A Distributed Multi-Tier Orchestration Framework for the Rapid Recovery of Virtualized Network Functions
DOI:
https://doi.org/10.64915/RADAP.2026.105.%25pKeywords:
network function virtualization (NFV), network slicing, local reconfiguration, distributed recovery, SLA-driven optimizationAbstract
The rapid adoption of 5G network slicing over the virtualized infrastructure has led to the identification of a key gap in the existing recovery frameworks whereby centralized orchestration approaches are unable to enforce the sub-second restoration mandates of the Ultra-Reliable Low-Latency Communication (URLLC) slices, while simple heuristics cannot observe the Service Level Agreement (SLA) heterogeneity across slice types. This paper presents the DLR (Distributed Latency-aware Recovery) framework – a three-tier hierarchical orchestration architecture for the prompt recovery of VNFs (Virtualized Network Functions) in sliced 5G. The framework is based on a control decomposition into a Global Orchestrator, Regional Slice Managers (RSMs), and Local Monitoring Agents (LMAs), where decisions on the recovery are made locally based on the SLA violations without any central bottlenecks. RSM placement is formulated as a multi-objective optimization problem with minimum node-to-manager latency, load imbalance, and inter-manager coordination overhead with a genetic algorithm solver producing 11–33% better combined quality scores than baseline strategies. Dedicated recovery algorithms deal with both a catastrophic failure of nodes by a hop limited and prioritized order of reassignment of VNFs as well as dealing with progressive overloads by using scoring-based proactive migration with given cutoff to expand or shorten the migration period. A novel four-component VNF Migration Cost Function is formulated based on jointly accounting for compute, network, migration, and SLA penalty costs, resulting in up to 44% fewer migration steps for progressive overload instances. For cross-domain scenarios, a lightweight consensus protocol coordinates with neighboring RSMs without central arbitration. Simulation results on a synthetic 10-node topology show that DLR has a median recovery time of 0.7 s (5.4 x faster than CGO), with URLLC SLA violations limited to 4% (compared to 14% for the Nearest Available Matching) and a 27% reduction in operational costs compared to latency-only approaches.
References
1. Li W., et al. (2019). Latency-Optimal VNF Resource Allocation for 5G Backhaul Transport Network Slicing. Applied Sciences, Vol. 9, Iss. 4, 701; DOI: 10.3390/app9040701.
2. Ordonez-Lucena J., et al. (2017). Network slicing for 5G with SDN/NFV: Concepts, architectures, and challenges. IEEE Communications Magazine, Vol. 55, No. 5, pp. 80--87, DOI: 10.1109/MCOM.2017.1600935.
3. Dobreff G., Bader A. and Pac A. (2025). A QoE and Availability-Aware Framework for Network Slice Placement and Resource Allocation. IEEE Access, Vol. 13, pp. 1481--1495, DOI: 10.1109/access.2024.3523900.
4. Gatzianas M., et al. (2021). Offline Joint Network and Computational Resource Allocation for Energy-Efficient 5G. Applied Sciences, Vol. 11, Iss. 22, 10547, DOI: 10.3390/app112210547.
5. Shinde S., Marabissi D. and Tarchi D. (2021). A network operator-biased approach for multi-service NF placement in 5G slicing. Computer Networks, Vol. 201, DOI: 10.1016/j.comnet.2021.108598.
6. Vidhya P., et al. (2025). Dynamic network slicing based resource management and VNF migration in 5G networks. Computer Networks, Vol. 259, DOI: 10.1016/j.comnet.2025.111064.
7. Di Mauro M., et al. (2025). Reliability and availability in virtualized networks: A survey. arXiv, DOI: 10.48550/arXiv.2503.22034.
8. Thien P., et al. (2024). Optimizing Placement and Scheduling for VNF by a Multi-objective Optimization Genetic Algorithm. International Journal of Computational Intelligence Systems, Vol. 17, DOI: 10.1007/s44196-024-00430-x.
9. Hirayama T., Jibiki M. and Kafle V. (2023). Salience-based VNF placement method for evolving multiple network slices. Computer Networks, Vol. 236, DOI: 10.1016/j.comnet.2023.110002.
10. Benkacem I., Taleb T., Bagaa M. and Flinck H. (2018). Optimal VNFs Placement in CDN Slicing Over Multi-Cloud Environment. IEEE Journal on Selected Areas in Communications, Vol. 36, Iss. 3, pp. 616--627, DOI: 10.1109/jsac.2018.2815441.
11. Addad R., et al. (2020). Optimization Model for Cross-Domain Network Slices in 5G Networks. IEEE Transactions on Mobile Computing, Vol. 19, Iss. 5, pp. 1156--1169, DOI: 10.1109/tmc.2019.2905599.
12. Nguyen T., et al. (2023). Optimizing Resource Allocation and VNF Embedding in RAN Slicing. IEEE Transactions on Network and Service Management, Vol. 21, Iss. 2, pp. 2187--2199, DOI: 10.1109/tnsm.2023.3319309.
13. Abdellatif A., et al. (2023). Intelligent-Slicing: An AI-Assisted Network Slicing Framework for 5G-and-Beyond. IEEE Transactions on Network and Service Management, Vol. 20, Iss. 2, pp. 1024--1039, DOI: 10.1109/tnsm.2023.3274236.
14. Panitsas I., et al. (2025). SlicePilot: Demystifying Network Slice Placement in Heterogeneous Cloud Infrastructures. arXiv, DOI: 10.48550/arXiv.2509.18545.
15. Chiosi M., et al. (2012). Network Functions Virtualisation: An Introduction. ETSI White Paper.
16. Gember-Jacobson A., et al. (2014). OpenNF: Enabling innovation in network function control. ACM SIGCOMM.
17. Cohen R., Lewin-Eytan L., Naor J.S. and Raz D. (2015). Near optimal placement of virtual network functions. Proc. IEEE INFOCOM, pp. 1346--1354, DOI: 10.1109/INFOCOM.2015.7218511.
18. Venkatapathy S., et al. (2024). Slice-aware 5G network orchestration framework based on dual-slice isolation (D-SIMS). Scientific Reports, Vol. 14, Article number: 18623, DOI: 10.1038/s41598-024-68892-9.
19. ETSI (2014). Network Functions Virtualisation (NFV); Management and Orchestration. GS NFV-MAN 001 v1.1.1. Sophia Antipolis.
20. Halabian H. (2019). Distributed Resource Allocation Optimization in 5G Virtualized Networks. IEEE Journal on Selected Areas in Communications, Vol. 37, Iss. 3, pp. 627--642, DOI: 10.1109/jsac.2019.2894305.
21. Motalleb M., et al. (2023). Resource Allocation in an Open RAN System Using Network Slicing. IEEE Transactions on Network and Service Management, Vol. 20, Iss 1, pp. 627--642, DOI: 10.1109/tnsm.2022.3205415.
22. Heller B., Sherwood R. and McKeown N. (2012). The controller placement problem. Proc. ACM SIGCOMM HotSDN, pp. 7--12, DOI: 10.1145/2342441.2342444.
23. Sahu V., Sahu N. and Sahu R. (2024). Challenges and opportunities of 5G network: A review. American Journal of Electrical and Computer Engineering, Vol. 8, No. 1, pp. 11--20. DOI: 10.11648/j.ajece.20240801.12.
24. ETSI (2016). Network Functions Virtualisation (NFV); Os-Ma-nfvo Interface and Information Model Specification. GS NFV-IFA 013 v2.1.1. Sophia Antipolis.
25. ITU-R (2023). IMT-2030 Framework. Recommendation ITU-R M.2160-0. Geneva.
26. Herrera J. G. and Botero J. F. (2016). Resource allocation in NFV: A comprehensive survey. IEEE Transactions on Network and Service Management, Vol. 13, No. 3, pp. 518--532, DOI: 10.1109/TNSM.2016.2598420.
27. Di Mauro M., et al. (2025). Reliability and availability in virtualized networks: A survey on standards, modeling approaches, and research challenges. arXiv preprint arXiv:2503.22034, DOI: 10.48550/arXiv.2503.22034.
28. 3GPP (2021). Service requirements for the 5G system. TS 22.261, Release 16.
29. Moens H. and De Turck F. (2014). VNF-P: A model for efficient placement of virtualized network functions. Proc. CNSM, Rio de Janeiro, pp. 418--423. DOI: 10.1109/CNSM.2014.7014205.
30. 3GPP (2023). System architecture for the 5G System (5GS). TS 23.501, Release 17.
31. 3GPP (2018). Telecommunication management; Study on management and orchestration of network slicing. TR 28.801, Release 15.
32. Sudhamani C., et al. (2023). A survey on 5G coverage improvement techniques. Sensors, Vol. 23, No. 4, Article 2356, DOI: 10.3390/s23042356.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 С. В. Суліма, Є. Д. Карашевич
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
