Skip to main content
SpringerLink
Log in

DRL-based admission control and resource allocation for 5G network slicing

  • Published:
Sādhanā Aims and scope Submit manuscript

Abstract

Network Slicing in 5G networks enables allocation and sharing of the underlying network resources of an infrastructure provider (INP) among multiple tenants of the INP. Each tenant generates slice requests specifying the resources required and the INP collects revenue from the tenants for hosting the admitted slices. In this paper, we have used the Prioritised Experience Replay-based Deep Q-Network with N step return (N-PERDQN) Reinforcement learning (RL) approach to solve the slice admission control and associated resource allocation problem, in a dynamic environment. The slices are classified as elastic and inelastic, with flexible and rigid service guarantee requirements respectively. We have also used Long Short-Term Memory (LSTM) to predict the admitted elastic slices’ future resource requirements to share resources among elastic slices more efficiently. This enables the INP to accept more slice requests to increase its revenue while maximizing system utilization. The system has been modeled using the SimPy discrete-event simulator and TensorFlow machine learning. Performance metrics studied include INP revenue generated, class-specific slice acceptance rate and resource utilization were studied on Materna data center trace. The proposed N-PERDQN approach performs better than other RL and heuristic methods by up to 10% on average, in terms of generating INP revenue. Also, N-PERDQN+LSTM accepts around 6% more slice requests than N-PERDQN without LSTM. We have also tested the performance of the proposed algorithm with an offline algorithm, that has complete set of incoming slice request information available beforehand. The proposed algorithm is able to achieve close to 82% to 86% in terms of revenue gain of the offline algorithm. The proposed mechanism was also studied using the Alibaba data center trace and similar improvements were observed.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12

Similar content being viewed by others

References

  1. Zhang S 2019 An overview of network slicing for 5G. IEEE Wirel. Commun. 26(3): 111–117

    Article  Google Scholar 

  2. Bega D, Gramaglia M, Banchs A, Sciancalepore V and Costa-Pérez X 2020 A machine learning approach to 5G infrastructure market optimization. IEEE Trans. Mob. Comput. 19: 498–512

    Article  Google Scholar 

  3. Aijaz A 2018 A radio resource slicing framework for 5G networks with haptic communications. IEEE Syst. J. 12: 2285–2296

    Article  Google Scholar 

  4. Mnih V, Kavukcuoglu K and Silver D et al 2015 Human-level control through deep reinforcement learning. Nature 518: 529–533

    Article  Google Scholar 

  5. Boutaba R, Salahuddin M A, Limam N, Ayoubi S, Shahriar N and Estrada-Solano F et al 2018 A comprehensive survey on machine learning for networking: evolution, applications and research opportunities. J. Internet Serv. Appl.. https://doi.org/10.1186/s13174-018-0087-2

    Article  Google Scholar 

  6. Francois-Lavet V, Henderson P, Islam R, Bellemare M G and Pineau J, 2018 An introduction to deep reinforcement learning. Found. Trends Mach. Learn. 11

  7. Li R, Zhao Z, Sun Q and I C L, Yang C and Chen X et al 2018 Deep reinforcement learning for resource management in network slicing. IEEE Access 6: 74429–74441

  8. Chen X, Zhao Z, Wu C, Bennis M, Liu H and Ji Y et al 2019 Multi-tenant cross-slice resource orchestration: a deep reinforcement learning approach. IEEE J. Sel. Areas Commun. 37(10): 2377–2392

    Article  Google Scholar 

  9. Huynh N V, Hoang D T, Nguyen D and Dutkiewicz E 2019 Optimal and fast real-time resource slicing with deep dueling neural networks. IEEE J. Sel. Areas Commun. 37: 1455–1470

    Article  Google Scholar 

  10. Bakri S, Brik B and Ksentini A 2021 On using reinforcement learning for network slice admission control in 5G: offline vs. online. Wiley Int. J. Commun. Syst. 34

  11. Saxena S and Sivalingam K M 2022 Slice admission control using overbooking for enhancing provider revenue in 5G Networks. In Proc. IEEE/IFIP Network Operations and Management Symposium (NOMS), pp. 1–7

  12. Schaul T, Quan J, Antonoglou I and Silver D 2016 Prioritized Experience Replay. https://arxiv.org/abs/1511.05952

  13. Google 2021. TensorFlow. https://www.tensorflow.org

  14. Materna GmbH Information & Communications G Dortmund 2017. MATERNA Dataset Trace. http://gwa.ewi.tudelft.nl/datasets/gwa-t-13-materna

  15. Luo S, Xu H, Ye K, Xu G, Zhang L and Yang G et al 2022 The power of prediction: Microservice auto scaling via workload learning. In Proceedings of the ACM Symposium on Cloud Computing, pp. 355–369

  16. Salvat J X, Zanzi L, Garcia-Saavedra A, Sciancalepore V and Costa X 2018 Overbooking network slices through yield-driven end-to-end orchestration. In Proc. ACM CoNEXT, pp. 353–365

  17. Han B, Sciancalepore V, Feng D, Costa X and Schotten H 2019 A utility-driven multi-queue admission control solution for network slicing. In Proc. of IEEE INFOCOM, pp. 55–63

  18. Raza M R, Rostami A, Wosinska L and Monti P 2019 A slice admission policy based on big data analytics for multi-tenant 5G networks. J. Lightw. Technol. 37: 1690–1697

    Article  Google Scholar 

  19. Caballero P, Banchs A, de Veciana G and Costa-Pérez X 2019 Network slicing games: enabling customization in multi-tenant mobile networks. IEEE/ACM Trans. Netw. 27: 662–675

    Article  Google Scholar 

  20. Sciancalepore V, Costa-Pérez X and Banchs A 2019 RL-NSB: reinforcement learning-based 5G network slice broker. IEEE/ACM Trans. Netw. 27: 1543–1557

    Article  Google Scholar 

  21. Jiang M, Condoluci M and Mahmoodi T 2016 Network slicing management and prioritization in 5G mobile systems. In Proc. of European Wireless Conference, pp. 1–6

  22. Soliman H M and Leon-Garcia A 2016 QoS-aware frequency-space network slicing and admission control for virtual wireless networks. In Proc. of IEEE GLOBECOM, pp. 1–6

  23. Vincenzi M, López-Aguilera E and Garcia-Villegas E 2019 Maximizing infrastructure providers’ revenue through network slicing in 5G. IEEE Access 7: 128283–128297

    Article  Google Scholar 

  24. Lee K, Hong S, Kim S J, Rhee I and Chong S 2009 SLAW: A New Mobility Model for Human Walks. In Proc. of IEEE INFOCOM, pp. 855–863

  25. Watkins C J C H and Dayan P 2004 Technical note: Q-learning. Mach. Learn. 8: 279–292

    Article  MATH  Google Scholar 

  26. Grondman I, Buşoniu L, Lopes G A D and Robert 2012 A survey of actor-critic reinforcement learning: standard and natural policy gradients. IEEE Trans. Syst. Man Cybernet. Part C (Appl. Rev.) 42: 1291–1307

  27. Gers F, Schmidhuber J and Cummins F 2000 Learning to forget: continual prediction with LSTM. Neural Comput. 12: 2451–2471

    Article  Google Scholar 

  28. Godfrey L B and Gashler M S 2018 Neural decomposition of time-series data for effective generalization. IEEE Trans. Neural Netw. Learn. Syst. 29: 2973–2985

    MathSciNet  Google Scholar 

  29. Mnih V, Kavukcuoglu K, Silver D, Graves A, Antonoglou I, Wierstra D et al 2013 Playing atari with deep reinforcement learning. arXiv:1312.5602

  30. Hasselt H V, Guez A and Silver D 2016 Deep reinforcement learning with double Q-learning. In Proc. AAAI, pp. 2094–2100

  31. Klaus G and Müller T V 2002 SimPy process-based discrete-event simulation framework. https://simpy.readthedocs.io/en/latest/

  32. Chollet F 2015 Keras: the Python deep learning API. https://keras.io

Download references

Acknowledgements

This work was supported a grant from VMWare University Research, USA. The authors gratefully thank the anonymous reviewers and members of DAWNLAB at IIT Madras for their valuable comments and suggestions.

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, Indian Institute of Technology Madras, Chennai, 600036, India

    Saurav Chakraborty & Krishna M Sivalingam

Authors
  1. Saurav Chakraborty
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Krishna M Sivalingam
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Krishna M Sivalingam.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chakraborty, S., Sivalingam, K.M. DRL-based admission control and resource allocation for 5G network slicing. Sādhanā 48, 155 (2023). https://doi.org/10.1007/s12046-023-02201-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s12046-023-02201-4

Keywords

Search

Navigation