近年來網路功能虛擬化(Network Function Virtualization, NFV) 因為其有不被專用硬體限制的特性，吸引了大量運營商的關注，運營商能在自己架設的伺服器上部屬各種的虛擬網路功能(Virtualized Network Function)，例如: 防火牆、網路監視器等等，因為不限制於專用硬體所以在部屬時能更彈性的使用資源。然而，運行在處理器核心上的多個虛擬化網路功能（VNF）會爭奪計算資源，導致性能下降。
在這篇論文中，我們提出了一種虛擬化網路功能（VNF）部署和遷移的演算法，無論是在同質核心架構或混合核心架構上，以防止核心被低效利用並實現負載平衡。為了防止核心低效利用，所有已使用的處理器核心中，除最多一個核心外，其餘核心的使用率都必須超過一定的使用百分比。為實現負載平衡，我們提出的演算法試圖最大化有效利用的核心數量。我們根據每個VNF 的流量強度預測CPU 核心的使用率。如果預測顯示特定的CPU 核心將超過其容量，我們的演算法會進行VNF 遷移的決策，使VNF 動態地在核心之間遷移。實驗測量和模擬結果均表明，我們提出的方法優於其他方法。

In recent years, Network Function Virtualization (NFV) has attracted a lot of operators’ attention because of its feature of not being restricted by dedicated hardware. Operators can deploy various virtualized network functions (e.g., firewall, network monitor, etc.) on their own servers and use resources more flexibly when deploying them because they are not restricted by dedicated hardware. However, multiple virtualized network functions (VNFs) running on the cores of a processor contend with each other for computational resources, resulting in performance degradation.
In this thesis, we propose an algorithm for VNF deployment and VNF migration, either on homogeneous cores or on hybrid cores, in order to prevent cores from underutilization and to achieve load balancing. To prevent core underutilization, all but at most one of the used processor cores has to exceed a given usage percentage. To achieve load balancing, our proposed algorithm attempts to maximize the number of cores that are properly utilized. We predict the CPU core usage based on traffic intensity of each VNF. If the prediction indicates that a specific CPU core will exceed its capacity, our algorithm makes decision for VNF migration, which dynamically migrate VNFs between cores. Both experiment measurement and simulation results show that our proposed method outperforms other methods.

[1] Intel, Intel Core i7-12700 Processor, https://www.intel.com/content/
www/us/en/products/sku/134591/intel- core- i712700- processor-
25m-cache-up-to-4-90-ghz/specifications.html, [Online; accessed
15-March-2023], 2022.
[2] ETSI, Network Functions Virtualisation (NFV); Architectural Framework,
https://www.etsi.org/deliver/etsi_gs/nfv/001_099/002/01.02.01_
60/gs_nfv002v010201p.pdf, [Online; accessed 16-March-2023], 2014.
[3] W. Zhang, J. Hwang, S. Rajagopalan, K. Ramakrishnan, and T. Wood,
“Flurries: Countless fine-grained nfs for flexible per-flow customization,” in
Proceedings of the 12th International on Conference on Emerging Networking
EXperiments and Technologies, 2016, pp. 3–17.
[4] C. Sun, J. Bi, Z. Zheng, H. Yu, and H. Hu, “Nfp: Enabling network function
parallelism in nfv,” in Proceedings of the ACM SIGCOMM Conference, Aug.
2017, pp. 43–56.
[5] A. Chowdhary and D. Huang, “Sdn based network function parallelism in
cloud,” in 2019 International Conference on Computing, Networking and
Communications (ICNC), 2019, pp. 486–490.
[6] D. R. Mafioletti, C. K. Dominicini, M. Martinello, M. R. N. Ribeiro, and
R. d. S. Villa¸ca, “Piaffe: A place-as-you-go in-network framework for flexible
embedding of vnfs,” in ICC 2020 - 2020 IEEE International Conference on
Communications (ICC), 2020, pp. 1–6.
[7] B. Deng, W. Wu, and L. Song, “Redundant logic elimination in network
functions,” in Proceedings of the Symposium on SDN Research, 2020, pp. 34–
40.
[8] A. Mohammadkhan, K. K. Ramakrishnan, and V. A. Jain, “Cleang—improving
the architecture and protocols for future cellular networks with nfv,” IEEE/ACM
Transactions on Networking, no. 6, pp. 2559–2572, 2020.
[9] S. G. Kulkarni, W. Zhang, J. Hwang, et al., “Nfvnice: Dynamic backpressure
and scheduling for nfv service chains,” IEEE/ACM Transactions on
Networking, vol. 28, no. 2, pp. 639–652, Feb. 2020.
[10] M. Ghaznavi, E. Jalalpour, B. Wong, R. Boutaba, and A. J. Mashtizadeh,
“Fault tolerant service function chaining,” in Proceedings of the Annual conference
of the ACM Special Interest Group on Data Communication (SIGCOMM)
on the applications, technologies, architectures, and protocols for
computer communication, Jul. 2020, pp. 198–210.
[11] M. Reininger, A. Arora, S. Herwig, et al., “Bento: Safely bringing network
function virtualization to tor,” in Proceedings of the 2021 ACM SIGCOMM
2021 Conference, 2021, pp. 821–835.
[12] J. Gil Herrera and J. F. Botero, “Resource allocation in nfv: A comprehensive
survey,” IEEE Transactions on Network and Service Management, pp. 518–
532, 2016.
[13] J. Li, W. Shi, Q. Ye, W. Zhuang, X. Shen, and X. Li, “Online joint vnf
chain composition and embedding for 5g networks,” in 2018 IEEE Global
Communications Conference (GLOBECOM), 2018, pp. 1–6.
[14] M. Savi, M. Tornatore, and G. Verticale, “Impact of processing-resource
sharing on the placement of chained virtual network functions,” IEEE Transactions
on Cloud Computing, pp. 1479–1492, 2021.
[15] P. Kr¨amer, P. Diederich, C. Kr¨amer, R. Pries, W. Kellerer, and A. Blenk,
“Sfc2cpu: Operating a service function chain platform with neural combinatorial
optimization,” in 2021 IFIP/IEEE International Symposium on Integrated
Network Management (IM), 2021.
[16] C. Chen, L. Nagel, L. Cui, and F. P. Tso, “B-scale: Bottleneck-aware vnf
scaling and flow routing in edge clouds,” in 2022 IEEE Symposium on Computers
and Communications (ISCC), 2022, pp. 1–6.
[17] D. Bhamare, A. Kassler, J. Vestin, M. A. Khoshkholghi, and J. Taheri,
“Intopt: In-band network telemetry optimization for nfv service chain monitoring,”
in ICC 2019 - 2019 IEEE International Conference on Communications
(ICC), 2019, pp. 1–7.
[18] J. Csirik, “Two simple algorithms for bin covering,” Acta Cybernetica, vol. 14,
no. 1, pp. 13–25, Jan. 1999.
[19] C.-F. Kuo, “Nfv performance prediction based on run-time instruction analysis,”
M.S. thesis, National Tsing Hua University, Jul. 2021.
[20] W. Zhang, G. Liu, W. Zhang, et al., “OpenNetVM: A Platform for High Performance
Network Service Chains,” in Proceedings of the ACM SIGCOMM
Workshop on Hot Topics in Middleboxes and Network Function Virtualization,
Aug. 2016.
[21] K. Wiles, The Pktgen Application, https://pktgen-dpdk.readthedocs.
io/en/latest, [Online; accessed 11-March-2023], 2010.
[22] C.-Y. Lee, “Resource allocation for nfv in hybrid multi-core architecture,”
M.S. thesis, National Tsing Hua University, Jul. 2022.
[23] Catherinemeadows, The Load Generator, https://github.com/sdnfv/
openNetVM/tree/master/examples/load_generator, [Online; accessed
27-April-2023], 2022.
[24] Catherinemeadows, The Simple Forward, https://github.com/sdnfv/
openNetVM/tree/master/examples/simple_forward, [Online; accessed
27-April-2023], 2022.
[25] Catherinemeadows, The AES encryption, https://github.com/sdnfv/
openNetVM/tree/master/examples/aes_encrypt, [Online; accessed 27-
April-2023], 2022.
[26] J. Zheng, C. Tian, H. Dai, et al., “Optimizing nfv chain deployment in
software-defined cellular core,” IEEE Journal on Selected Areas in Communications,
no. 2, pp. 248–262, 2020.