pith. sign in

arxiv: 2604.12642 · v1 · submitted 2026-04-14 · 💻 cs.SE

Pricing-Driven Resource Allocation in the Computing Continuum

Alejandro Garc\'ia-Fern\'andez , Boris Sedlak , Jos\'e Antonio Parejo , Pantelis Frangoudis , Antonio Ruiz-Cort\'es , Schahram Dustdar This is my paper

Pith reviewed 2026-05-10 15:15 UTC · model grok-4.3

classification 💻 cs.SE
keywords resource allocationcomputing continuumpricing structuresconfiguration spacesheterogeneous infrastructurecost optimizationdeployment constraints
0
0 comments X

The pith

Pricings can represent the configuration spaces of heterogeneous infrastructure to solve resource allocation in the computing continuum.

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

The paper claims that the combinatorial problem of choosing nodes for application deployment across distributed and varied computing resources can be reframed using pricing structures. Instead of building custom optimization models for topologies and demands, the approach treats infrastructure options and constraints as plans and add-ons that implicitly define what configurations are possible. A reader would care because existing methods do not generalize well as scale and heterogeneity increase, while pricing representations could reuse analysis tools to find cost-optimal solutions that still satisfy performance and location rules. The work supplies a direct formulation of the allocation task in pricing terms, a workflow to compute solutions, methods to create synthetic test cases, and a collection of 9,600 precomputed scenarios.

Core claim

Pricings, structures whose plans and add-ons implicitly define the configuration space of possible subscriptions, can serve as general-purpose representations for the resource allocation problem in the computing continuum, allowing infrastructure nodes and their functional and non-functional constraints to be expressed through pricing options so that cost-optimal deployments can be computed by exploring those spaces.

What carries the argument

Pricings, consisting of plans and add-ons that implicitly define configuration spaces of infrastructure selections and constraints.

If this is right

  • Resource allocation can be formulated directly as selection within a pricing structure rather than as an ad-hoc constrained optimization problem.
  • Existing pricing analysis engines can be used to explore the space and return cost-optimal deployments that meet functional and non-functional constraints.
  • Synthetic generation processes can produce varied infrastructure topologies and workload demands for testing pricing-based allocation methods.
  • A dataset of 9,600 precomputed scenarios provides a common benchmark for comparing pricing-driven and traditional approaches.

Where Pith is reading between the lines

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

  • If the representation holds, providers could adjust pricing add-ons to steer allocations toward preferred nodes without changing the underlying optimization logic.
  • The same pricing framing might apply to other configuration tasks in distributed systems, such as service composition or workflow placement.
  • Standardized pricing models could eventually allow infrastructure descriptions to be shared and analyzed across different management platforms.

Load-bearing premise

The constraints and capabilities of geographically distributed heterogeneous nodes can be encoded into pricing plans and add-ons without significant loss of expressiveness or computational efficiency.

What would settle it

A concrete infrastructure topology and demand set in which a valid low-cost allocation satisfying all location and performance constraints cannot be expressed as any combination of pricing plans and add-ons, or in which the pricing-based search returns an allocation that violates a constraint.

Figures

Figures reproduced from arXiv: 2604.12642 by Alejandro Garc\'ia-Fern\'andez, Antonio Ruiz-Cort\'es, Boris Sedlak, Jos\'e Antonio Parejo, Pantelis Frangoudis, Schahram Dustdar.

Figure 1
Figure 1. Figure 1: An example of a pricing: the Zoom platform. It [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Pricing-driven resource allocation workflow [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Mapping of a sample node into the pricing domain [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Sample topology generated around the Melbourne city [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Execution time of the proposed approach under increasing problem size. Each point represents the median solving time [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
read the original abstract

Deploying applications across the computing continuum requires selecting infrastructure nodes from geographically distributed and heterogeneous environments while satisfying constraints (e.g., performance, location). This decision problem is an important facet of resource allocation. As infrastructures grow in scale and heterogeneity, the resulting decision space becomes inherently combinatorial. Existing approaches typically formulate this problem as a constrained optimization task using ad-hoc representations of infrastructure topologies and demand, which hinders generalization across solutions. In contrast, Software as a Service ecosystems address a structurally similar configuration problem through pricings -structures whose plans and add-ons implicitly define the configuration space of possible subscriptions. Building on this observation, this work explores the potential of pricings as general-purpose representations of configuration spaces, positioning them as a promising alternative for addressing configuration problems, such as resource allocation, across the computing continuum. To this end, the paper presents the following contributions: i) a pricing-based formulation of the resource allocation problem in the computing continuum, enabling infrastructure configuration spaces to be represented using pricings; ii) a workflow that leverages PRIME, a pricing analysis engine, to explore these spaces and compute cost-optimal deployments satisfying functional and non-functional constraints; iii) generation processes for synthetic infrastructure topologies and workload demands; and iv) a dataset comprising 9,600 precomputed resource allocation scenarios to support benchmarking.

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 / 0 minor

Summary. The paper claims that pricing structures (plans and add-ons) from SaaS ecosystems can serve as general-purpose representations of configuration spaces for resource allocation in the computing continuum. It presents (i) a pricing-based formulation of the problem, (ii) a PRIME-based workflow to explore spaces and find cost-optimal deployments under functional/non-functional constraints, (iii) processes for generating synthetic infrastructure topologies and workloads, and (iv) a dataset of 9,600 precomputed scenarios for benchmarking.

Significance. If the central claim holds, the work would offer a more uniform and potentially generalizable alternative to ad-hoc topology/demand representations for combinatorial configuration problems in heterogeneous, geo-distributed systems. The synthetic generation processes and the 9,600-scenario dataset constitute concrete, reusable contributions that could support future empirical comparisons in the field.

major comments (1)
  1. [Contributions (i) and (ii)] Contributions (i) and (ii): The pricing-based formulation and PRIME workflow rest on the assumption that the full configuration space—including interdependent constraints across geographically distributed nodes (e.g., pairwise latencies, collective resource bounds, location-dependent performance)—can be faithfully encoded as independent plans and add-ons without material loss of expressiveness or combinatorial explosion. No concrete mapping, completeness argument, or worked example is supplied to substantiate this for the computing-continuum setting; the skeptic note correctly identifies this as load-bearing for the generality claim.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback and for identifying the need for greater substantiation of the core generality claim. We address the major comment below and will revise the manuscript to strengthen this aspect.

read point-by-point responses

  1. Referee: Contributions (i) and (ii): The pricing-based formulation and PRIME workflow rest on the assumption that the full configuration space—including interdependent constraints across geographically distributed nodes (e.g., pairwise latencies, collective resource bounds, location-dependent performance)—can be faithfully encoded as independent plans and add-ons without material loss of expressiveness or combinatorial explosion. No concrete mapping, completeness argument, or worked example is supplied to substantiate this for the computing-continuum setting; the skeptic note correctly identifies this as load-bearing for the generality claim.

    Authors: We agree that a concrete mapping and worked example are essential to support the claim that pricing structures can faithfully represent configuration spaces with interdependent constraints. In the current manuscript, infrastructure nodes are modeled as plans with add-ons capturing per-node attributes (resources, location, base performance), while inter-node constraints such as pairwise latencies and collective bounds are expressed as separate functional/non-functional requirements passed to the PRIME optimizer. This separation avoids encoding all dependencies directly inside the pricing but does require an explicit translation step. We will add a dedicated subsection with a worked example (e.g., a 5-node geo-distributed topology with latency and collective CPU bounds) that shows the precise mapping from continuum elements to plans/add-ons, how PRIME enforces the remaining constraints, and a brief analysis of expressiveness versus combinatorial growth. This revision will directly address the load-bearing assumption. revision: yes

Circularity Check

0 steps flagged

No significant circularity in the pricing-based resource allocation formulation

full rationale

The paper proposes a pricing-based formulation for resource allocation in the computing continuum as an alternative to ad-hoc representations, along with a workflow leveraging PRIME, synthetic generation processes, and a benchmarking dataset. No derivation chain, equations, fitted parameters presented as predictions, or self-citation load-bearing uniqueness claims appear in the provided abstract or contributions. The work is a conceptual reframing and generative effort that remains self-contained without any step reducing outputs to inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the unproven premise that pricing structures can serve as lossless or near-lossless encodings of arbitrary infrastructure topologies and constraints. No free parameters, invented entities, or additional axioms are stated in the abstract.

axioms (1)
  • domain assumption Pricing plans and add-ons can implicitly define the full configuration space of possible infrastructure subscriptions without loss of necessary constraints.
    Invoked in the contrast with ad-hoc representations and the positioning of pricings as general-purpose alternatives.

pith-pipeline@v0.9.0 · 5554 in / 1247 out tokens · 52146 ms · 2026-05-10T15:15:58.156603+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages

  1. [1]

    Cloud continuum: The definition,

    S. Moreschini, F. Pecorelli, X. Li, S. Naz, D. H ¨astbacka, and D. Taibi, “Cloud continuum: The definition,”IEEE Access, vol. 10, pp. 131 876– 131 886, 2022

    work page 2022

  2. [2]

    Resource allocation challenges in the cloud and edge continuum,

    P. Soumplis, P. Kokkinos, A. Kretsis, P. Nicopolitidis, G. Papadimitriou, and E. Varvarigos, “Resource allocation challenges in the cloud and edge continuum,” inAdvances in Computing, Informatics, Networking and Cybersecurity: A Book Honoring Professor Mohammad S. Obaidat’s Significant Scientific Contributions. Springer, 2022, pp. 443–464

    work page 2022

  3. [3]

    Resource allocation optimization model for computing continuum,

    M. Mihaiu, B.-C. Mocanu, C. Negru, A. Petrescu-Nit ,˘a, and F. Pop, “Resource allocation optimization model for computing continuum,” Mathematics, vol. 13, no. 3, p. 431, 2025

    work page 2025

  4. [4]

    A comprehensive survey on resource allocation strategies in fog/cloud environments,

    J. Vergara, J. Botero, and L. Fletscher, “A comprehensive survey on resource allocation strategies in fog/cloud environments,”Sensors, vol. 23, no. 9, p. 4413, 2023

    work page 2023

  5. [5]

    Pricing4SaaS: Towards a pricing model to drive the operation of SaaS,

    A. Garc ´ıa-Fern´andez, J. A. Parejo, and A. Ruiz-Cort ´es, “Pricing4SaaS: Towards a pricing model to drive the operation of SaaS,” inInternational Conference on Advanced Information Systems Engineering, 2024, pp. 47–54

    work page 2024

  6. [6]

    Trends in industry support for pricing-driven devops in saas,

    A. Garc ´ıa-Fern´andez, J. A. Parejo, F. J. Cavero, and A. Ruiz-Cort ´es, “Trends in industry support for pricing-driven devops in saas,”IEEE Transactions on Services Computing, vol. 19, no. 1, pp. 726–737, 2026

    work page 2026

  7. [7]

    Automated Analysis of Pricings in SaaS-Based Information Systems,

    A. Garc ´ıa-Fern´andez, J. A. Parejo, P. Trinidad, and A. Ruiz-Cort ´es, “Automated Analysis of Pricings in SaaS-Based Information Systems,” inInternational Conference on Advanced Information Systems Engineer- ing, 2025, pp. 223–239

    work page 2025

  8. [8]

    Pricing-Driven Resource Allocation in the Cloud Continuum – Laboratory Package,

    Anonymous, “Pricing-Driven Resource Allocation in the Cloud Continuum – Laboratory Package,” 2026. [Online]. Available: https: //doi.org/10.5281/zenodo.19063676

    work page doi:10.5281/zenodo.19063676 2026

  9. [9]

    Integrating terms of service and service level agreements for automating cloud service man- agement,

    E. Molino-Pe ˜na, J. M. Garc´ıa, and A. Ruiz-Cort´es, “Integrating terms of service and service level agreements for automating cloud service man- agement,” inInternational Conference on Service-Oriented Computing. Springer, 2025, pp. 19–34

    work page 2025

  10. [10]

    Modeling and programming with gecode,

    C. Schulte, G. Tack, and M. Z. Lagerkvist, “Modeling and programming with gecode,”Schulte, Christian and Tack, Guido and Lagerkvist, Mikael, vol. 1, 2010

    work page 2010

  11. [11]

    Choco-solver: A Java library for constraint programming,

    C. Prud’homme and J.-G. Fages, “Choco-solver: A Java library for constraint programming,”Journal of Open Source Software, vol. 7, no. 78, p. 4708, 2022

    work page 2022

  12. [12]

    A- MINT: An LLM Pipeline for Automated Modeling of iPricings from SaaS Pricing Pages,

    F. J. Cavero, J. A. Parejo, J. C. Alonso, and A. Ruiz-Cort ´es, “A- MINT: An LLM Pipeline for Automated Modeling of iPricings from SaaS Pricing Pages,” inInternational Conference on Service-Oriented Computing. Springer, 2026, pp. 83–91

    work page 2026

  13. [13]

    What can i do here? iot service discovery in smart cities,

    E. Wang and R. Chow, “What can i do here? iot service discovery in smart cities,” in2016 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), 2016, pp. 1–6

    work page 2016

  14. [14]

    A decentralized approach for resource dis- covery using metadata replication in edge networks,

    I. Murturi and S. Dustdar, “A decentralized approach for resource dis- covery using metadata replication in edge networks,”IEEE Transactions on Services Computing, vol. 15, no. 5, pp. 2526–2537, 2022

    work page 2022

  15. [15]

    Qos-aware middleware for web services composition,

    L. Zeng, B. Benatallah, A. H. Ngu, M. Dumas, J. Kalagnanam, and H. Chang, “Qos-aware middleware for web services composition,”IEEE Transactions on software engineering, vol. 30, no. 5, pp. 311–327, 2004

    work page 2004

  16. [16]

    A survey on mobile edge computing: The communication perspective,

    Y . Mao, C. You, J. Zhang, K. Huang, and K. B. Letaief, “A survey on mobile edge computing: The communication perspective,”IEEE communications surveys & tutorials, vol. 19, no. 4, pp. 2322–2358, 2017

    work page 2017

  17. [17]

    Trust-based resource allocation in web services,

    K. Xiong and H. Perros, “Trust-based resource allocation in web services,” in2006 IEEE international conference on web services (ICWS’06). IEEE, 2006, pp. 663–672

    work page 2006

  18. [18]

    Markov blanket composition of slos,

    B. Sedlak, V . C. Pujol, P. K. Donta, and S. Dustdar, “Markov blanket composition of slos,” in2024 IEEE International Conference on Edge Computing and Communications (EDGE), 2024, pp. 128–138

    work page 2024

  19. [19]

    PhuLai/eua-dataset,

    P. Lai, “PhuLai/eua-dataset,” Dec. 2025, original-date: 2018- 06-05T08:32:30Z. [Online]. Available: https://github.com/PhuLai/ eua-dataset

    work page 2025

  20. [20]

    D. A. Menasc ´e, V . A. F. Almeida, and L. W. Dowdy,Capacity Planning for Web Services: Metrics, Models, and Methods. Prentice Hall, 2002

    work page 2002

  21. [21]

    L. A. Barroso, J. Clidaras, and U. H ¨olzle,The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines, ser. Synthesis Lectures on Computer Architecture. Morgan & Claypool Publishers, 2013

    work page 2013

  22. [22]

    Cloudscale: Elastic resource scaling for multi-tenant cloud systems,

    Z. Shen, S. Subbiah, X. Gu, and J. Wilkes, “Cloudscale: Elastic resource scaling for multi-tenant cloud systems,”IEEE Transactions on Parallel and Distributed Systems, vol. 23, no. 2, pp. 259–266, 2011

    work page 2011

  23. [23]

    Clipper: A low-latency online prediction serving system,

    D. Crankshaw, X. Wang, G. Zhou, M. J. Franklin, J. E. Gonzalez, and I. Stoica, “Clipper: A low-latency online prediction serving system,” inProceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation (NSDI). USENIX Association, 2017

    work page 2017

  24. [24]

    MS-GD-P: priority-based service deploy- ment for cloud-edge-end scenarios,

    H. Jin, H. Wang, and J. Luo, “MS-GD-P: priority-based service deploy- ment for cloud-edge-end scenarios,”The Journal of Supercomputing, vol. 80, no. 18, pp. 25 713–25 735, Dec. 2024

    work page 2024

  25. [25]

    A survey on resource allocation strategies in cloud computing,

    V . V . Vinothina, R. Sridaran, and P. Ganapathi, “A survey on resource allocation strategies in cloud computing,”International Journal of Advanced Computer Science and Applications, vol. 3, no. 6, 2012

    work page 2012

  26. [26]

    SLA Management in Intent-Driven Service Management Systems: A Taxon- omy and Future Directions,

    Y . Sharma, D. Bhamare, N. Sastry, B. Javadi, and R. Buyya, “SLA Management in Intent-Driven Service Management Systems: A Taxon- omy and Future Directions,”ACM Comput. Surv., vol. 55, no. 13s, pp. 292:1–292:38, Jul. 2023

    work page 2023

  27. [27]

    Intelli- gent resource allocation scheme for cloud-edge-end framework aided multi-source data stream,

    Y . Wu, C. Cai, X. Bi, J. Xia, C. Gao, Y . Tang, and S. Lai, “Intelli- gent resource allocation scheme for cloud-edge-end framework aided multi-source data stream,”EURASIP Journal on Advances in Signal Processing, vol. 2023, May 2023

    work page 2023

  28. [28]

    Designing Reconfigurable Intelligent Systems with Markov Blankets,

    B. Sedlak, V . C. Pujol, P. K. Donta, and S. Dustdar, “Designing Reconfigurable Intelligent Systems with Markov Blankets,” inService- Oriented Computing, Rome, Italy, 2023, pp. 42–50

    work page 2023

  29. [29]

    A survey on web services composition,

    S. Dustdar and W. Schreiner, “A survey on web services composition,” International journal of web and grid services, vol. 1, no. 1, pp. 1–30, 2005

    work page 2005

  30. [30]

    Web service composition: a survey of techniques and tools,

    A. L. Lemos, F. Daniel, and B. Benatallah, “Web service composition: a survey of techniques and tools,”ACM Computing Surveys (CSUR), vol. 48, no. 3, pp. 1–41, 2015

    work page 2015

  31. [31]

    Towards network- aware service composition in the cloud,

    S. Wang, A. Zhou, F. Yang, and R. N. Chang, “Towards network- aware service composition in the cloud,”IEEE Transactions on Cloud Computing, vol. 8, no. 4, pp. 1122–1134, 2016

    work page 2016