28 June 2018 in Security and Communication Networks
This paper presents VISKA, a cloud security service for dynamically detecting malicious switching elements in software defined networking (SDN) infrastructures. The main contributions of VISKA lie in (1) utilizing network programming and secure probabilistic sketching in SDN environments to dynamically detect and isolate parts of the data plane that experience malicious behavior, (2) applying a set of focused packet probing and sketching mechanisms on isolated network partitions/views rather than focusing the security mechanisms on the whole physical network, (3) efficiently analyzing the network behavior of the resulting views by recursively partitioning them in a divide-and-conquer fashion to logarithmically reduce the problem size in order to localize abnormal/malicious switching units, and (4) providing an attack categorization module that analyzes live ingress/egress traffic of the maliciously detected switch(es) solely to identify the specific type of attack, rather than inspecting the whole network traffic as is done in traditional intrusion detection systems. This significantly enhances the performance of attack detection and reduces the load on the controller. A testbed prototype implementation is realized on the Mininet network emulator. The experimental analysis corroborated the algorithms’ convergence property using the linear and FatTree topologies with network sizes of up to 250 switches. Moreover, an implementation of the attack categorization module is realized and achieved an accuracy rate of over 90% for the different attack types supported.
24 February 2025 in IEEE Transactions on Dependable and Secure Computing
This paper presents DCGUARD, a unified security approach for detecting and isolating misbehaving computing and forwarding nodes in multi-tenant virtualized cloud data centers. DCGUARD employs technological advancements in Virtual Machine Introspection (VMI), Software-Defined Networking (SDN), and secure probabilistic sketching to detect and isolate parts of the Virtual Machines (VMs) and network switches experiencing malicious behavior dynamically. The main contribution lies in designing a divide-and-conquer strategy that utilizes VMI and network programmability to apply focused distributed task and packet probing mechanisms on portions of the data center network rather than focusing the security functions on the entire physical network. The processing VMs and network switches are recursively partitioned into independent logical groups inspected individually to localize abnormal/malicious computing and switching nodes incrementally. This remarkably enhances the efficiency of the detection mechanisms, which opportunistically approaches a logarithmic time complexity in the number of protocol steps towards convergence (compared to a linear time complexity in traditional intrusion detection systems) when a relatively low number of hostile VMs and switches are present. Real experiments are evaluated, and a test-bed blueprint of the proposed design is emulated in a virtualized cloud environment using the Mininet emulator. The performance, convergence, and accuracy benchmarks corroborate the analytical advantage of the proposed security approach.