Multi-Controller Placement Optimization in SDNs Using Enhanced Density-Based Clustering Techniques

Document Type : Research Article

Author

Department of Computer Engineering, Payame Noor University (PNU), Tehran, Iran.

10.22108/jcs.2025.142694.1149

Abstract

One of the major challenges in large-scale Software-Defined Networks (SDN) is determining the optimal placement and number of controllers. Existing methods often face issues such as high latency, localization problems, and parameter configuration complexity. While density-based clustering algorithms like DBSCAN offer several advantages, they also encounter challenges, such as parameter tuning and noise caused by unnecessary distance computations. This paper introduces an enhanced algorithm named FIDBSCAN (Fast and Improved Density-Based Clustering with Noise) as an innovative approach to solving the controller placement problem. The algorithm accelerates the neighborhood search process by early pruning of noisy points and automatic input parameter determination, resulting in effective clustering for controller placement. The objective of this algorithm is to minimize the average propagation delays and the worst-case delay between controllers and switches. To evaluate the proposed method, experiments were conducted on real-world topologies, specifically Viatel and TW Telecom, obtained from the Internet Topology Zoo database. The results demonstrated that on the Viatel network, the proposed algorithm achieved a precision of 1.0, a recovery rate of 0.97, and an F-measure of 0.98, outperforming algorithms such as OPTICS, GDBSCAN, DBSCAN, DDSC, and ODBSCAN. Additionally, the clustering error rate of the proposed algorithm was 0.01, and it identified the optimal number of clusters (5 clusters) in both the Viatel and TW Telecom networks. For statistical analysis, ANOVA and Tukey tests were employed. The ANOVA test revealed significant differences in the performance of the algorithms across all metrics (Precision, Recovery Rate, F-Measure, and Clustering Error) with F=24.6 and P=0.001. The Tukey test further identified that the FIDBSCAN algorithm significantly outperformed the other algorithms in all metrics. These results, combined with high accuracy and low error rates, validate the findings and establish FIDBSCAN as an ideal solution for addressing the controller placement problem in SDN networks.

Keywords

Main Subjects

References

[1] Deep, G. and Tripathi, V. and Dumka, A.. A review on controller placement problem in software defined networking. AIP Conference Proceedings. 2521(1): 1--8, AIP Publishing. 2023. [DOI ]
[2] Rashid, S. J. and Alkababji, A. M. and Khidhir, A. S.. Performance evaluation of software-defined networking controllers in wired and wireless networks. TELKOMNIKA (Telecommunication Computing Electronics and Control). 21(1): 49--59, 2023. [DOI ]
[3] Ahmed, M. R. and Shatabda, S. and Islam, A. M. and Robin, M. T.. Intrusion Detection System in Software-Defined Networks Using Machine Learning and Deep Learning Techniques--A Comprehensive Survey. Authorea Preprints. 2023. [DOI ]
[4] Taha, M.. An efficient software defined network controller based routing adaptation for enhancing QoE of multimedia streaming service. Multimedia Tools and Applications. 82(22): 33865--33888, 2023. [DOI ]
[5] Chowdhury, S. and Helian, N. and de Amorim, R. C.. Feature weighting in DBSCAN using reverse nearest neighbours. Pattern Recognition. 137: 109314, 2023. [DOI ]
[6] Ghamkhar, H. and Ghazizadeh, M. J. and Mohajeri, S. H. and Moslehi, I. and Yousefi-Khoshqalb, E.. An unsupervised method to exploit low-resolution water meter data for detecting end-users with abnormal consumption: Employing the DBSCAN and time series complexity. Sustainable Cities and Society. 94: 104516, 2023. [DOI ]
[7] . The Internet Topology Zoo Database. . http://www.topology-zoo.org. : 2023.
[8] Singh, A. K. and Srivastava, S. and Banerjea, S.. Evaluating heuristic techniques as a solution of controller placement problem in SDN. Journal of Ambient Intelligence and Humanized Computing. 14(9): 11729--11746, 2023. [DOI ]
[9] Kanodia, K. and Mohanty, S. and Kurroliya, K. and Sahoo, B.. CCPGWO: A meta-heuristic strategy for link failure aware placement of controller in SDN. 2020 International Conference on Inventive Computation Technologies (ICICT). 859--863, IEEE. 2020. [DOI ]
[10] Salam, R. and Bhattacharya, A.. Efficient greedy heuristic approach for fault-tolerant distributed controller placement in scalable SDN architecture. Cluster Computing. 25(6): 4543--4572, 2022. [DOI ]
[11] Li, C. and Jiang, K. and Luo, Y.. Dynamic placement of multiple controllers based on SDN and allocation of computational resources based on heuristic ant colony algorithm. Knowledge-Based Systems. 241: 108330, 2022. [DOI ]
[12] He, D. and Chen, J. and Qiu, X.. A density algorithm for controller placement problem in software defined wide area networks. The Journal of Supercomputing. 79(5): 5374--5402, 2023. [DOI ]
[13] Ibrahim, A. A. and Hashim, F. and Sali, A. and Noordin, N. K. and Fadul, S. M.. A modified genetic algorithm for controller placement problem in SDN distributed network. 2021 26th IEEE Asia-Pacific Conference on Communications (APCC). 83--88, IEEE. 2021. [DOI ]
[14] Radam, N. S. and Al-Janabi, S. T. and Jasim, K. S.. Multi-controllers placement optimization in SDN by the hybrid HSA-PSO algorithm. Computers. 11(7): 111, 2022. [DOI ]
[15] Keshari, S. K. and Kansal, V. and Kumar, S.. An intelligent way for optimal controller placements in software-defined IoT networks for smart cities. Computers \& Industrial Engineering. 162: 107667, 2021. [DOI ]
[16] Benoudifa, O. and Wakrime, A. A. and Benaini, R.. Autonomous solution for controller placement problem of software-defined networking using MuZero based intelligent agents. Journal of King Saud University-Computer and Information Sciences. 35(10): 101842, 2023. [DOI ]
[17] Singh, G. D. and Tripathi, V. and Dumka, A. and Rathore, R. S. and Bajaj, M. and Escorcia-Gutierrez, J. and Prokop, L.. A novel framework for capacitated SDN controller placement: Balancing latency and reliability with PSO algorithm. Alexandria Engineering Journal. 87: 77--92, 2024. [DOI ]
[18] Zobary, F.. Optimizing SDN Controller to Switch Latency for Controller Placement Problem. Informatica. 48(8): 2024. [DOI ]
[19] Frdiesa, M.. A Controller Placement Algorithm Using Ant Colony Optimization in Software-Defined Network. International Journal of Wireless Information Networks. 31(2): 142--154, 2024. [DOI ]
[20] Zadedehbalaei, A. and Bagheri, A. and Afshar, H.. A study on DBSCAN Clustering algorithm issues and a survey on its improvements. Soft Computing Journal. 6(1): 2--37, 2021.
[21] Adekoya, O. and Aneiba, A.. A Stochastic Computational Graph with Ensemble Learning Model for solving Controller Placement Problem in Software-Defined Wide Area Networks. Journal of Network and Computer Applications. 225: 103869, 2024. [DOI ]
[22] Alouache, L. and Yassa, S. and Ahfir, A.. A multi-objective optimization approach for SDVN controllers placement problem. 2022 13th International Conference on Network of the Future (NoF). 1--9, IEEE. 2022. [DOI ]
[23] Al Samara, M. and Bennis, I. and Abouaissa, A. and Lorenz, P.. Complete outlier detection and classification framework for WSNs based on OPTICS. Journal of Network and Computer Applications. 211: 103563, 2023. [DOI ]
[24] Monko, G. and Kimura, M.. Optimized DBSCAN Parameter Selection: Stratified Sampling for Epsilon and GridSearch for Minimum Samples. Computer Science \& Information Technology (CS \& IT). 43--61, 2023. [DOI ]
[25] Rus, A. M. M. and Othman, Z. A. and Bakar, A. A. and Zainudin, S.. A Hierarchical ST-DBSCAN with Three Neighborhood Boundary Clustering Algorithm for Clustering Spatio--temporal Data. International Journal of Advanced Computer Science and Applications. 13(12): 2022. [DOI ]
[26] Li, X. and Liu, Q.. DDSC-SMOTE: an imbalanced data oversampling algorithm based on data distribution and spectral clustering. The Journal of Supercomputing. 1--30, 2024. [DOI ]
[27] Li, J. and Zheng, A. and Guo, W. and Bandyopadhyay, N. and Zhang, Y. and Wang, Q.. Urban flood risk assessment based on DBSCAN and K-means clustering algorithm. Geomatics, Natural Hazards and Risk. 14(1): 2250527, 2023. [DOI ]
[28] Dholakiya, D. and Kshirsagar, T. and Nayak, A.. Survey of mininet challenges, opportunities, and application in software-defined network (sdn). Information and Communication Technology for Intelligent Systems: Proceedings of ICTIS 2020, Volume 2. 213--221, 2021. [DOI ]
[29] Niazai, S. and Rahimzai, A. A. and Atifnigar, H.. Applications of MATLAB in natural sciences: a comprehensive review. Eur J Theoret Appl Sci. 1(5): 1006--15, 2023. [DOI ]
[30] Karthika, P. and Arockiasamy, K.. Simulation of SDN in Mininet and detection of DDoS attack using machine learning. Bulletin of Electrical Engineering and Informatics. 12(3): 1797--1805, 2023. [DOI ]
[31] Chay, Z. E. and Lee, C. H. and Lee, K. C. and Oon, J. S. and Ling, M. H.. Russel and Rao coefficient is a suitable substitute for Dice coefficient in studying restriction mapped genetic distances of Escherichia coli. arXiv preprint. 2023. [DOI ]
[32] Titz, J. . The relationship between the phi coefficient and the unidimensionality index H: Improving psychological scaling from the ground up. Psychological Methods. 2025. [DOI ]
[33] Sundqvist, M. and Chiquet, J. and Rigaill, G. . Adjusting the adjusted rand index: a multinomial story. Computational Statistics. 38(1): 327--347, 2023. [DOI ]
[34] Wang, Y. and Zhang, Q. and Liu, M.. Analysis of variance. Textbook of Medical Statistics: For Medical Students. 99--124, 2024. [DOI ]
[35] Ravichandran, C. and Padmanaban, G.. A numerical simulation-based method to predict floor wise distribution of cooling loads in Indian residences using Tukey honest significant difference test. Advances in Building Energy Research. 17(1): 1--29, 2023. [DOI ]
Journal of Computing and Security
Volume 11, Issue 2 - Serial Number 2
September 2024
Pages 61-83

Files

History

  • Receive Date: 02 September 2024
  • Revise Date: 22 February 2025
  • Accept Date: 06 April 2025

Share

How to cite

Statistics