pith. sign in

arxiv: 2605.27106 · v1 · pith:QJH3OXZGnew · submitted 2026-05-26 · 💻 cs.DC · cs.GT· cs.MA· cs.NI

Autonomic Federated-Market Orchestration for the Edge-Cloud Continuum

Lauri Lov\'en , Roberto Morabito , Abhishek Kumar , Susanna Pirttikangas , Jukka Riekki , Sasu Tarkoma This is my paper

Pith reviewed 2026-06-29 15:34 UTC · model grok-4.3

classification 💻 cs.DC cs.GTcs.MAcs.NI
keywords federated marketedge-cloud continuumWalrasian equilibriumgross substitutesservice dependency DAGsautonomic orchestrationdecentralised allocationMAPE-K loop
0
0 comments X

The pith

Decentralised price-based allocation matches centralised welfare on tree and series-parallel service graphs under gross-substitutes valuations.

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

The paper presents Neural Pub/Sub, a federated-broker system for edge-cloud resource orchestration that organises itself through market price signals rather than central commands. Its central proposition states that when service valuations satisfy the gross-substitutes property on tree or series-parallel dependency DAGs, the decentralised price mechanism reaches the same total welfare as an omniscient central planner. This matters for autonomous administrative domains because it supports scalable allocation while enforcing data sovereignty and handling failures without a single point of control. Empirical runs on a multi-domain testbed show the market approach preserving completion rates where simpler schedulers collapse under load and matching or exceeding sharded central oracles at equal process counts.

Core claim

The paper establishes a Walrasian convergence proposition: under gross-substitutes valuations on tree and series-parallel service-dependency DAGs, decentralised price-based allocation matches the welfare of a centralised oracle. The claim is embedded in the Plan step of a MAPE-K loop that performs marginal-cost clearing-price analysis and placement over a polymatroidal feasibility region, then dispatches across federated brokers using bounded-staleness price signals.

What carries the argument

The Walrasian convergence proposition for price-based allocation, which equates decentralised and centralised welfare on restricted DAG structures.

If this is right

  • The market mechanism sustains completion rates above 98 percent under broker death and network partition.
  • Sovereignty enforcement adds no measurable runtime overhead.
  • Market allocation preserves completion rates across increasing arrival rates where round-robin drops from 98.8 percent to 3.3 percent.
  • At equal process count the federated market stays within plus or minus 1.5 percent of a four-shard centralised orchestrator.

Where Pith is reading between the lines

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

  • The same price-signal substrate could coordinate placement decisions across additional administrative boundaries once the dependency restriction is met.
  • Admission control and information completeness emerge as direct consequences of the price-discovery process rather than separate policy layers.
  • If real service graphs can be approximated by series-parallel reductions, the convergence result would cover a wider class of practical workflows.

Load-bearing premise

Service-dependency graphs are restricted to trees or series-parallel structures and valuations satisfy the gross-substitutes property.

What would settle it

A workload on a non-tree, non-series-parallel DAG or with valuations that violate gross substitutes in which the market mechanism produces strictly lower total welfare than the central oracle.

Figures

Figures reproduced from arXiv: 2605.27106 by Abhishek Kumar, Jukka Riekki, Lauri Lov\'en, Roberto Morabito, Sasu Tarkoma, Susanna Pirttikangas.

Figure 1
Figure 1. Figure 1: Conceptual overview of Neural Pub/Sub. A pipeline (gray) is dispatched stage-by-stage (orange) onto [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: MAPE-K control loop closed at each broker. Solid arrows: control flow per allocation epoch. Dashed [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Consolidated experimental results: (a) near-optimality (market vs single-process oracle, all cells); [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
read the original abstract

The edge-cloud computing continuum demands self-management mechanisms that scale across autonomous administrative domains while honouring tenant- and operator-specified data sovereignty. We present Neural Pub/Sub, a federated-broker autonomic substrate whose self-organising behaviour emerges from market-based price signals rather than centralised control. Its MAPE-K control loop closes over per-broker health and load monitoring, marginal-cost clearing-price analysis, placement planning over a polymatroidal feasibility region, federated cross-domain dispatch, and shared peer subscription summaries with bounded-staleness price signals. The Plan step is anchored in a Walrasian convergence proposition: under gross-substitutes valuations on tree and series-parallel service-dependency DAGs, decentralised price-based allocation matches the welfare of a centralised oracle. We evaluate the substrate on a 4-VM, 4-domain, 48-worker federated edge-cloud testbed (single data centre, 50 ms emulated WAN) in a 1005-run campaign augmented by a fair-process-count sharded-oracle comparator. The federated market dominates a single-process oracle by 2-4% with 45 of 45 per-seed wins (sign-test p ~ 2.8e-14, Hodges-Lehmann median -39.6 ms); against a four-shard centralised orchestrator at equal process count the gap stays within +/-1.5% across all nine (pipeline, load) cells. Round-robin completion rate collapses 98.8% -> 22.4% -> 3.3% across arrival rates 5/10/15 pps while the market preserves completion; the advantage decomposes into three Walrasian properties (information completeness, admission control, price discovery). Federation withstands broker death and network partition (completion rate >= 98.7% across 75 cells), and sovereignty enforcement adds no measurable runtime overhead across 60 governance-grid runs. Heterogeneous-domain stressors and cross-site WAN deployment remain future work.

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 presents Neural Pub/Sub, a federated-broker autonomic substrate for edge-cloud continuum orchestration. It uses a MAPE-K control loop driven by market price signals for self-organization across domains while enforcing data sovereignty. The Plan step is anchored in a Walrasian convergence proposition asserting that, under gross-substitutes valuations on tree and series-parallel service-dependency DAGs, decentralized price-based allocation matches the welfare of a centralized oracle. Evaluation on a 4-VM/4-domain/48-worker testbed (1005 runs) shows the federated market outperforming single-process and sharded oracles by 2-4% in latency with statistical significance, preserving completion rates under load where round-robin fails, and maintaining robustness to broker failures and partitions.

Significance. If the Walrasian equivalence holds, the work supplies a principled, mechanism-design foundation for scalable self-management in multi-administrative-domain edge-cloud systems. The experimental decomposition into information completeness, admission control, and price discovery, together with the sovereignty and fault-tolerance results, would strengthen the case for market-based autonomic substrates over purely centralized or heuristic alternatives.

major comments (1)
  1. [Abstract] Abstract (Plan step): the Walrasian convergence proposition is stated without a theorem formulation, proof sketch, or derivation steps showing how gross-substitutes plus the tree/series-parallel restriction imply equivalence to the centralized oracle. This is load-bearing for the claim that the market mechanism is theoretically anchored rather than heuristic.
minor comments (2)
  1. [Evaluation] The testbed description (4-VM, 4-domain, 50 ms emulated WAN) should explicitly state whether the service-dependency DAGs used in the 1005 runs satisfy the tree or series-parallel restriction required by the proposition.
  2. [Plan step] Notation for the polymatroidal feasibility region and marginal-cost clearing prices is introduced without a compact reference to the precise mathematical definition employed.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive summary and significance assessment. We address the major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (Plan step): the Walrasian convergence proposition is stated without a theorem formulation, proof sketch, or derivation steps showing how gross-substitutes plus the tree/series-parallel restriction imply equivalence to the centralized oracle. This is load-bearing for the claim that the market mechanism is theoretically anchored rather than heuristic.

    Authors: We agree the abstract states the proposition concisely without a full theorem or derivation due to length limits. The manuscript body (Section 4) formulates it as Proposition 1 with a proof sketch: gross-substitutes valuations ensure a unique equilibrium price vector via tatonnement, while the tree/series-parallel DAG restriction allows a dynamic-programming recursion over sub-DAGs that computes the welfare-maximizing allocation in polynomial time, matching the centralized optimum. We will revise the abstract to cite 'Proposition 1' and note the key assumptions enabling equivalence. If the referee requests, we can lengthen the proof sketch in the main text. revision: yes

Circularity Check

0 steps flagged

No circularity: central claim invokes external economic theorem without internal reduction.

full rationale

The paper anchors its Plan step in a stated Walrasian convergence proposition under gross-substitutes valuations on restricted DAGs, presented as a standard result rather than derived or fitted within the manuscript. No equations, parameters, or self-citations reduce the welfare-equivalence claim to the paper's own inputs by construction. Experiments compare implementations but do not substitute for or circularly validate the proposition. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central theoretical claim rests on domain assumptions about valuation properties and graph structure drawn from economic theory rather than new axioms invented by the paper.

axioms (1)
  • domain assumption Gross-substitutes valuations hold on tree and series-parallel service-dependency DAGs
    Directly invoked to establish the Walrasian convergence proposition in the Plan step.

pith-pipeline@v0.9.1-grok · 5930 in / 1262 out tokens · 47490 ms · 2026-06-29T15:34:04.787825+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

90 extracted references · 21 canonical work pages · 5 internal anchors

  1. [1]

    2024.3GPP TS 23.288: Architecture enhancements for 5G System (5GS) to support network data analytics services

    3GPP. 2024.3GPP TS 23.288: Architecture enhancements for 5G System (5GS) to support network data analytics services. Technical Report TS 23.288 V18.6.0. 3rd Generation Partnership Project. Release 18

    2024

  2. [2]

    2024.3GPP TS 28.541: Management and orchestration; 5G Network Resource Model (NRM); Stage 2 and stage 3

    3GPP. 2024.3GPP TS 28.541: Management and orchestration; 5G Network Resource Model (NRM); Stage 2 and stage 3. Technical Report TS 28.541 V18.7.0. 3rd Generation Partnership Project. Release 18

    2024

  3. [3]

    3GPP. 2024. 5G; Architecture enhancements for 5G System (5GS) to support network data analytics services (3GPP TS 23.288). Technical Specification, Release 18. Accessed 2026-05-04

    2024

  4. [4]

    3GPP. 2024. 5G; Procedures for the 5G System (5GS) (3GPP TS 23.502). Technical Specification, Release 18

    2024

  5. [5]

    3GPP. 2024. 5G; System architecture for the 5G System (5GS) (3GPP TS 23.501). Technical Specification, Release 18

    2024

  6. [6]

    3GPP. 2024. Management and orchestration; Concepts, use cases and requirements (3GPP TS 28.530). Technical Specification, Release 18

    2024

  7. [7]

    Anthropic. 2024. Model Context Protocol (MCP). https://modelcontextprotocol.io

    2024

  8. [8]

    Lawrence M Ausubel. 2004. An efficient ascending-bid auction for multiple objects.American Economic Review94, 5 (2004), 1452–1475

    2004

  9. [9]

    Katz, Ben Clifford, Rohan Kumar, Lukasz Lacinski, Ryan Chard, Justin M

    Yadu Babuji, Anna Woodard, Zhuozhao Li, Daniel S. Katz, Ben Clifford, Rohan Kumar, Lukasz Lacinski, Ryan Chard, Justin M. Wozniak, Ian Foster, Michael Wilde, and Kyle Chard. 2019. Parsl: pervasive parallel programming in Python. InProceedings of the 28th International Symposium on High-Performance Parallel and Distributed Computing (HPDC). 25–36

    2019

  10. [10]

    Franklin, Joseph M

    Peter Bailis, Shivaram Venkataraman, Michael J. Franklin, Joseph M. Hellerstein, and Ion Stoica. 2014. Quantifying eventual consistency with PBS. InCommunications of the ACM, Vol. 57. 93–102. 0:32 Lovén et al

    2014

  11. [11]

    Faizul Bari, Shihabur Rahman Chowdhury, Reaz Ahmed, Raouf Boutaba, and Otto Carlos M

    Md. Faizul Bari, Shihabur Rahman Chowdhury, Reaz Ahmed, Raouf Boutaba, and Otto Carlos M. B. Duarte. 2016. Orchestrating virtualized network functions.IEEE Transactions on Network and Service Management13, 4 (2016), 725–739

    2016

  12. [12]

    Pete Beckman, Jack Dongarra, Nicola Ferrier, Geoffrey Fox, Terry Moore, Dan Reed, and Micah Beck. 2020. Harnessing the computing continuum for programming our world.Fog Computing: Theory and Practice(2020), 215–230

    2020

  13. [13]

    BerriAI. 2024. LiteLLM: Call All LLM APIs Using the OpenAI Format. Accessed 2026-05-04. https://docs.litellm.ai/ docs/routing

    2024

  14. [14]

    2019.Kubeflow and Kubeflow Pipelines

    Ekaba Bisong. 2019.Kubeflow and Kubeflow Pipelines. Apress, 671–685

    2019

  15. [15]

    Eirina Bourtsoulatze, David Burth Kurka, and Deniz Gündüz. 2019. Deep joint source-channel coding for wireless image transmission. InIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 4514–4518

    2019

  16. [16]

    Radu Calinescu, Carlo Ghezzi, Marta Kwiatkowska, and Raffaela Mirandola. 2012. Self-adaptive software needs quantitative verification at runtime.Commun. ACM55, 9 (2012), 69–77

    2012

  17. [17]

    Paris Carbone, Stephan Ewen, Gyula Fóra, Seif Haridi, Stefan Richter, and Kostas Tzoumas. 2017. State management in Apache Flink: Consistent stateful distributed stream processing.Proceedings of the VLDB Endowment10, 12 (2017), 1718–1729

    2017

  18. [18]

    Antonio Carzaniga and et al. 2001. Design and evaluation of a wide-area event notification service.ACM Transactions on Computer Systems (TOCS)19, 3 (2001), 332–383

    2001

  19. [19]

    Betty H. C. Cheng, Rogério de Lemos, Holger Giese, Paola Inverardi, Jeff Magee, Jesper Andersson, Basil Becker, Nelly Bencomo, Yuriy Brun, Bojan Cukic, et al. 2009. Software Engineering for Self-Adaptive Systems: A Research Roadmap. InSoftware Engineering for Self-Adaptive Systems. Lecture Notes in Computer Science, Vol. 5525. Springer, 1–26

    2009

  20. [20]

    CrewAI. 2024. CrewAI: Framework for orchestrating role-playing autonomous AI agents. https://www.crewai.com

    2024

  21. [21]

    Jeffrey Dean and Sanjay Ghemawat. 2008. MapReduce: Simplified data processing on large clusters.Commun. ACM 51, 1 (2008), 107–113

    2008

  22. [22]

    Maechling, Rajiv Mayani, Weiwei Chen, Rafael Ferreira da Silva, Miron Livny, and Kent Wenger

    Ewa Deelman, Karan Vahi, Gideon Juve, Mats Rynge, Scott Callaghan, Philip J. Maechling, Rajiv Mayani, Weiwei Chen, Rafael Ferreira da Silva, Miron Livny, and Kent Wenger. 2015. Pegasus, a workflow management system for science automation.Future Generation Computer Systems46 (2015), 17–35

    2015

  23. [23]

    R. J. Duffin. 1965. Topology of series-parallel networks.J. Math. Anal. Appl.10, 2 (1965), 303–318

    1965

  24. [24]

    Jack Edmonds. 1970. Submodular functions, matroids, and certain polyhedra. InCombinatorial Structures and their Applications (Proc. Calgary Internat. Conf., 1969). Gordon and Breach, 69–87

    1970

  25. [25]

    ETSI. 2014. Network Functions Virtualisation (NFV); Management and Orchestration (GS NFV-MAN 001). Group Specification

    2014

  26. [26]

    ETSI. 2019. Zero-touch network and Service Management (ZSM); Reference Architecture (GS ZSM 002). Group Specification

    2019

  27. [27]

    Patrick Th Eugster and et al. 2003. The many faces of publish/subscribe.ACM computing surveys (CSUR)35, 2 (2003), 114–131

    2003

  28. [28]

    2005.Submodular Functions and Optimization(2 ed.)

    Satoru Fujishige. 2005.Submodular Functions and Optimization(2 ed.). Elsevier

    2005

  29. [29]

    Gagan Goel, Vahab Mirrokni, and Renato Paes Leme. 2015. Polyhedral clinching auctions and the adwords polytope. J. ACM62, 3 (2015), 1–27

    2015

  30. [30]

    Google. 2025. Agent2Agent (A2A) Protocol. https://a2a-protocol.org

    2025

  31. [31]

    R. L. Graham. 1966. Bounds for certain multiprocessing anomalies.Bell System Technical Journal45, 9 (1966), 1563–1581

    1966

  32. [32]

    Faruk Gul and Ennio Stacchetti. 1999. Walrasian equilibrium with gross substitutes.Journal of Economic Theory87, 1 (1999), 95–124

    1999

  33. [33]

    Taicheng Guo, Xiuying Chen, Yaqi Wang, Ruidi Chang, Shichao Pei, Nitesh V Chawla, Olaf Wiest, and Xiangliang Zhang

  34. [34]

    Large language model based multi-agents: A survey of progress and challenges.arXiv preprint arXiv:2402.01680 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  35. [35]

    IBM. 2006. An Architectural Blueprint for Autonomic Computing. IBM White Paper. https://www-03.ibm.com/ autonomic/pdfs/AC%20Blueprint%20White%20Paper%20V7.pdf

    2006

  36. [36]

    Alexander S Kelso and Vincent P Crawford. 1982. Job matching, coalition formation, and gross substitutes.Econometrica 50, 6 (1982), 1483–1504

    1982

  37. [37]

    Kephart and David M

    Jeffrey O. Kephart and David M. Chess. 2003. The Vision of Autonomic Computing.Computer36, 1 (2003), 41–50. doi:10.1109/MC.2003.1160055

    work page doi:10.1109/mc.2003.1160055 2003

  38. [38]

    Changhoon Kim, Anirudh Sivaraman, Naga Katta, Antonin Bas, Advait Dixit, and Lawrence J. Wobker. 2015. In-band Network Telemetry via Programmable Dataplanes. InACM SIGCOMM Demo

    2015

  39. [39]

    Hyunjoo Kim and Manish Parashar. 2011. CometCloud: an autonomic cloud engine. InCloud Computing: Principles and Paradigms. Wiley, 275–297. Autonomic Federated-Market Orchestration for the Edge–Cloud Continuum 0:33

    2011

  40. [40]

    Henna Kokkonen, Susanna Pirttikangas, and Lauri Lovén. 2023. EISim: A Platform for Simulating Intelligent Edge Orchestration Solutions. InProc. IEEE/ACM Symposium on Edge Computing (SEC), Poster Session. 1–2. Open-source: https://github.com/hennas/EISim

    2023

  41. [41]

    KServe Contributors. 2024. KServe: Highly scalable and standards based model inference platform on Kubernetes. https://kserve.github.io/website/. Accessed 2026-05-04

    2024

  42. [42]

    Kubernetes SIG Scheduling. 2024. Kueue: Kubernetes-native Job Queueing. https://kueue.sigs.k8s.io/. Accessed 2026-05-04

    2024

  43. [43]

    Yu-Kwong Kwok and Ishfaq Ahmad. 1999. Static scheduling algorithms for allocating directed task graphs to multiprocessors.Comput. Surveys31, 4 (1999), 406–471

    1999

  44. [44]

    LangChain. 2024. LangGraph: Agent Runtime and Orchestration Framework. Accessed 2026-05-04. https://www. langchain.com/langgraph

    2024

  45. [45]

    Benny Lehmann, Daniel Lehmann, and Noam Nisan. 2006. Combinatorial auctions with decreasing marginal utilities. Games and Economic Behavior55, 2 (2006), 270–296

    2006

  46. [46]

    Letaief, Yuanming Shi, Jianmin Lu, and Jianhua Lu

    Khaled B. Letaief, Yuanming Shi, Jianmin Lu, and Jianhua Lu. 2022. Edge Artificial Intelligence for 6G: Vision, Enabling Technologies, and Applications.IEEE Journal on Selected Areas in Communications40, 1 (2022), 5–36. doi:10.1109/JSAC.2021.3126076

    work page doi:10.1109/jsac.2021.3126076 2022

  47. [47]

    Linux Foundation Networking. 2024. ONAP Architecture Overview (Open Network Automation Platform). https: //www.onap.org/. Accessed 2026-05-04

    2024

  48. [48]

    Xi Liu, Jun Liu, Hong Wu, and Jing Dong. 2024. A family of truthful mechanisms for resource allocation with multi- attribute demand in mobile edge computing.Cluster Computing27 (2024), 11595–11610. doi:10.1007/s10586-024-04552-x

    work page doi:10.1007/s10586-024-04552-x 2024

  49. [49]

    Lauri Lovén and et al. 2023. How can ai be distributed in the computing continuum? introducing the neural pub/sub paradigm.arXiv preprint arXiv:2309.02058(2023)

    work page arXiv 2023

  50. [50]

    Lauri Lovén, Abhishek Kumar, Alexander Engelhardt, Alaa Saleh, Roberto Morabito, Xiaoli Liu, Naser Hossein Motlagh, and Sasu Tarkoma. 2026. Neural Router: Semantic Content Matching for Agentic AI.arXiv preprint arXiv:2605.25701 (2026)

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  51. [51]

    Lauri Lovén, Ella Peltonen, Erkki Harjula, and Susanna Pirttikangas. 2021. Weathering the Reallocation Storm: Large-Scale Analysis of Edge Server Workload. In2021 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit). IEEE, 317–322. doi:10.1109/EuCNC/6GSummit51104.2021.9482593

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1109/eucnc/6gsummit51104.2021.9482593 2021

  52. [52]

    Lauri Lovén, Ella Peltonen, Leena Ruha, Erkki Harjula, and Susanna Pirttikangas. 2022. A dark and stormy night: Reallocation storms in edge computing.EURASIP Journal on Wireless Communications and Networking2022, 1 (2022),

    2022

  53. [53]

    doi:10.1186/s13638-022-02170-y

    work page doi:10.1186/s13638-022-02170-y

  54. [54]

    Lauri Lovén, Alaa Saleh, Reza Farahani, Ilir Murturi, Miguel Bordallo López, Praveen Kumar Donta, and Schahram Dustdar. 2026. Real-Time AI Service Economy: A Framework for Agentic Computing Across the Continuum.arXiv preprint arXiv:2603.05614(2026)

    work page arXiv 2026

  55. [55]

    Haoxiang Luo, Yinqiu Liu, Ruichen Zhang, Jiacheng Wang, Gang Sun, Dusit Niyato, Hongfang Yu, Zehui Xiong, Xianbin Wang, and Xuemin Shen. 2025. Toward edge general intelligence with multiple-large language model (Multi-LLM): Architecture, trust, and orchestration.arXiv preprint arXiv:2507.00672(2025)

    work page arXiv 2025

  56. [56]

    Dimitrios Michael Manias, Ali Chouman, and Abdallah Shami. 2024. Semantic routing for enhanced performance of LLM-assisted intent-based 5G core network management and orchestration. InIEEE Global Communications Conference (GLOBECOM). IEEE, 1–6

    2024

  57. [57]

    Paul Milgrom and John Roberts. 1990. Rationalizability, learning, and equilibrium in games with strategic complemen- tarities.Econometrica58, 6 (1990), 1255–1277

    1990

  58. [58]

    Philipp Moritz, Robert Nishihara, Stephanie Wang, Alexey Tumanov, Richard Liaw, Eric Liang, Melih Elibol, Zongheng Yang, William Paul, Michael I Jordan, and Ion Stoica. 2018. Ray: A distributed framework for emerging AI applications. InUSENIX Symposium on Operating Systems Design and Implementation (OSDI). 561–577

    2018

  59. [59]

    Morris, Volodymyr Kuleshov, Vitaly Shmatikov, and Alexander M

    John X. Morris, Volodymyr Kuleshov, Vitaly Shmatikov, and Alexander M. Rush. 2023. Text Embeddings Reveal (Almost) As Much As Text. InProc. EMNLP

    2023

  60. [60]

    Noam Nisan and Ilya Segal. 2006. The communication requirements of efficient allocations and supporting prices. Journal of Economic Theory129, 1 (2006), 192–224

    2006

  61. [61]

    Not Diamond. 2024. Not Diamond: Intelligent AI Model Routing. Accessed 2026-05-04. https://docs.notdiamond.ai/ docs/what-is-model-routing

    2024

  62. [62]

    NVIDIA. 2024. NVIDIA Triton Inference Server. https://developer.nvidia.com/triton-inference-server. Accessed 2026-05-04

    2024

  63. [63]

    O-RAN Alliance. 2024. O-RAN Architecture Description (O-RAN.WG1.OAD). Technical Specification, R003. Accessed 2026-05-04

    2024

  64. [64]

    O-RAN Alliance. 2024. O-RAN Non-RT RIC Architecture (O-RAN.WG2.Non-RT-RIC-ARCH). Technical Specification. Accessed 2026-05-04. 0:34 Lovén et al

    2024

  65. [65]

    O-RAN Alliance. 2024. O-RAN Security Requirements and Controls Specification (O-RAN.WG11.Security- Requirements-Spec). Technical Specification. Accessed 2026-05-04

    2024

  66. [66]

    RouteLLM: Learning to Route LLMs with Preference Data

    Isaac Ong, Amjad Almahairi, Vincent Wu, Wei-Lin Chiang, Tianhao Wu, Joseph E. Gonzalez, M. Waleed Kadous, and Ion Stoica. 2024. RouteLLM: Learning to Route LLMs with Preference Data. arXiv preprint arXiv:2406.18665

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  67. [67]

    OpenAI. 2025. OpenAI Agents SDK. https://github.com/openai/openai-agents-python

    2025

  68. [68]

    OpenRouter. 2025. OpenRouter: Unified API for AI Models with Intelligent Provider Routing. Accessed 2026-05-04. https://openrouter.ai/docs/guides/routing/provider-selection

    2025

  69. [69]

    P4 Language Consortium. 2023. P4Runtime Specification. https://p4.org/p4-spec/p4runtime/main/P4Runtime-Spec. html. Accessed 2026-05-04

    2023

  70. [70]

    Manish Parashar. 2025. Everywhere and Nowhere: Envisioning a Computing Continuum for Science.Computing in Science & Engineering27, 1 (2025), 51–56. doi:10.1109/MCSE.2025.3543924

    work page doi:10.1109/mcse.2025.3543924 2025

  71. [71]

    Esa Piri, Pekka Ruuska, Teemu Kanstrén, Jukka Mäkelä, Jari Korva, Atso Hekkala, Ari Pouttu, Olli Liinamaa, Matti Latva- aho, Kari Vierimaa, and Harri Valasma. 2016. 5GTN: A test network for 5G application development and testing. In 2016 European Conference on Networks and Communications (EuCNC). IEEE, 313–318. doi:10.1109/EuCNC.2016.7561054

    work page doi:10.1109/eucnc.2016.7561054 2016

  72. [72]

    Zhijin Qin, Xiaoming Tao, Jianhua Lu, Wen Tong, and Geoffrey Ye Li. 2022. Semantic communications: Principles and challenges.arXiv preprint arXiv:2201.01389(2022)

    work page arXiv 2022

  73. [73]

    Alaa Saleh, Roberto Morabito, Schahram Dustdar, Sasu Tarkoma, Susanna Pirttikangas, and Lauri Lovén. 2025. Towards Message Brokers for Generative AI: Survey, Challenges, and Opportunities.Comput. Surveys58, 1 (2025), 1–37. doi:10.1145/3742891

    work page doi:10.1145/3742891 2025

  74. [74]

    Alaa Saleh, Sasu Tarkoma, Anders Lindgren, Praveen Kumar Donta, Schahram Dustdar, Susanna Pirttikangas, and Lauri Lovén. 2025. Memindex: Agentic event-based distributed memory management for multi-agent systems.ACM Transactions on Autonomous and Adaptive Systems(2025). doi:10.1145/3774946

    work page doi:10.1145/3774946 2025

  75. [75]

    Mazeiar Salehie and Ladan Tahvildari. 2009. Self-Adaptive Software: Landscape and Research Challenges.ACM Transactions on Autonomous and Adaptive Systems (TAAS)4, 2 (2009), 1–42. doi:10.1145/1516533.1516538

    work page doi:10.1145/1516533.1516538 2009

  76. [76]

    Emilio Calvanese Strinati and Sergio Barbarossa. 2021. 6G Networks: Beyond Shannon Towards Semantic and Goal-Oriented Communications.Computer Networks190 (2021), 107930. doi:10.1016/j.comnet.2021.107930

    work page doi:10.1016/j.comnet.2021.107930 2021

  77. [77]

    Villegas, Hausi A

    Gabriel Tamura, Norha M. Villegas, Hausi A. Müller, João Pedro Sousa, Basil Becker, Gabor Karsai, Serge Mankovskii, Mauro Pezzè, Bradley Schmerl, Ladan Tahvildari, and Kenny Wong. 2013. Towards practical runtime verification and validation of self-adaptive software systems. InSoftware Engineering for Self-Adaptive Systems II (LNCS 7475). Springer, 108–132

    2013

  78. [78]

    2012.Publish/subscribe systems: design and principles

    Sasu Tarkoma. 2012.Publish/subscribe systems: design and principles. John Wiley & Sons

    2012

  79. [79]

    Sasu Tarkoma, Roberto Morabito, and Jaakko Sauvola. 2023. AI-native Interconnect Framework for Integration of Large Language Model Technologies in 6G Systems.arXiv preprint arXiv:2311.05842(2023)

    work page arXiv 2023

  80. [80]

    Haluk Topcuoglu, Salim Hariri, and Min-You Wu. 2002. Performance-effective and low-complexity task scheduling for heterogeneous computing.IEEE Transactions on Parallel and Distributed Systems13, 3 (2002), 260–274

    2002

Showing first 80 references.