Home / Current Issue / Paper 1707731
Topo Guard Mitigating Virtual Machine Poisoning Attacks in Software-Defined Networking
Subject area: Science,Engineering and Technology · Area of research: Network Security
Abstract
A novel networking model referred to as software-defined networking, or SDN, provides a controller and applications with the ability to view the entire network and to design it in a programmable manner. This provides an opportunity for innovative advancements in network protocols and applications. Most SDN applications rely on the logically centralized control plane of SDN to provide the entire network visibility, which is one of its key advantages. For the first time in the literature, we provide new attack vectors unique to SDN that significantly erode this foundation. Our new attacks share some ?lan with spoofing attacks in traditional networks (such as the ARP poisoning attack), but they differ significantly in that they exploit some vulnerabilities that result from how modern SDN operates differently from traditional networks. The virtual machine knowledge, which is an essential building block for both topology-aware SDN applications and core SDN components, can be efficiently poisoned by successful attacks.
Keywords
SDN, DDOS Attacks, Machine Learning
How to cite this paper
@article{1707731,
author = {Praneish R, Niranjan S, Navanitha Krishna R A},
title = {Topo Guard Mitigating Virtual Machine Poisoning Attacks in Software-Defined Networking},
journal = {Iconic Research And Engineering Journals},
year = {2025},
volume = {8},
number = {10},
pages = {106-112},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/},
abstract = {A novel networking model referred to as software-defined networking, or SDN, provides a controller and applications with the ability to view the entire network and to design it in a programmable manner. This provides an opportunity for innovative advancements in network protocols and applications. Most SDN applications rely on the logically centralized control plane of SDN to provide the entire network visibility, which is one of its key advantages. For the first time in the literature, we provide new attack vectors unique to SDN that significantly erode this foundation. Our new attacks share some ?lan with spoofing attacks in traditional networks (such as the ARP poisoning attack), but they differ significantly in that they exploit some vulnerabilities that result from how modern SDN operates differently from traditional networks. The virtual machine knowledge, which is an essential building block for both topology-aware SDN applications and core SDN components, can be efficiently poisoned by successful attacks.},
keywords = {SDN, DDOS Attacks, Machine Learning},
month = {April},
}