arxiv: 2605.04316 · v1 · submitted 2026-05-05 · 💻 cs.DC

Recognition: unknown

Orchestrating Serverless Applications in the Edge Cloud Space Continuum: What Breaks and What is Next?

Hadi Tabatabaee Malazi , Reza Farahani , Nitinder Mohan , Schahram Dustdar

Authors on Pith no claims yet

Pith reviewed 2026-05-08 17:06 UTC · model grok-4.3

classification 💻 cs.DC

keywords serverless orchestrationedge cloud continuumLEO satellitesdynamic graphsfunction placementdecentralized stateconstraint-aware scaling

0 comments

The pith

Serverless orchestration breaks in the edge cloud space continuum because LEO systems create time-varying contact graphs and strict resource constraints.

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

The paper argues that serverless computing works well in stable edge and cloud settings but cannot be applied directly when the infrastructure expands to include Low Earth Orbit satellites. It identifies ten assumptions that fail under these conditions and groups them into three challenges around executing functions over changing graphs, placing and scaling them under tight limits, and keeping correct results when state updates arrive late and from many places. If the claim holds, developers could run demand-driven functions across orbits, ground edges, and clouds for tasks that need global reach without constant connectivity. Readers would care because this extension opens serverless to applications that must operate in remote, mobile, or disaster-prone environments where traditional orchestration collapses.

Core claim

The paper establishes that ten assumptions underlying current serverless orchestration no longer hold in the edge cloud space continuum. LEO constellations introduce time-varying contact graphs, intermittent links, and hard limits on energy, memory, communication, and cost. These broken assumptions fall into three challenges: spatiotemporal execution over dynamic graphs, constraint-aware function placement and scaling, and correctness and progress under decentralized and delayed state. The authors propose an architecture that addresses the challenges and illustrate it with a flood-response use case that spans satellites, edges, and clouds.

What carries the argument

The ten broken assumptions, grouped into the three core challenges of spatiotemporal execution over dynamic graphs, constraint-aware function placement and scaling, and correctness under decentralized delayed state.

Load-bearing premise

The proposed architecture will deliver robust and efficient serverless execution across the continuum when implemented beyond the single illustrative use case.

What would settle it

A simulation of LEO contact schedules and resource limits that runs the proposed architecture on the flood-response workflow and shows whether functions complete correctly without violating feasibility constraints.

read the original abstract

Serverless computing has matured into an effective execution model for edge cloud environments, enabling function level decomposition, demand driven scaling, and workflow execution across stable, well provisioned infrastructure. This success motivates extending it to the edge cloud space continuum, where Low Earth Orbit (LEO) constellations are increasingly explored as distributed compute substrates. However, existing serverless orchestration is not directly applicable in this setting, where LEO systems impose time varying contact graphs, intermittent link availability, and strict feasibility constraints on energy, memory, communication, and operational cost. This article identifies ten broken assumptions in existing serverless orchestration and organizes them into three core challenges: spatiotemporal execution over dynamic graphs, constraint aware function placement and scaling, and correctness and progress under decentralized and delayed state. It then proposes an architecture that enables robust and efficient serverless execution across the continuum, grounded in these challenges and demonstrated through a representative flood response use case.

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.

Desk Editor's Note private letter to a colleague

The paper usefully identifies why serverless breaks in LEO constellations but the proposed architecture lacks concrete mechanisms or evaluation.

read the letter

Serverless orchestration as we know it won't transfer cleanly to LEO satellite setups. The paper spells out ten broken assumptions, groups them into three challenges around dynamic execution graphs, constraint-based placement, and delayed state management, and then outlines an architecture to handle the space continuum. The new part is the specific list of assumptions tailored to LEO properties such as intermittent links and energy limits. This organizes scattered issues from prior edge and satellite computing papers into one place. The flood response use case makes the problems concrete without overclaiming. The architecture stays conceptual. It gets described at a high level with no algorithms for function placement, no protocols for state consistency across delays, and no numbers on latency or energy use. Without those, we can't tell if it actually repairs the broken assumptions or just restates the problems. This paper fits readers working on edge computing extensions or emerging space-based systems. It gives a clear map of the issues even if it doesn't solve them. The thinking is straightforward and engages the literature directly. I would send it for peer review. The challenges section alone could help shape follow-up work, and referees might push for more specifics on the architecture.

Referee Report

2 major / 1 minor

Summary. The paper claims that serverless orchestration assumptions from stable edge clouds do not hold in the LEO-based edge cloud space continuum due to time-varying contact graphs, intermittent connectivity, and tight constraints on energy, memory, communication, and cost. It identifies ten broken assumptions organized into three core challenges (spatiotemporal execution over dynamic graphs, constraint-aware placement/scaling, and correctness/progress under decentralized/delayed state), proposes a high-level architecture to address them, and demonstrates the approach via an illustrative flood-response use case.

Significance. If validated, the work could help define research directions for serverless computing in dynamic orbital environments by systematically exposing mismatches with terrestrial assumptions and sketching a continuum-wide execution model. The conceptual framing and use-case grounding provide a useful starting point for subsequent algorithm and protocol development, though the absence of mechanisms or metrics currently limits its immediate engineering impact.

major comments (2)

[architecture proposal section] The section proposing the architecture (following the three challenges): the architecture is described only at a conceptual level with no component specifications, placement algorithms, state-synchronization protocols, or decision procedures for handling dynamic contact graphs and feasibility constraints; without these, it is not possible to determine whether the design actually repairs the ten broken assumptions.
[use-case demonstration section] The flood-response use-case section: the scenario is presented purely as an illustration with no quantitative results (latency, energy, success rate, overhead), no baseline comparisons, and no explicit mapping showing how each of the three challenges or ten assumptions is resolved; this leaves the central claim that the architecture 'enables robust and efficient serverless execution' unsupported by evidence.

minor comments (1)

The paper would benefit from explicit section numbering and a table summarizing the ten broken assumptions with their corresponding challenges and proposed mitigations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We are grateful to the referee for the constructive feedback. Our manuscript is a vision paper whose primary contribution is to identify ten broken assumptions in serverless orchestration when moving from stable edge clouds to the LEO-based edge-cloud-space continuum and to sketch a high-level architecture that could address the resulting challenges. We respond to each major comment below.

read point-by-point responses

Referee: [architecture proposal section] The section proposing the architecture (following the three challenges): the architecture is described only at a conceptual level with no component specifications, placement algorithms, state-synchronization protocols, or decision procedures for handling dynamic contact graphs and feasibility constraints; without these, it is not possible to determine whether the design actually repairs the ten broken assumptions.

Authors: We agree that the architecture is presented at a conceptual level. The paper's goal is to systematically expose mismatches between existing serverless assumptions and the requirements imposed by time-varying contact graphs, intermittent connectivity, and tight resource constraints in the continuum, then to outline the necessary architectural elements rather than to deliver a fully specified system. Full component interfaces, placement algorithms, and synchronization protocols would constitute a separate systems contribution. In revision we will expand the architecture section with additional detail on component responsibilities, high-level decision flows for contact-graph handling and constraint checking, and explicit traceability from each element back to the ten broken assumptions. revision: partial
Referee: [use-case demonstration section] The flood-response use-case section: the scenario is presented purely as an illustration with no quantitative results (latency, energy, success rate, overhead), no baseline comparisons, and no explicit mapping showing how each of the three challenges or ten assumptions is resolved; this leaves the central claim that the architecture 'enables robust and efficient serverless execution' unsupported by evidence.

Authors: The flood-response scenario is intended to ground the three challenges in a concrete, realistic setting rather than to serve as an empirical evaluation. We acknowledge that the section currently lacks quantitative metrics, baseline comparisons, and an explicit mapping. In the revised manuscript we will add a dedicated subsection that maps each challenge and each of the ten broken assumptions to specific elements of the use case. We will also revise the surrounding text to clarify that the architecture is proposed as a direction that can enable robust execution, with concrete performance validation left to future implementation and simulation studies. revision: partial

Circularity Check

0 steps flagged

No significant circularity in conceptual challenge identification and high-level architecture proposal

full rationale

The paper identifies ten broken assumptions from stated properties of LEO systems (time-varying contact graphs, intermittent links, energy/memory/communication constraints) and limitations of existing serverless orchestration, organizes them into three challenges, and proposes a high-level architecture illustrated by an example use case. No mathematical derivations, equations, fitted parameters, or predictions appear in the provided text. Claims rest on external domain knowledge rather than self-referential definitions, self-citation chains, or renamings that reduce to inputs by construction. The use case is explicitly illustrative, not a quantitative or fitted validation.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The central claim rests on domain assumptions about LEO system properties and current serverless limitations, with the proposed architecture introduced as the main new element without independent empirical support.

axioms (2)

domain assumption Serverless computing has matured into an effective execution model for edge cloud environments with function level decomposition and demand driven scaling.
Stated as background motivation in the abstract.
domain assumption LEO constellations impose time varying contact graphs, intermittent link availability, and strict feasibility constraints on energy, memory, communication, and operational cost.
Core premise used to explain why existing orchestration breaks.

invented entities (1)

Proposed architecture for serverless execution across the edge cloud space continuum no independent evidence
purpose: To enable robust and efficient execution by addressing the three core challenges
The architecture is introduced as the solution but lacks implementation or independent verification beyond the use case.

pith-pipeline@v0.9.0 · 5469 in / 1413 out tokens · 22808 ms · 2026-05-08T17:06:45.805256+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

20 extracted references

[1]

Satellite Computing: Vision and Challenges,

S. Wang and Q. Li, “Satellite Computing: Vision and Challenges,”IEEE Internet of Things Journal, 2023

2023
[2]

Orbital Edge Computing: Nanosatellite Constellations as a New Class of Computer System,

B. Denby and B. Lucia, “Orbital Edge Computing: Nanosatellite Constellations as a New Class of Computer System,” inInternational Conference on Architectural Support for Programming Languages and Operating Systems. Association for Computing Machinery, 2020

2020
[3]

Krios: Scheduling Abstractions and Mechanisms for Enabling a LEO Compute Cloud,

V. Bhosale, A. Gavrilovska, and K. Bhardwaj, “Krios: Scheduling Abstractions and Mechanisms for Enabling a LEO Compute Cloud,” inACM Symposium on Cloud Computing. Association for Computing Machinery, 2024

2024
[4]

LEOCraft: Towards designing performant LEO networks,

S. Basak, A. Pal, and D. Bhattacherjee, “LEOCraft: Towards designing performant LEO networks,” in Annual Technical Conference, ser. USENIX ATC ’25. USENIX Association, 2025

2025
[5]

HyperDrive: Scheduling serverless functions in the edge- cloud-space 3D continuum,

T. Pusztai, C. Marcelino, and S. Nastic, “HyperDrive: Scheduling serverless functions in the edge- cloud-space 3D continuum,” in2024 IEEE/ACM Symposium on Edge Computing. IEEE, 2024

2024
[6]

A Survey on Satellite Computing: Connecting the Dots Between Networks and Applications,

C. Guimarães, A. Netti, M. Sauer, F . Zeiger, H.- P . Huth, and E. Boriskova, “A Survey on Satellite Computing: Connecting the Dots Between Networks and Applications,”IEEE Communications Surveys & Tutorials, 2026

2026
[7]

SaTE: Low-latency traffic engineering for satellite networks,

H. Wu, Y . Han, M. Rajpal, Q. Zhang, and J. Wang, “SaTE: Low-latency traffic engineering for satellite networks,” inACM SIGCOMM 2025 Conference. Association for Computing Machinery, 2025. 10 Orchestrating Serverless Applications in the Edge-Cloud-Space Continuum May 2026 IEEE Internet Computing

2025
[8]

Komet: A serverless platform for low-earth orbit edge services,

T. Pfandzelter and D. Bermbach, “Komet: A serverless platform for low-earth orbit edge services,” in2024 ACM Symposium on Cloud Computing. Association for Computing Machinery, 2024

2024
[9]

Deciphering the Enigma of Satellite Computing with COTS Devices: Measurement and Analysis,

R. Xing, M. Xu, A. Zhou, Q. Li, Y . Zhang, F . Qian, and S. Wang, “Deciphering the Enigma of Satellite Computing with COTS Devices: Measurement and Analysis,” inAnnual International Conference on Mobile Computing and Networking. Association for Computing Machinery, 2024

2024
[10]

Gaia: Hybrid Hardware Acceleration for Serverless AI in the 3D Compute Continuum,

M. Reisecker, C. Marcelino, T. Pusztai, and S. Nastic, “Gaia: Hybrid Hardware Acceleration for Serverless AI in the 3D Compute Continuum,” inIEEE/ACM 12th International Conference on Big Data Computing, Applications and Technologies. Association for Computing Machinery, 2025

2025
[11]

Databelt: A continuous data path for serverless workflows in the 3D compute continuum,

C. Marcelino, L. Guelmino, T. Pusztai, and S. Nastic, “Databelt: A continuous data path for serverless workflows in the 3D compute continuum,”Journal of Systems Architecture, 2025

2025
[12]

The Serverless Computing Survey: A Technical Primer for Design Architecture,

Z. Li, L. Guo, J. Cheng, Q. Chen, B. He, and M. Guo, “The Serverless Computing Survey: A Technical Primer for Design Architecture,”ACM Computing Surveys, 2022

2022
[13]

Space Computing: Architectures, Challenges, and Future Directions,

E. Xue, Z. Zhang, J. Xue, H. Wang, I. E. Carvajal- Roca, Z. He, H. Zhang, H. Wang, Z. Wan, and C. Li, “Space Computing: Architectures, Challenges, and Future Directions,”Intelligent Computing, 2025

2025
[14]

ComEdge: Cloud-native platform for integrated computing and communication in satellite–terrestrial network,

H. Shi, X. Zhang, P . Wu, J. Chen, and Y . Zhang, “ComEdge: Cloud-native platform for integrated computing and communication in satellite–terrestrial network,”Electronics, 2023

2023
[15]

Space Computing Constellation: System Architecture, Implementations and Challenges,

H. Wang, K. Y ao, L. Gong, Y . Jin, P . Zhang, Y . Liu, Y . Yuan, J. Xue, Z. Wan, Y . Liu, X. Li, C. Li, and Z. Zhao, “Space Computing Constellation: System Architecture, Implementations and Challenges,”IEEE Internet of Things Journal, 2026

2026
[16]

Utility optimization for computation offloading and splitting in time-varying hap and LEO satellite integrated mec networks,

R. Zhang, J. Zhang, X. Wang, and W. Shi, “Utility optimization for computation offloading and splitting in time-varying hap and LEO satellite integrated mec networks,”Computer Networks, 2024

2024
[17]

Emulating space computing networks with RHONE,

L. Wang, Q. Liu, Q. Li, S. Wang, X. Liu, and C. Xu, “Emulating space computing networks with RHONE,” inAnnual International Conference on Mobile Computing and Networking, 2025

2025
[18]

Resource Allocation on Low-Earth Orbit Edge Infrastructure: Taxonomy, Survey, and Research Challenges,

F . D. Rossi, P . S. S. d. Souza, and M. C. Luizelli, “Resource Allocation on Low-Earth Orbit Edge Infrastructure: Taxonomy, Survey, and Research Challenges,”IEEE Access, 2025

2025
[19]

A Comprehensive Survey of Orbital Edge Computing: Systems, Applications, and Algorithms,

Z. Yin, C. Wu, C. Guo, Y . Li, M. Xu, W. Gao, and C. Chi, “A Comprehensive Survey of Orbital Edge Computing: Systems, Applications, and Algorithms,” Chinese Journal of Aeronautics, 2025

2025
[20]

Device-Satellite- Satellite Collaborative Task Offloading Computing and Resource Allocation in 6G Satellite-Ground Edge Computing Network,

S. Liu, Z. Zhang, and S. Zeadally, “Device-Satellite- Satellite Collaborative Task Offloading Computing and Resource Allocation in 6G Satellite-Ground Edge Computing Network,”Computer Networks, 2026. HADI TABATABAEE MALAZIis an Assistant Professor in the School of Computer Science at University College Dublin, Ireland. He leads the Sustainable Orchestrati...

2026