Aggregation-Guided Buffering and Orchestration for Wide-Area Mixture-of-Agents in the Optical Internet of Agents

Authors

  • Xiao Lin College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author
  • Zuqi Teng College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author
  • Wang Li College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author
  • Ziang Feng College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author
  • Weiling Lu College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author
  • Enguo Chen College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China Author

DOI:

https://doi.org/10.64509/jicn.24.140

Keywords:

Optical Internet of Agents, Mixture-of-Agents, spatio-temporal orchestration, hierarchical aggregation, store-and- forward

Abstract

Wide-area Mixture-of-Agents (MoA) collaboration requires the exchange of large payloads over the Optical Internet of Agents (O-IoA) under fluctuating traffic conditions. To increase agent participation within collaboration deadlines under network resource contention, we propose the Aggregation-Guided Buffering and Orchestration (AGBO) framework. AGBO coordinates hierarchical update aggregation, intermediate buffering, and optical transmission using a capacitated time-expanded resource graph (TERG). We develop analytical models showing that temporal buffering reduces collaboration failure probability at the cost of higher candidate-path search complexity. We also develop an AGBO algorithm combining two-stage maximum-flow orchestration with Meta Agent (MA) selection. Simulation results show that AGBO retains more Work Agents (WAs) across different wavelength-utilization levels, payload sizes, and collaboration deadlines. At a wavelength utilization of 60%, it improves the WA participation ratio by at least 57 percentage points over the two baselines. In a convolutional neural network (CNN)-based collaborative-learning proxy, AGBO also achieves faster accuracy improvement and higher global test accuracy alongside higher WA participation.

Downloads

Download data is not yet available.

References

[1] Boateng, G.O., Sami, H., Alagha, A., Elmekki, H., Hammoud, A., Mizouni, R., Mourad, A., Otrok, H., Bentahar, J., Muhaidat, S., et al.: A Survey on Large Language Models for Communication, Network, and Service Management: Application Insights, Challenges, and Future Directions. IEEE Communications Surveys & Tutorials 28(1), 527–566 (2026). https://doi.org/10.1109/COMST.2025.3564333

[2] Wang, Y., Guo, S., Pan, Y., Su, Z., Chen, F., Luan, T.H., Li, P., Kang, J., Niyato, D.: Internet of Agents: Fundamentals, Applications, and Challenges. IEEE Transactions on Cognitive Communications and Networking 12, 4476–4501 (2026). https://doi.org/10.1109/TCCN.2025.3623369

[3] Qian, C., Xie, Z., Wang, Y., Liu, W., Zhu, K., Xia, H., Dang, Y., Du, Z., Chen, W., Yang, C., et al.: Scaling Large Language Model-based Multi-Agent Collaboration. In Proceedings of The Thirteenth International Conference on Learning Representations (ICLR 2025), pp. 1–18 (2025)

[4] Yao, D., Rabbani, T., Li, T.: Federation over Text: Insight Sharing for Multi-Agent Reasoning. In Proceedings of The Fourteenth International Conference on Learning Representations (ICLR 2026), pp. 1–46 (2026).

[5] Wang, J., Wang, J., Athiwaratkun, B., Zhang, C., Zou, J.Y.: Mixture-of-Agents Enhances Large Language Model Capabilities. In Proceedings of The Thirteenth International Conference on Learning Representations (ICLR 2025), pp. 1–20 (2025).

[6] Fu, T., Min, Z., Zhang, H., Yan, J., Dai, G., Ouyang, W., Wang, Y.: Cache-to-Cache: Direct Semantic Communication Between Large Language Models. In Proceedings of The Fourteenth International Conference on Learning Representations (ICLR 2026), pp. 1–29 (2026).

[7] Zhang, C., Li, C., Shen, H., Zhang, J., Zou, J., Xie, N., Wang, P., Sun, Q., Bae, S.-H., Wang, X., et al.: Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization. arXiv preprint arXiv:2603.12933 (2026). https://doi.org/10.48550/arXiv.2603.12933

[8] Lin, X., Yang, H., Huang, Y., Chen, E.: Design of a Bulk Data Transfer Approach for Deadline-Critical Tasks in Optical Computing Power Networks. In Proceedings of 2025 13th International Conference on Information and Communication Networks (ICICN), pp. 239–244 (2025). https://doi.org/10.1109/ICICN67355.2025.11430506

[9] Zarei, M.H., Hadi, M., Pakravan, M.R.: Dynamic Routing and Spectrum Assignment in Elastic Optical Networks Using Graph Signal Correlation. Journal of Optical Communications and Networking 18(8), 848–861 (2026). https://doi.org/10.1364/JOCN.592663

[10] Fan, Y., Li, Y., Natalino, C., Guo, J., Wu, W., Ruan, R., Wang, W., Zhao, Y., Zhang, J.: Distributed model training task migration for hotspot management in intelligent computing center interconnection with tidal characteristics. IEEE Transactions on Network and Service Management 22(6), 5489–5502 (2025). https://doi.org/10.1109/TNSM.2025.3590011

[11] Chen, L., Li, Y., Natalino, C., Li, Y., Zhang, B., Fan, Y., Wang, W., Zhao, Y., Zhang, J.: Reliable and efficient RAR-based distributed model training in computing power network. Journal of Optical Communications and Networking 16(5), 527–540 (2024). https://doi.org/10.1364/JOCN.511165

[12] Lian, M., Zhao, Y., Li, X., Liu, W., Li, Y., Tornatore, M., Zhang, J.: Resource allocation in flexible-bandwidth fine-grained optical transport networks for geo-distributed machine learning. IEEE Internet of Things Journal 12(13), 25601–25619 (2025). https://doi.org/10.1109/JIOT.2025.3558933

[13] Hu, Q., Wang, W., Huang, C., Wang, X., Li, Y., Zhao, Y., Zheng, Y., Tan, Y., Zhang, J.: Task placement and traffic interleaving for cross-datacenter LLM training over optical networks. Journal of Optical Communications and Networking 18(2), 137–149 (2026). https://doi.org/10.1364/JOCN.579324

[14] Jiang, D., Ren, X., Lin, B.Y.: LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, pp. 14165–14178 (2023). https://doi.org/10.18653/v1/2023.acl-long.792

[15] Tekin, S. F., Ilhan, F., Huang, T., Hu, S., Liu, L.: LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity. In Findings of the Association for Computational Linguistics: EMNLP 2024, pp. 11951–11966 (2024). https://doi.org/10.18653/v1/2024.findings-emnlp.698

[16] Wang, Z., Moriyama, S., Wang, W.-Y., Gangopadhyay, B., Takamatsu, S.: Talk Structurally, Act Hierarchically: A Collaborative Refinement Framework for LLM Multi-Agent Systems. IEEE Transactions on Artificial Intelligence 7(9), 4993–5003 (2026). https://doi.org/10.1109/TAI.2026.3676025

[17] Ma, X., Ma, Y., Lin, C., Yan, S., Bi, J., Cao, Z., Tian, Y., Tresp, V., Schuetze, H.: Self-Evolving Multi-Agent Systems via Textual Backpropagation. In Findings of the Association for Computational Linguistics: ACL 2026, pp. 9918–9951 (2026). https://doi.org/10.18653/v1/2026.findings-acl.483

[18] Liu, Z., Zhang, Y., Li, P., Liu, Y., Yang, D.: A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration. In Proceedings of Conference on Language Modeling (COLM), pp. 1–30 (2024)

[19] Ong, I., Almahairi, A., Wu, V., Chiang, W.-L., Wu, T., Gonzalez, J.E., Kadous, M.W., Stoica, I.: RouteLLM: Learning to Route LLMs with Preference Data. In Proceedings of The Thirteenth International Conference on Learning Representations (ICLR 2025), pp. 1–16 (2025).

[20] Wang, Y., Kong, L., Zhu, M., Gu, J., Cai, Y., Zhang, J.: Dynamic SFC mapping and time scheduling with store-and-forward scheme based on DRL in inter-datacenter elastic optical networks. IEEE Transactions on Networking 34, 4110–4125 (2026). https://doi.org/10.1109/TON.2026.3674479

[21] Lin, X., Cai, Z., Zeng, C., Feng, Z., Lu, W., Chen, Y., Chen, E.: Fluctuation-Aware Routing and Scheduling for Bulk Transfers in Computing Power Optical Networks. In Proceedings of 2026 11th International Conference on Computer and Communication Systems (ICCCS), pp. 119–124 (2026). https://doi.org/10.1109/ICCCS69761.2026.11613512

[22] Lin, X., Sun, W., Wang, X., Yue, S., Veeraraghavan, M., Hu, W.: Time-space decoupled SnF scheduling of bulk transfers across inter-datacenter optical networks. IEEE Access 8, 24829–24846 (2020). https://doi.org/10.1109/ACCESS.2020.2970542

JICN140

Downloads

Published

2026-11-13

Issue

Section

Articles

How to Cite

Lin, X., Teng, Z., Li, W., Feng, Z., Lu, W., & Chen, E. (2026). Aggregation-Guided Buffering and Orchestration for Wide-Area Mixture-of-Agents in the Optical Internet of Agents. Journal of Intelligent Computing and Networking, 2(4), 1-17. https://doi.org/10.64509/jicn.24.140

Similar Articles

1-10 of 14

You may also start an advanced similarity search for this article.