EUVIMEDEUVIMED entdecken
Crossref

Evaluating Attack Detection and Prevention in SDN NFV Networks: A Comprehensive Security Solution Analysis

Varun Venkatesh Dandasi

Journal of Artificial intelligence and Machine Learning · 2023

DOI: 10.55124/jaim.v1i3.282 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Artificial intelligence and Machine Learning, 2023. Verfasst von Varun Venkatesh Dandasi.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study delves into the intricate security and privacy challenges inherent in Software-Defined Networking (SDN) and Network Functions Virtualization (NFV) ecosystems, highlighting an urgent imperative for robust security solutions in contemporary network architectures. As the Internet of Things (IoT) proliferates and the complexity of networks escalates, the integration of effective security measures emerges as an essential safeguard for protecting network infrastructures against the relentless tide of evolving cyber threats. The significance of this research lies in its comprehensive evaluation and comparative analysis of six leading security solutions—OpenFlow, FortNOX, FRESCO, Rosemary, VeriFlow, and SE-Floodlight—offering network administrators and security professionals critical quantitative insights to facilitate informed decision-making. Employing the Weighted Sum Method (WSM) as its primary research methodology, the study presents a systematic framework for multi-criteria decision analysis specifically tailored to network security implementations. The evaluation criteria encompass a range of parameters, including the attack detection rate, attack prevention rate, network overhead, false positive and false negative rates, and encryption latency. Furthermore, alternative metrics such as network scalability, resource utilization, and implementation complexity are considered across diverse network environments. To ensure an objective comparison of all solutions, the research methodology utilizes normalized matrices coupled with weighted calculations, thereby enhancing the rigor and depth of the analysis. The results of the analysis illuminate that VeriFlow not only excels but also establishes a benchmark with an impressive preference score of 0.8907. This remarkable system showcases formidable capabilities in attack detection, boasting a rate of 90%, alongside a commendable prevention rate of 85%. However, this prowess comes at a slight cost, as evidenced by an increased encryption latency of 20 ms. In contrast, FortNOX emerges as a strong contender, securing the second position with a score of 0.7878. This system strikes a commendable balance, delivering a robust suite of security features while maintaining a moderate latency of 15 ms. Meanwhile, Rosemary trails in third place with a score of 0.7260, reflecting its comparatively lesser efficacy. The findings unveil a discernible relationship between the effectiveness of security measures and the concomitant system overhead. This interplay suggests that optimal security solutions frequently necessitate a delicate equilibrium—wherein enhanced protection levels inevitably lead to trade-offs concerning network performance. Ultimately, the study posits that while high-security implementations may engender additional latency and overhead, the exemplary cases presented by VeriFlow and FortNOX underscore the feasibility of achieving formidable security without severely compromising network functionality. These revelations significantly enrich the existing corpus of knowledge in the realm of SDN NFV security, offering valuable insights for organizations striving to bolster their network infrastructure’s security posture.

Zitieren

APA

Dandasi, V. V. (2023). Evaluating Attack Detection and Prevention in SDN NFV Networks: A Comprehensive Security Solution Analysis. Journal of Artificial intelligence and Machine Learning. https://doi.org/10.55124/jaim.v1i3.282

Vancouver

Dandasi VV. Evaluating Attack Detection and Prevention in SDN NFV Networks: A Comprehensive Security Solution Analysis. Journal of Artificial intelligence and Machine Learning. 2023. doi:10.55124/jaim.v1i3.282

BibTeX
@article{Dandasi2023,
  author = {Dandasi, Varun Venkatesh},
  title = {Evaluating Attack Detection and Prevention in SDN NFV Networks: A Comprehensive Security Solution Analysis},
  journal = {Journal of Artificial intelligence and Machine Learning},
  year = {2023},
  doi = {10.55124/jaim.v1i3.282}
}
Kostenlos starten →