BI-TE: achieving GNN-based bandwidth indistinguishable traffic engineering in multi-domain SDN

Abstract

Software Defined Networking (SDN) offers Traffic Engineering (TE) great flexibility by decoupling the control and data plane. As network services become more diverse, the single-domain control architecture is no longer sufficient to meet scalability requirements, and the multi-domain and multi-level distributed control architecture is gaining popularity. However, traffic engineering across multiple domains poses challenges, particularly when each administrative domain is unwilling to disclose its network topology and resource information due to privacy concerns. To address this issue, this paper adopts the concept of differential privacy and perturbs the domain information to achieve bandwidth indistinguishable TE. Unfortunately, perturbations may decrease the accuracy of the TE algorithm’s resource allocation, negatively affecting performance. To mitigate this problem, we propose BI-TE, which utilizes a GNN-based bandwidth utilization prediction model to assist the controller in selecting the optimal forwarding path, thereby enhancing TE efficiency. Experimental results demonstrate that compared to abstraction-based hierarchical TE, BI-TE can reduce the processing time by nearly 24.35% while ensuring network bandwidth utilization close to 90%. Additionally, the fairness of allocation is also guaranteed.

Graphical abstract

Keywords

SDN / traffic engineering / privacy protection / distributed control / GNN

References

[1] H ong C Y, K andula S, M ahajan R, Z hang M, G ill V, N anduri M, W attenhofer R . Achieving high utilization with software-driven WAN. ACM SIGCOMM Computer Communication Review, 2013, 43( 4): 15–26

[2] G uo Z, D ou S, W u W, X ia Y . Toward flexible and predictable path programmability recovery under multiple controller failures in software-defined WANs. IEEE/ACM Transactions on Networking, 2023, 31( 5): 1965–1980

[3] A l-Tam F, C orreia N . Fractional switch migration in multi-controller software-defined networking. Computer Networks, 2019, 157: 1–10

[4] D as T, G urusamy M . Controller placement for resilient network state synchronization in multi-controller SDN. IEEE Communications Letters, 2020, 24( 6): 1299–1303

[5] Joshi K D, Kataoka K. PRIME-Q: privacy aware end-to-end QOS framework in multi-domain SDN. In: Proceedings of 2019 IEEE Conference on Network Softwarization. 2019, 169−177

[6] C hen Q, S hi S, L i X, Q ian C, Z hong S . SDN-based privacy preserving cross domain routing. IEEE Transactions on Dependable and Secure Computing, 2019, 16( 6): 930–943

[7] Andrés M E, Bordenabe N E, Chatzikokolakis K, Palamidessi C. Geo-indistinguishability: differential privacy for location-based systems. In: Proceedings of 2013 ACM SIGSAC Conference on Computer & Communications Security. 2013, 901−914

[8] T ogou M A, C hekired D A, K houkhi L, M untean G M . A hierarchical distributed control plane for path computation scalability in large scale software-defined networks. IEEE Transactions on Network and Service Management, 2019, 16( 3): 1019–1031

[9] Zhao L, Hua J, Liu Y, Qu W, Zhang S, Zhong S. Distributed traffic engineering for multi-domain software defined networks. In: Proceedings of the 39th IEEE International Conference on Distributed Computing Systems. 2019, 492−502

[10] El Houda Z A, Khoukhi L. A hierarchical fog computing framework for network attack detection in SDN. In: Proceedings of IEEE International Conference on Communications. 2022, 4366−4371

[11] Berde P, Gerola M, Hart J, Higuchi Y, Kobayashi M, Koide T, Lantz B, O’Connor B, Radoslavov P, Snow W, Parulkar G. ONOS: towards an open, distributed SDN OS. In: Proceedings of the 3rd Workshop on Hot Topics in Software Defined Networking. 2014, 1−6

[12] S anoussi N, C hetioui K, O rhanou G, E l Hajji S . ITC: intrusion tolerant controller for multicontroller SDN architecture. Computers & Security, 2023, 132: 103351

[13] Z hou H, Z heng Y, J ia X, S hu J . Collaborative prediction and detection of DDoS attacks in edge computing: a deep learning-based approach with distributed SDN. Computer Networks, 2023, 225: 109642

[14] C hen C, L iao Z, J u Y, H e C, Y u K, W an S . Hierarchical domain-based multicontroller deployment strategy in SDN-enabled space−air−ground integrated network. IEEE Transactions on Aerospace and Electronic Systems, 2022, 58( 6): 4864–4879

[15] Ferguson A D, Gribble S D, Hong C Y, Killian C E, Mohsin W, Mühe H, Ong J, Poutievski L, Singh A, Vicisano L, Alimi R, Chen S S, Conley M, Mandal S, Nagaraj K, Bollineni K N, Sabaa A, Zhang S, Zhu M, Vahdat A. Orion: Google’s software-defined networking control plane. In: Proceedings of the 18th USENIX Symposium on Networked Systems Design and Implementation. 2021, 83−98

[16] M oufakir T, Z hani M F, G herbi A, B ouachir O . Collaborative multi-domain routing in SDN environments. Journal of Network and Systems Management, 2022, 30( 1): 23

[17] N ayyer A, S harma A K, A wasthi L K . Laman: a supervisor controller based scalable framework for software defined networks. Computer Networks, 2019, 159: 125–134

[18] L in P, B i J, W olff S, W ang Y, X u A, C hen Z, H u H, L in Y . A west-east bridge based SDN inter-domain testbed. IEEE Communications Magazine, 2015, 53( 2): 190–197

[19] C hiang S H, W ang C H, Y ang D N, L iao W, C hen W T . Distributed multicast traffic engineering for multi-domain software-defined networks. IEEE Transactions on Parallel and Distributed Systems, 2023, 34( 2): 446–462

[20] L iu Y, Z hao L, H ua J, Q u W, Z hang S, Z hong S . Distributed traffic engineering for multi-domain SDN without trust. IEEE Transactions on Cloud Computing, 2022, 10( 4): 2481–2496

[21] H ua J, Z hao L, Z hang S, L iu Y, G e X, Z hong S . Topology-preserving traffic engineering for hierarchical multi-domain SDN. Computer Networks, 2018, 140: 62–77

[22] J oshi K D, K ataoka K . pSMART: a lightweight, privacy-aware service function chain orchestration in multi-domain NFV/SDN. Computer Networks, 2020, 178: 107295

[23] Dolson D, Homma S, López D, Boucadair M. Hierarchical Service Function Chaining (hSFC). RFC, 2018

[24] X iang Q, T ony Wang X, J ensen Zhang J, N ewman H, R ichard Yang Y, J ace Liu Y . Unicorn: unified resource orchestration for multi-domain, geo-distributed data analytics. Future Generation Computer Systems, 2019, 93: 188–197

[25] Z hang J, L iu Y, L i Z, L u Y . Forecast-assisted service function chain dynamic deployment for SDN/NFV-enabled cloud management systems. IEEE Systems Journal, 2023, 17( 3): 4371–4382

[26] M ao B, T ang F X, F adlullah Z M, K ato N . An intelligent route computation approach based on real-time deep learning strategy for software defined communication systems. IEEE Transactions on Emerging Topics in Computing, 2021, 9( 3): 1554–1565

[27] M ajdoub M, E l Kamel A, Y oussef H . DQR: an efficient deep q-based routing approach in multi-controller software defined wan (SD-WAN). Journal of Interconnection Networks, 2020, 20( 4): 2150002

[28] S un P, G uo Z, L an J, L i J, H u Y, B aker T . ScaleDRL: a scalable deep reinforcement learning approach for traffic engineering in SDN with pinning control. Computer Networks, 2021, 190: 107891

[29] S carselli F, G ori M, T soi A C, H agenbuchner M, M onfardini G . The graph neural network model. IEEE Transactions on Neural Networks, 2009, 20( 1): 61–80

[30] Li Y, Tarlow D, Brockschmidt M, Zemel R S. Gated graph sequence neural networks. In: Proceedings of the 4th International Conference on Learning Representations. 2016

[31] Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph attention networks. 2018, arXiv preprint arXiv: 1710.10903

[32] Ge Z, Hou J, Nayak A. Forecasting SDN end-to-end latency using graph neural network. In: Proceedings of 2023 International Conference on Information Networking. 2023, 293−298

[33] Y an B, L iu Q, S hen J, L iang D . Flowlet-level multipath routing based on graph neural network in openflow-based SDN. Future Generation Computer Systems, 2022, 134: 140–153

[34] L i J, S un P, H u Y . Traffic modeling and optimization in datacenters with graph neural network. Computer Networks, 2020, 181: 107528

[35] A lmasan P, S uárez-Varela J, R usek K, B arlet-Ros P, C abellos-Aparicio A . Deep reinforcement learning meets graph neural networks: exploring a routing optimization use case. Computer Communications, 2022, 196: 184–194

[36] G eyer F . DeepComNet: performance evaluation of network topologies using graph-based deep learning. Performance Evaluation, 2019, 130: 1–16

[37] H erbster M, W armuth M K . Tracking the best expert. Machine Learning, 1998, 32( 2): 151–178

[38] J ain S, K umar A, M andal S, O ng J, P outievski L, S ingh A, V enkata S, W anderer J, Z hou J, Z hu M, Z olla J, H ölzle U, S tuart S, V ahdat A . B4: experience with a globally-deployed software defined wan. ACM SIGCOMM Computer Communication Review, 2013, 43( 4): 3–14

[39] Gilmer J, Schoenholz S S, Riley P F, Vinyals O, Dahl G E. Neural message passing for quantum chemistry. In: Proceedings of the 34th International Conference on Machine Learning. 2017, 1263−1272

[40] Rusek K, Suárez-Varela J, Mestres A, Barlet-Ros P, Cabellos-Aparicio A. Unveiling the potential of graph neural networks for network modeling and optimization in SDN. In: Proceedings of 2019 ACM Symposium on SDN Research. 2019, 140−151

[41] Chung J, Gulcehre C, Cho K, Bengio Y. Empirical evaluation of gated recurrent neural networks on sequence modeling. 2014, arXiv preprint arXiv: 1412.3555

[42] Klambauer G, Unterthiner T, Mayr A, Hochreiter S. Self-normalizing neural networks. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, 972−981

[43] S pring N, M ahajan R, W etherall D . Measuring ISP topologies with rocketfuel. ACM SIGCOMM Computer Communication Review, 2002, 32( 4): 133–145

[44] Roughan M, Greenberg A, Kalmanek C, Rumsewicz M, Yates J, Zhang Y. Experience in measuring backbone traffic variability: models, metrics, measurements and meaning. In: Proceedings of the 2nd ACM SIGCOMM Workshop on Internet Measurment. 2002, 91−92

[45] H ei X, Z hang J, B ensaou B, C heung C C . Wavelength converter placement in least-load-routing-based optical networks using genetic algorithms. Journal of Optical Networking, 2004, 3( 5): 363–378

[46] O rlowski S, W essäly R, P ióro M, T omaszewski A . Sndlib 1. 0—survivable network design library. Networks: An International Journal, 2010, 55( 3): 276–286

[47] Barreto F, Wille E C G, Nacamura Jr L. Fast emergency paths schema to overcome transient link failures in OSPF routing. 2012, arXiv preprint arXiv: 1204.2465

[48] Pedro J, Santos J, Pires J. Performance evaluation of integrated OTN/DWDM networks with single-stage multiplexing of optical channel data units. In: Proceedings of the 13th International Conference on Transparent Optical Networks. 2011, 1−4

RIGHTS & PERMISSIONS

Higher Education Press