pith. sign in

arxiv: 2605.15335 · v1 · pith:72DMPU7Anew · submitted 2026-05-14 · 💻 cs.DC · cs.SY· eess.SY

Designing Dense Satellite Clusters for Distributed Space-based Datacenters

Jules P\'enot , Hamsa Balakrishnan This is my paper

Pith reviewed 2026-05-20 19:54 UTC · model grok-4.3

classification 💻 cs.DC cs.SYeess.SY
keywords satellite datacentersLEO orbital clusters3D satellite packinginter-satellite linksspace-based computingClos network mapping
0
0 comments X

The pith

3D orbital designs let satellite datacenters scale with the cube of their radius-to-spacing ratio

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

The paper proposes orbital designs for clusters of satellites acting as distributed datacenters in low Earth orbit. It presents a planar layout and a three-dimensional layout that satisfy constraints on spacing, solar power access, and communication links. The 3D design allows the number of satellites to grow with the cube of the size ratio, enabling larger scale than flat arrangements. This matters because it could support more computing power in space where power and cooling are available from the sun and vacuum.

Core claim

The proposed 3D architecture allows for the number of datacenter satellites to scale proportional to (R_max/R_min)^3, an improvement over all previous LEO datacenter cluster designs. Both the planar and 3D cluster orbital designs remain consistent with the inter-satellite spacing, unobstructed sun-vector, and inter-satellite line of sight constraints throughout the cluster's entire orbit, as shown by construction and numerical analysis. For a given satellite cluster, an integer optimization problem maps a VL2-like Clos network datacenter switching fabric onto the satellites and their corresponding set of feasible ISLs, confirming that there are sufficiently many permanently unobstructed ISLs

What carries the argument

the 3D cluster orbital design parametrized by minimum inter-satellite spacing R_min and cluster radius R_max that maintains all constraints by construction

Load-bearing premise

The proposed planar and 3D cluster orbital designs remain consistent with the inter-satellite spacing, unobstructed sun-vector, and inter-satellite line of sight constraints throughout the cluster's entire orbit

What would settle it

A simulation or observation showing any satellite in the proposed cluster losing unobstructed sun exposure or line of sight to its neighbors at any point during the full orbit would disprove the designs

read the original abstract

Recent proposals for datacenters in sun-synchronous Low Earth Orbit rely on a large number of compute satellites formation-flying in dense clusters. Designing such satellite clusters requires optimizing the satellites' orbital geometry under several safety and operational constraints applied throughout the cluster's entire orbit. These constraints include guaranteeing a minimum inter-satellite spacing, obstruction-less solar power for every satellite, and that each satellite have a stable set of nearest neighbors with which it can maintain inter-satellite links (ISLs). In this work, we propose two main cluster orbital designs, parametrized by the minimum inter-satellite spacing $R_{min}$ and the cluster radius $R_{max}$: a planar cluster, and a 3D cluster. We show by construction and numerical analysis that both cluster orbital designs are consistent with the inter-satellite spacing, unobstructed sun-vector, and inter-satellite line of sight constraints. The proposed planar architecture is the most efficient packing of satellites in a plane for given $R_{min}$ and $R_{max}$ values, and our 3D architecture allows for the number of datacenter satellites to scale proportional to $(R_{max}/R_{min})^3$, an improvement over all previous LEO datacenter cluster designs. Finally, for a given satellite cluster, we formulate and solve an integer optimization problem that maps a VL2-like Clos network datacenter switching fabric onto the satellites and their corresponding set of feasible ISLs. We confirm that for both the planar and 3D architectures, there are sufficiently many permanently unobstructed ISLs within the cluster to replicate the switching fabric of terrestrial datacenters. We also examine the tradeoff between the number of ISLs each satellite can simultaneously sustain, and the corresponding number of cluster satellites that must be dedicated as aggregation and intermediate switches.

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

1 major / 2 minor

Summary. The manuscript proposes two orbital designs for dense LEO satellite clusters for space-based datacenters, parametrized by minimum inter-satellite spacing R_min and cluster radius R_max: a planar design and a 3D volumetric design. It asserts that both satisfy inter-satellite spacing, unobstructed sun-vector, and stable ISL line-of-sight constraints by construction and numerical analysis over the orbit. The 3D design is claimed to enable satellite count scaling proportional to (R_max/R_min)^3, an improvement over prior LEO designs. An integer program is formulated and solved to embed a VL2-like Clos network onto the satellites and their feasible ISLs, confirming sufficient permanent links and exploring ISL capacity tradeoffs.

Significance. If the continuous orbit-wide constraint satisfaction is established, the work would advance scalable space datacenters by providing explicit geometric constructions with cubic scaling in 3D and a concrete network mapping that replicates terrestrial fabrics. The parameter-free geometric basis and integer-program embedding are strengths that support reproducibility.

major comments (1)
  1. [3D cluster orbital design and numerical verification] The central scaling claim in the 3D architecture section—that satellite number scales as (R_max/R_min)^3 while maintaining all constraints throughout the orbit—rests on 'construction and numerical analysis.' The analysis samples discrete epochs, but with shared semi-major axis and mean motion, differential precession and eccentricity vectors induce continuous shearing of the formation. It is unclear whether sampling covers all true anomalies, particularly near orbital nodes or sun-synchronous terminator crossings, where minimum distances or LOS could be violated; this gap directly undermines sustained validity of the volumetric scaling.
minor comments (2)
  1. [Abstract] Abstract states numerical verification but omits error bars, exact parameter values for R_min/R_max, and the number/density of sampled epochs used in the orbit coverage.
  2. [Orbital design sections] Notation for orbital elements (inclination, eccentricity vectors, argument of perigee) should be defined explicitly with the values employed in the numerical checks.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful review and for identifying a key point regarding verification of continuous constraint satisfaction in the 3D cluster design. We address this comment directly below and will strengthen the manuscript accordingly.

read point-by-point responses

  1. Referee: The central scaling claim in the 3D architecture section—that satellite number scales as (R_max/R_min)^3 while maintaining all constraints throughout the orbit—rests on 'construction and numerical analysis.' The analysis samples discrete epochs, but with shared semi-major axis and mean motion, differential precession and eccentricity vectors induce continuous shearing of the formation. It is unclear whether sampling covers all true anomalies, particularly near orbital nodes or sun-synchronous terminator crossings, where minimum distances or LOS could be violated; this gap directly undermines sustained validity of the volumetric scaling.

    Authors: We thank the referee for this observation. The 3D volumetric design is constructed from relative orbital elements that enforce minimum inter-satellite spacing R_min and unobstructed line-of-sight by geometry, independent of specific true anomaly; the numerical sampling was intended only as corroboration at representative epochs. We agree, however, that discrete sampling alone leaves open the possibility of transient violations due to differential precession or at critical points such as nodes and terminator crossings. In the revised manuscript we will replace the existing numerical section with a denser, orbit-wide verification that explicitly samples at 1-degree true-anomaly intervals and includes targeted checks at orbital nodes and sun-synchronous terminator crossings. This will either confirm or bound any residual violations, thereby rigorously supporting the claimed cubic scaling throughout the full orbit. revision: yes

Circularity Check

0 steps flagged

No significant circularity; claims rest on explicit design construction plus independent numerical checks

full rationale

The paper parametrizes both planar and 3D cluster designs directly by the geometric inputs R_min and R_max, then states that the resulting formations satisfy the spacing, sun-vector, and ISL constraints 'by construction and numerical analysis.' The reported cubic scaling N ∝ (R_max/R_min)^3 is the direct volumetric consequence of placing satellites at minimum spacing R_min inside a sphere of radius R_max; it is presented as a property of the proposed architecture rather than a prediction derived from fitted data or prior results. No self-citations, uniqueness theorems, or ansatzes are invoked to justify the central claims. The numerical analysis is described as an independent verification step, keeping the derivation self-contained against external orbital constraints.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claims rest on two tunable geometric parameters and the assumption that sun-synchronous LEO formation flying can be maintained under the stated constraints.

free parameters (2)
  • R_min
    Minimum inter-satellite spacing used to parametrize both cluster designs.
  • R_max
    Cluster radius used to parametrize both cluster designs and to derive the cubic scaling.
axioms (1)
  • domain assumption Satellites can maintain stable formation flying in sun-synchronous LEO while satisfying continuous solar illumination and line-of-sight constraints.
    Invoked when stating that the designs are consistent with the constraints throughout the orbit.

pith-pipeline@v0.9.0 · 5866 in / 1212 out tokens · 35920 ms · 2026-05-20T19:54:06.010776+00:00 · methodology

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

15 extracted references · 15 canonical work pages

  1. [1]

    Why we should train AI in space,

    A. Oltean and P. Johnston, “Why we should train AI in space,” tech. rep., Starcloud, 2024

    work page 2024

  2. [2]

    Space solar power generation: A viable system proposal and technoeconomic analysis,

    O. S. Mizrahi, P. Jahelka, E. Gdoutos, J. Brunet, A. Ayling, A. Fikes, A. Wu, R. Madonna, H. A. Atwater, S. Pellegrino, and A. Hajimiri, “Space solar power generation: A viable system proposal and technoeconomic analysis,”Joule, V ol. 9, June 2025, 10.1016/j.joule.2025.101928

    work page doi:10.1016/j.joule.2025.101928 2025

  3. [3]

    Big Tech Dreams of Putting Data Centers in Space,

    S. Hurwitz, “Big Tech Dreams of Putting Data Centers in Space,”Wall Street Journal, 2025

    work page 2025

  4. [4]

    Towards a future space-based, highly scalable AI infrastructure system design,

    B. A. y. Arcas, T. Beals, M. Biggs, J. V . Bloom, T. Fischbacher, K. Gromov, U. K ¨oster, R. Pravahan, and J. Manyika, “Towards a future space-based, highly scalable AI infrastructure system design,” Nov

    work page

  5. [5]

    arXiv:2511.19468 [cs], 10.48550/arXiv.2511.19468

    work page doi:10.48550/arxiv.2511.19468

  6. [6]

    Bezos and Musk Race to Bring Data Centers to Space,

    M. Maidenberg and B. Peterson, “Bezos and Musk Race to Bring Data Centers to Space,”Wall Street Journal, 2025

    work page 2025

  7. [7]

    A first look at starlink performance,

    F. Michel, M. Trevisan, D. Giordano, and O. Bonaventure, “A first look at starlink performance,”Pro- ceedings of the 22nd ACM Internet Measurement Conference, IMC ’22, New York, NY , USA, Associ- ation for Computing Machinery, Oct. 2022, pp. 130–136, 10.1145/3517745.3561416. 18

    work page doi:10.1145/3517745.3561416 2022

  8. [8]

    Orbital Optimization of a Distributed Heliocentric Relay Network for Mars-Earth Communications,

    J. P ´enot and H. Balakrishnan, “Orbital Optimization of a Distributed Heliocentric Relay Network for Mars-Earth Communications,”2026 IEEE Aerospace Conference, Big Sky, MT, Mar. 2026. Forthcom- ing in proceedings

    work page 2026

  9. [9]

    D’Amico,Autonomous formation flying in low earth orbit

    S. D’Amico,Autonomous formation flying in low earth orbit. Ridderkerk: Ridderprint, 2010. OCLC: 840430463

    work page 2010

  10. [10]

    Vallado,Fundamentals Of Astrodynamics And Applications

    D. Vallado,Fundamentals Of Astrodynamics And Applications. Microcosm Press, 4th ed., 2013

    work page 2013

  11. [11]

    Heliocentric Mutually Orbiting Groups: Parametrization, Phasing Ma- neuvers, and Orbital Insertion,

    J. P ´enot and H. Balakrishnan, “Heliocentric Mutually Orbiting Groups: Parametrization, Phasing Ma- neuvers, and Orbital Insertion,”2025 AAS/AIAA Astrodynamics Specialist Conference, Boston, MA, AAS, Aug. 2025. Forthcoming in proceedings

    work page 2025

  12. [12]

    Trajectory Design for the ESA LISA Mission,

    W. Martens and E. Joffre, “Trajectory Design for the ESA LISA Mission,”The Journal of the Astro- nautical Sciences, V ol. 68, June 2021, pp. 402–443. arXiv:2101.03040 [gr-qc], 10.1007/s40295-021- 00263-2

    work page doi:10.1007/s40295-021- 2021

  13. [13]

    Gr ¨unbaum and G

    B. Gr ¨unbaum and G. C. G. C. Shephard,Tilings and patterns. New York : W.H. Freeman, 1987

    work page 1987

  14. [14]

    VL2: a scalable and flexible data center network,

    A. Greenberg, J. R. Hamilton, N. Jain, S. Kandula, C. Kim, P. Lahiri, D. A. Maltz, P. Patel, and S. Sen- gupta, “VL2: a scalable and flexible data center network,”SIGCOMM Comput. Commun. Rev., V ol. 39, Aug. 2009, pp. 51–62, 10.1145/1594977.1592576

    work page doi:10.1145/1594977.1592576 2009

  15. [15]

    Gurobi Optimizer Reference Manual,

    Gurobi Optimization, LLC, “Gurobi Optimizer Reference Manual,” 2024. 19

    work page 2024