

    Delfosse and N

    N. Delfosse and N. H. Nickerson, Quantum 5, 595 (2021)

    work page 2021

  21. [21]

    Litinski, Quantum 3, 128 (2019)

    D. Litinski, Quantum 3, 128 (2019)

    work page 2019

  22. [22]

    Bravyi and A

    S. Bravyi and A. Kitaev, Physical Review A—Atomic, Molecular, and Optical Physics 71, 022316 (2005)

    work page 2005

  23. [23]

    N. Lauk, N. Sinclair, S. Barzanjeh, J. P. Covey, M. Saffman, M. Spiropulu, and C. Simon, Quantum Science and Technology 5, 020501 (2020)

    work page 2020

  24. [24]

    Kurpiers, P

    P. Kurpiers, P. Magnard, T. Walter, B. Royer, M. Pechal, J. Heinsoo, Y. Salathé, A. Akin, S. Storz, J.-C. Besse, et al. , Nature 558, 264 (2018)

    work page 2018

  25. [25]

    Tomita and K

    Y. Tomita and K. M. Svore, Physical Review A 90, 062320 (2014). 21

    work page 2014

  26. [26]

    J. M. Baker, C. Duckering, A. Hoover, and F. T. Chong, in Proceedings of the 17th ACM International Conference on Computing Frontiers (2020) pp. 98–107

    work page 2020

  27. [27]

    Pompili, S

    M. Pompili, S. L. Hermans, S. Baier, H. K. Beukers, P. C. Humphreys, R. N. Schouten, R. F. Vermeulen, M. J. Tiggelman, L. dos Santos Martins, B. Dirkse, et al. , Science 372, 259 (2021)

    work page 2021

  28. [28]

    Panteleev and G

    P. Panteleev and G. Kalachev, in Proceedings of the 54th annual ACM SIGACT symposium on theory of computing (2022) pp. 375–388

    work page 2022

  29. [29]

    Bravyi, A

    S. Bravyi, A. W. Cross, J. M. Gambetta, D. Maslov, P. Rall, and T. J. Yoder, Nature 627, 778 (2024)

    work page 2024

  30. [30]

    Bartolucci, P

    S. Bartolucci, P. Birchall, H. Bombin, H. Cable, C. Dawson, M. Gimeno-Segovia, E. Johnston, K. Kieling, N. Nickerson, M. Pant, et al. , Nature Communications 14, 912 (2023)

    work page 2023

  31. [31]

    Wehner, D

    S. Wehner, D. Elkouss, and R. Hanson, Science 362, eaam9288 (2018)

    work page 2018

  32. [32]

    M. Pant, H. Krovi, D. Towsley, L. Tassiulas, L. Jiang, P. Basu, D. Englund, and S. Guha, npj Quantum Information 5, 25 (2019)

    work page 2019

  33. [33]

    Lamport, in Operating Systems, An Advanced Course (1978) pp

    L. Lamport, in Operating Systems, An Advanced Course (1978) pp. 558–565

    work page 1978

  34. [34]

    X. Fu, M. Rol, C. C. Bultink, J. Van Someren, N. Khammassi, I. Ashraf, R. Vermeulen, J. De Sterke, W. Vlothuizen, R. Schouten, et al. , IEEE Micro 38, 40 (2018)

    work page 2018

  35. [35]

    Patra, J

    B. Patra, J. P. Van Dijk, S. Subramanian, A. Corna, X. Xue, C. Jeon, F. Sheikh, E. Juarez-Hernandez, B. P. Esparza, H. Rampurawala, et al. , in 2020 IEEE International Solid-State Circuits Conference-(ISSCC) (IEEE, 2020) pp. 304–306

    work page 2020

  36. [36]

    X. Xue, B. Patra, J. P. van Dijk, N. Samkharadze, S. Subramanian, A. Corna, B. Paquelet Wuetz, C. Jeon, F. Sheikh, E. Juarez-Hernandez, et al. , Nature 593, 205 (2021)

    work page 2021

  37. [37]

    Cross, A

    A. Cross, A. Javadi-Abhari, T. Alexander, N. De Beaudrap, L. S. Bishop, S. Heidel, C. A. Ryan, P. Sivarajah, J. Smolin, J. M. Gambetta, et al. , ACM Transactions on Quantum Computing 3, 1 (2022)

    work page 2022

  38. [38]

    J. N. Gray, in Operating systems: An advanced course (Springer, 2005) pp. 393–481

    work page 2005

  39. [39]

    M. Y. Siraichi, V. F. dos Santos, S. Collange, and F. M. Q. de M. Pereira, in Proceedings of the 2018 International Symposium on Code Generation and Optimization (ACM, 2018) pp. 113–125

    work page 2018

  40. [40]

    Murali, N

    P. Murali, N. M. Linke, M. Martonosi, A. J. Abhari, N. H. Nguyen, and C. H. Alderete, in 2019 ACM/IEEE 46th Annual International Symposium on Computer Architecture (ISCA) (IEEE, 2019) pp. 527–540

    work page 2019

  41. [41]

    D. K. Tuckett, S. D. Bartlett, and S. T. Flammia, Physical review letters 120, 050505 (2018)

    work page 2018

  42. [42]

    S. Ma, G. Liu, P. Peng, B. Zhang, S. Jandura, J. Claes, A. P. Burgers, G. Pupillo, S. Puri, and J. D. Thompson, Nature 622, 279 (2023)

    work page 2023

  43. [43]

    Y. Wu, S. Kolkowitz, S. Puri, and J. D. Thompson, Nature communications 13, 4657 (2022)

    work page 2022

  44. [44]

    Baranes, M

    G. Baranes, M. Cain, J. P. B. Ataides, D. Bluvstein, J. Sinclair, V. Vuletić, H. Zhou, and M. D. Lukin, Physical Review X 16, 011002 (2026)

    work page 2026

  45. [45]

    G. Q. AI et al. , Nature 638, 920 (2025)

    work page 2025

  46. [46]

    Salathé, P

    Y. Salathé, P. Kurpiers, T. Karg, C. Lang, C. K. Andersen, A. Akin, S. Krinner, C. Eichler, and A. Wallraff, Physical Review Applied 9, 034011 (2018)

    work page 2018

  47. [47]

    D. Xu, A. B. Özgüler, G. Di Guglielmo, N. Tran, G. N. Perdue, L. Carloni, and F. Fahim, in 2022 IEEE/ACM Third International Workshop on Quantum Computing Software (QCS) (IEEE, 2022) pp. 43–49

    work page 2022

  48. [48]

    R. D. Oliveira, S. Bahrani, E. Arabul, R. Wang, R. Nejabati, and D. Simeonidou, in IEEE INFOCOM 2023-IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) (IEEE, 2023) pp. 1–6

    work page 2023

  49. [49]

    Liyanage, Y

    N. Liyanage, Y. Wu, E. Houghton, and L. Zhong, in 2024 IEEE International Conference on Quantum Computing and Engineering (QCE) , Vol. 2 (IEEE, 2024) pp. 622–623

    work page 2024