Abstract
Software-defined networking and network functions virtualisation are making networks programmable and consequently much more flexible and agile. To meet service-level agreements, achieve greater utilisation of legacy networks, faster service deployment, and reduce expenditure, telecommunications operators are deploying increasingly complex service function chains (SFCs). Notwithstanding the benefits of SFCs, increasing heterogeneity and dynamism from the cloud to the edge introduces significant SFC placement challenges, not least adding or removing network functions while maintaining availability, quality of service, and minimising cost. In this paper, an availability- and energy-aware solution based on reinforcement learning (RL) is proposed for dynamic SFC placement. Two policy-aware RL algorithms, Advantage Actor-Critic (A2C) and Proximal Policy Optimisation (PPO), are compared using simulations of a ground truth network topology based on the Rede Nacional de Ensino e Pesquisa Network, Brazil’s National Teaching and Research Network backbone. The simulation results show that PPO generally outperformed A2C and a greedy approach in terms of both acceptance rate and energy consumption. The biggest difference in the PPO when compared to the other algorithms relates to the SFC availability requirement of 99.965%; the PPO algorithm median acceptance rate is 67.34% better than the A2C algorithm. A2C outperforms PPO only in the scenario where network servers had a greater number of computing resources. In this case, the A2C is 1% better than the PPO.
Similar content being viewed by others
References
Ali HMM, Lawey AQ, El-Gorashi TE, Elmirghani JM (2015) Energy efficient disaggregated servers for future data centers. In: 2015 20th European Conference on Networks and Optical Communications-(NOC), IEEE pp. 1–6.
Andrade E, Nogueira B, Matos R, Callou G, Maciel P (2017) Availability modeling and analysis of a disaster-recovery-as-a-service solution. Computing 99(10):929–954
Araujo J, Maciel P, Andrade E, Callou G, Alves V, Cunha P (2018) Decision making in cloud environments: an approach based on multiple-criteria decision analysis and stochastic models. J Cloud Comput 7(1):7
Araujo J, Maciel P, Torquato M, Callou G, Andrade E (2014) Availability evaluation of digital library cloud services. In: Dependable Systems and Networks (DSN), 2014 44th Annual IEEE/IFIP International Conference on, IEEE pp. 666–671
Arulkumaran K, Deisenroth MP, Brundage M, Bharath AA (2017) A brief survey of deep reinforcement learning. arXiv preprint arXiv:1708.05866
Bhamare D, Jain R, Samaka M, Erbad A (2016) A survey on service function chaining. J Netw Comput Appl 75:138–155
Cai J, Huang Z, Luo J, Liu Y, Zhao H, Liao L (2020) Composing and deploying parallelized service function chains. J Netw Comput Appl 163:102637
Chai H, Zhang J, Wang Z, Shi J, Huang T (2019) A parallel placement approach for service function chain using deep reinforcement learning. In: 2019 IEEE 5th International Conference on Computer and Communications (ICCC), IEEE pp. 2123–2128
Costa I, Araujo J, Dantas J, Campos E, Silva FA, Maciel P (2016) Availability evaluation and sensitivity analysis of a mobile backend-as-a-service platform. Qual Reliab Eng Int 32(7):2191–2205
Dâmaso A, Rosa N, Maciel P (2014) Reliability of wireless sensor networks. Sensors 14(9):15760–15785
Dâmaso A, Rosa N, Maciel P (2017) Integrated evaluation of reliability and power consumption of wireless sensor networks. Sensors 17(11):2547
Fan J, Guan C, Zhao Y, Qiao C (2017) Availability-aware mapping of service function chains. In: IEEE INFOCOM 2017-IEEE Conference on Computer Communications, IEEE pp. 1–9
Farshin A, Sharifian S (2019) A modified knowledge-based ant colony algorithm for virtual machine placement and simultaneous routing of nfv in distributed cloud architecture. J Supercomput 75(8):5520–5550
Gens F (2013) The 3rd platform: enabling digital transformation, vol 209. IDC, USA
Gissler B, Shrivastava P (2015) A system for design decisions based on reliability block diagrams. In: 2015 Annual Reliability and Maintainability Symposium (RAMS), pp. 1–6. IEEE
Gomez-Rodriguez MA, Sosa-Sosa VJ, Carretero J, Gonzalez JL (2020) Cloudbench: an integrated evaluation of vm placement algorithms in clouds. J Supercomput :1–34
Goodfellow I, Bengio Y, Courville A, Bengio Y (2016) Deep learning, vol 1. MIT press, Cambridge
ISG, NFV. Network functions virtualisation (nfv)-network operator perspectives on industry progress. ETSI GS NFV-SEC 001 V1.1.1. 2013. Available at https://www.etsi.org/deliver/etsi_gs/nfv-sec/001_099/001/01.01.01_60/gs_nfv-sec001v010101p.pdf. Accessed Apr 2021
Guo S, Dai Y, Xu S, Qiu X, Qi F (2019) Trusted cloud-edge network resource management: Drl-driven service function chain orchestration for iot. IEEE Internet Things J 7(7):6010–6022
Hasselt HV (2010) Double q-learning. Adv Neural Inf Process Syst 23:2613–2621
He W, Chen X, Qiu X, Guo S, Yu P (2019) Asco: an availability-aware service chain orchestration. In: 2019 IFIP/IEEE Symposium on Integrated Network and Service Management (IM), IEEE pp. 590–593
Høyland A, Rausand M (2009) System reliability theory: models and statistical methods, vol 420. Wiley, New Jersey
Jain R (1991) The art of computer systems performance analysis. Wiley, New Jersey
Jim M (2015) Nfv applications - key considerations for profitability. https://web.dialogic.com/making-nfv-profitable
Kaur K, Mangat V, Kumar K (2020) A comprehensive survey of service function chain provisioning approaches in sdn and nfv architecture. Comput Sci Rev 38:100298
Khezri HR, Moghadam PA, Farshbafan MK, Shah-Mansouri V, Kebriaei H, Niyato D (2019) Deep reinforcement learning for dynamic reliability aware nfv-based service provisioning. In: 2019 IEEE Global Communications Conference (GLOBECOM), IEEE pp. 1–6
Kouah R, Alleg A, Laraba A, Ahmed T (2018) Energy-aware placement for iot-service function chain. In: 2018 IEEE 23rd International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD), IEEE pp. 1–7
Kumar A, Pant S, Ram M (2017) System reliability optimization using gray wolf optimizer algorithm. Qual Reliab Eng Int 33(7):1327–1335
Leong YC, Radulescu A, Daniel R, DeWoskin V, Niv Y (2017) Dynamic interaction between reinforcement learning and attention in multidimensional environments. Neuron 93(2):451–463
Lesort T, Díaz-Rodríguez N, Goudou JF, Filliat D (2018) State representation learning for control: an overview. Neural Netw 108:379–392
Li G, Zhou H, Feng B, Li G (2018) Context-aware service function chaining and its cost-effective orchestration in multi-domain networks. IEEE Access 6:34976–34991
Li G, Zhou H, Feng B, Zhang Y, Yu S (2019) Efficient provision of service function chains in overlay networks using reinforcement learning. IEEE Trans Cloud Comput
Lima PA, Neto ASB, Maciel P (2020) Data centers’ services restoration based on the decision-making of distributed agents. Telecommun Syst :1–12
Luo Z, Wu C, Li Z, Zhou W (2019) Scaling geo-distributed network function chains: a prediction and learning framework. IEEE J Sel Areas Commun 37(8):1838–1850
Lynn T, Gourinovitch A, Svorobeh S, Endo PT (2018) Software defined networking and network functions virtualization - market briefing. https://recap-project.eu/media/market-briefings/
Mann ZÁ (2015) Allocation of virtual machines in cloud data centers-a survey of problem models and optimization algorithms. ACM Comput Surv (CSUR) 48(1):1–34
Matos R, Dantas J, Araujo J, Trivedi KS, Maciel P (2017) Redundant eucalyptus private clouds: availability modeling and sensitivity analysis. J Grid Comput 15(1):1–22
Mirjalily G, Zhiquan L (2018) Optimal network function virtualization and service function chaining: a survey. Chin J Electron 27(4):704–717
Mnih V, Badia AP, Mirza M, Graves A, Lillicrap T, Harley T, Silver D, Kavukcuoglu K (2016) Asynchronous methods for deep reinforcement learning. In: International Conference on Machine Learning, pp. 1928–1937
Moualla G, Turletti T, Saucez D (2018) An availability-aware sfc placement algorithm for fat-tree data centers. In: 2018 IEEE 7th International Conference on Cloud Networking (CloudNet), IEEE pp. 1–4
Mousavi SS, Schukat M, Howley E (2017) Traffic light control using deep policy-gradient and value-function-based reinforcement learning. IET Intell Transp Syst 11(7):417–423
Mundie C, de Vries P, Haynes P, Corwine M (2002) Trustworthy computing. Tech. rep, Technical report, p 10
Osband I, Blundell C, Pritzel A, Van Roy B (2016) Deep exploration via bootstrapped dqn. arXiv preprint http://arxiv.org/abs/1602.04621
Palhares A, Santos M, Endo P, Vitalino J, Rodrigues M, Gonçalves G, Sadok D, Sefidcon A, Wuhib F (2014) Joint allocation of nodes and links with load balancing in network virtualization. In: 2014 IEEE 28th International Conference on Advanced Information Networking and Applications, pp. 148–155. IEEE
Pan J, Wang X, Cheng Y, Yu Q (2018) Multisource transfer double dqn based on actor learning. IEEE Trans Neural Netw Learn Syst 29(6):2227–2238
Peng B, Li X, Gao J, Liu J, Chen YN, Wong KF (2018) Adversarial advantage actor-critic model for task-completion dialogue policy learning. In: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6149–6153. IEEE
Qiu Z, Zhang J, Ning P, Wen X (2017) Reliability modeling and analysis of sic mosfet power modules. In: IECON 2017-43rd Annual Conference of the IEEE Industrial Electronics Society, IEEE pp. 1459–1463
Ravichandiran S (2018) Hands-on reinforcement learning with Python: master reinforcement and deep reinforcement learning using OpenAI gym and tensorFlow. Packt Publishing Ltd, England
Santos GL, Endo PT, da Silva Lisboa MFF, da Silva LGF, Sadok D, Kelner J, Lynn T et al (2018) Analyzing the availability and performance of an e-health system integrated with edge, fog and cloud infrastructures. J Cloud Comput 7(1):16
Sayadnavard MH, Haghighat AT, Rahmani AM (2019) A reliable energy-aware approach for dynamic virtual machine consolidation in cloud data centers. J Supercomput 75(4):2126–2147
Schulman J, Wolski F, Dhariwal P, Radford A, Klimov O (2017) Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347
de Sousa NFS, Perez DAL, Rosa RV, Santos MA, Rothenberg CE (2019) Network service orchestration: a survey. Comput Commun 142:69–94
Sun G, Li Y, Yu H, Vasilakos AV, Du X, Guizani M (2019) Energy-efficient and traffic-aware service function chaining orchestration in multi-domain networks. Future -Gener Comput Syst 91:347–360
Sun P, Lan J, Li J, Guo Z, Hu Y (2020) Combining deep reinforcement learning with graph neural networks for optimal vnf placement. IEEE Commun Lett
Sutton RS, Barto AG (2018) Reinforcement learning: an introduction. MIT press, Cambridge
Tavakoli-Someh S, Rezvani MH (2019) Multi-objective virtual network function placement using nsga-ii meta-heuristic approach. J Supercomput 75(10):6451–6487
Torquato M, Torquato L, Maciel P, Vieira M (2019) Iaas cloud availability planning using models and genetic algorithms. In: 2019 9th Latin-American Symposium on Dependable Computing (LADC), IEEE pp. 1–10
Troia S, Alvizu R, Maier G (2019) Reinforcement learning for service function chain reconfiguration in nfv-sdn metro-core optical networks. IEEE Access 7:167944–167957
Xiao Y, Zhang Q, Liu F, Wang J, Zhao M, Zhang Z, Zhang J (2019) Nfvdeep: Adaptive online service function chain deployment with deep reinforcement learning. In: Proceedings of the International Symposium on Quality of Service, pp. 1–10
Xu Z, Zhang X, Yu S, Zhang J (2018) Energy-efficient virtual network function placement in telecom networks. In: 2018 IEEE International Conference on Communications (ICC), IEEE pp. 1–7
Zhang J, Wang Z, Ma N, Huang T, Liu Y (2018) Enabling efficient service function chaining by integrating nfv and sdn: architecture, challenges and opportunities. IEEE Netw 32(6):152–159
Zhang X, Xu Z, Fan L, Yu S, Qu Y (2019) Near-optimal energy-efficient algorithm for virtual network function placement. IEEE Trans Cloud Comput
Zheng Y, Li X, Xu L (2020) Balance control for the first-order inverted pendulum based on the advantage actor-critic algorithm. Int J Control Autom Syst 18(12):3093–3100
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Santos, G.L., Lynn, T., Kelner, J. et al. Availability-aware and energy-aware dynamic SFC placement using reinforcement learning. J Supercomput 77, 12711–12740 (2021). https://doi.org/10.1007/s11227-021-03784-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-021-03784-7