網絡功能虛擬化（NFV）的出現徹底改變了網絡體系結構的基礎架構和服務管理。通過虛擬網絡功能（VNF），網絡運營商不僅可以降低其資本支出（CAPEX）和運營支出（OPEX），而且還可以縮短網絡服務部署的時間，也增加部署的彈性。然而，網絡功能虛擬化的主要挑戰之一是虛擬網絡功能放置問題，因放置問題會影響部署成本和服務質量。虛擬網絡功能放置問題是一個眾所周知的NP完全問題，在過去的研究中，許多的學者將放置問題表示為整數線性規劃（ILP）問題，並使用整數線性規劃求解器找到最佳放置位置。由於解整數線性規劃的問題十分耗時，因此在問題的規模較大時和動態工作負載情況下使用整數線性規劃求解器是不可行的。所以，許多學者也提出各種貪婪算法以尋找近似解。然而，這兩種方法都無法針對動態的流量快速地做出準確的放置決策。因此，在此論文中，我們提出了一種混合方法，該方法使用更少的時間來最大化網絡服務部署的總體利潤。由實驗結果表明，與貪婪算法相比，我們的混合方法可取得高達36％利潤的提升，與整數線性規劃解器相比，可提高30倍的速度。

The emergence of network function virtualization~(NFV) has revolutionized the infrastructure and service management of network architecture. Through virtual network function~(VNF), network operators not only can reduce their cost on Capital expenditures~(CAPEX), operating expenses~(OPEX), and power consumption, but also improve the time and flexibility of network service deployment. However, one of the key challenges of NFV is the VNF placement problem, which can affect both deployment cost and service quality. VNF placement is a well-known NP-complete problem; hence many previous studies formulate the problem as an Integer Linear Programming~(ILP) Problem and find the best placement using an ILP solver. Since solving ILP is time-consuming, using an ILP solver is infeasible for large scaled and dynamic workloads in real-time. As a result, various greedy algorithms have also been proposed to find approximate solutions. However, neither of these approaches can make quick and accurate placement decisions for dynamic traffic workload. Therefore, we propose a hybrid method that uses less time to maximize the overall profit of network service deployment. Our evaluations based on real backbone network traffic and topology show that our hybrid approach can achieve up to 36\% profit improvement compared to a pure greedy approach, while achieving x30 times computation time speedup over a pure ILP approach.

[1] Alleg, A., Kouah, R., Moussaoui, S., and Ahmed, T. Virtual network functions
placement and chaining for real­time applications. In 2017 IEEE 22nd International Workshop on Computer Aided Modeling and Design of Communication
Links and Networks (CAMAD) (2017), pp. 1–6.
[2] Asgari, S., Jamali, S., Fotohi, R., and Nooshyar, M. Performance­aware placement and chaining scheme for virtualized network functions: a particle swarm
optimization approach. The Journal of Supercomputing (04 2021).
[3] Bari, M. F., Chowdhury, S. R., Ahmed, R., and Boutaba, R. On orchestrating
virtual network functions. In 2015 11th International Conference on Network
and Service Management (CNSM) (2015), pp. 50–56.
[4] Carpio, F., Jukan, A., and Pries, R. Balancing the migration of virtual network
functions with replications in data centers. 1–8.
[5] Cho, D., Taheri, J., Zomaya, A. Y., and Bouvry, P. Real­time virtual network
function (vnf) migration toward low network latency in cloud environments.
In 2017 IEEE 10th International Conference on Cloud Computing (CLOUD)
(2017), pp. 798–801.
[6] Cohen, R., Lewin­Eytan, L., Naor, J. S., and Raz, D. Near optimal placement
of virtual network functions. In 2015 IEEE Conference on Computer Communications (INFOCOM) (2015), pp. 1346–1354.
[7] Cplex, I. I. V12. 1: User’s manual for cplex. International Business Machines
Corporation 46, 53 (2009), 157.
[8] Eramo, V., Ammar, M., and Lavacca, F. G. Migration energy aware reconfigurations of virtual network function instances in nfv architectures. IEEE
Access 5 (2017), 4927–4938.
[9] et al, M. C. Network functions virtualization. [Online]. Available:
https://portal.etsi.org/nfv/nfv_white_paper.pdf.
[10] Farshin, A., and Sharifian, S. A modified knowledge­based ant colony algorithm for virtual machine placement and simultaneous routing of nfv in distributed cloud architecture. The Journal of Supercomputing 75 (08 2019).
[11] Gember­Jacobson, A., and Akella, A. Improving the safety, scalability, and
efficiency of network function state transfers. pp. 43–48.
[12] Gupta, A., Farhan Habib, M., Mandal, U., Chowdhury, P., Tornatore, M., and
Mukherjee, B. On service­chaining strategies using virtual network functions
in operator networks. Computer Networks 133 (2018), 1–16.
[13] Gupta, A., Habib, M. F., Chowdhury, P., Tornatore, M., and Mukherjee, B.
Joint virtual network function placement and routing of traffic in operator networks.
[14] Han, B., Gopalakrishnan, V., Ji, L., and Lee, S. Network function virtualization: Challenges and opportunities for innovations. IEEE Communications
Magazine 53, 2 (2015), 90–97.
[15] Houidi, O., Soualah, O., Louati, W., Mechtri, M., Zeghlache, D., and Kamoun,
F. An efficient algorithm for virtual network function scaling. In GLOBECOM
2017 ­ 2017 IEEE Global Communications Conference (2017), pp. 1–7.
[16] Isyaku, B., Mohd Zahid, M. S., Bte Kamat, M., Abu Bakar, K., and Ghaleb,
F. A. Software defined networking flow table management of openflow
switches performance and security challenges: A survey. Future Internet 12,
9 (2020).
[17] Lee, Y.­C., and Zomaya, A. Energy efficient utilization of resources in cloud
computing systems. The Journal of Supercomputing 60 (05 2010), 268–280.
[18] Li, D., Hong, P., Xue, K., and j. Pei. Virtual network function placement
considering resource optimization and sfc requests in cloud datacenter. IEEE
Transactions on Parallel and Distributed Systems 29, 7 (2018), 1664–1677.
[19] Manias, D., Jammal, M., Hawilo, H., Shami, A., Heidari, P., Larabi, A., and
Brunner, R. Machine learning for performance­aware virtual network function
placement. pp. 1–6.
[20] Mehraghdam, S., Keller, M., and Karl, H. Specifying and placing chains of virtual network functions. In 2014 IEEE 3rd International Conference on Cloud
Networking (CloudNet) (2014), pp. 7–13.
[21] Meng, Z., Bi, J., Sun, C., Xu, A., and Hu, H. Pram: Priority­aware flow
migration scheme in nfv networks. SOSR ’17, Association for Computing
Machinery, p. 183–184.
[22] Moens, H., and Turck, F. D. Vnf­p: A model for efficient placement of virtualized network functions. In 10th International Conference on Network and
Service Management (CNSM) and Workshop (2014), pp. 418–423.
[23] Orlowski, S., Pióro, M., Tomaszewski, A., and Wessäly, R. SNDlib 1.0–
Survivable Network Design Library. In Proceedings of the 3rd International
Network Optimization Conference (INOC 2007), Spa, Belgium (April 2007).
http://sndlib.zib.de, extended version accepted in Networks, 2009.
[24] Orlowski, S., Pióro, M., Tomaszewski, A., and Wessäly, R. SNDlib 1.0–
Survivable Network Design Library. Networks 55, 3 (2010), 276–286.
[25] Pei, J., Hong, P., Pan, M., Liu, J., and Zhou, J. Optimal vnf placement via
deep reinforcement learning in sdn/nfv­enabled networks. IEEE Journal on
Selected Areas in Communications 38, 2 (2020), 263–278.
[26] Quinn, P., and Nadeau, T. Problem statement for service function chaining
(rfc7498). [Online]. Available: https://rfc­editor.org/rfc/rfc7498.txt.
[27] Racheg, W., Ghrada, N., and Zhani, M. F. Profit­driven resource provisioning in nfv­based environments. In 2017 IEEE International Conference on
Communications (ICC) (2017), pp. 1–7.
[28] Ruiz, L., Barroso, R. J. D., De Miguel, I., Merayo, N., Aguado, J. C.,
De La Rosa, R., Fernández, P., Lorenzo, R. M., and Abril, E. J. Genetic algorithm for holistic vnf­mapping and virtual topology design. IEEE Access 8
(2020), 55893–55904.
[29] Strom, D., and van der Zwet, J. F. ”truth and lies about latency in the cloud,”
interxiontm white paper.
[30] Sun, P., Lan, J., Li, J., Guo, Z., and Hu, Y. Combining deep reinforcement
learning with graph neural networks for optimal vnf placement. IEEE Communications Letters 25, 1 (2021), 176–180.
[31] Sun, Q., Lu, P., Lu, W., and Zhu, Z. Forecast­assisted nfv service chain deployment based on affiliation­aware vnf placement. In 2016 IEEE Global Communications Conference (GLOBECOM) (2016), pp. 1–6.
[32] Tang, L., He, X., Zhao, P., Zhao, G., Zhou, Y., and Chen, Q. Virtual network
function migration based on dynamic resource requirements prediction. IEEE
Access 7 (2019), 112348–112362.
[33] Varasteh, A., De Andrade, M., Machuca, C. M., Wosinska, L., and Kellerer, W.
Power­aware virtual network function placement and routing using an abstraction technique. In 2018 IEEE Global Communications Conference (GLOBECOM) (2018), pp. 1–7.
[34] Wang, Y., Xie, G., Li, Z., He, P., and Salamatian, K. Transparent flow migration for nfv. In 2016 IEEE 24th International Conference on Network Protocols (ICNP) (2016), pp. 1–10.