SDN controllers become the main advantage of the architecture because they present a centralized control decision-making and general view of the network. They are, however, also a critical point that an attacker could exploit. More review of the body of research is needed regarding the types of attacks on SDN controllers, methods to detect them, and mitigation techniques directed specifically to the controller, particularly considering the approach of machine learning detection methods. This survey addresses the topics of attacks targeting the SDN controller, methods for their detection, what types of controllers are used in different studies, and datasets used in machine learning detection methods. The findings highlight that most attacks exploit vulnerabilities inherent in the OpenFlow protocol, while the detection methodologies remain primarily statistical and machine learning approaches. Additionally, the review shows that while outdated controllers like Floodlight and Ryu are still widely used in studies, actively supported controllers such as ONOS and ODL are used much less. Finally, the survey finds only two publicly available datasets tailored for SDN environments, none considering attacks directed at the controllers, illustrating a notable gap in the existing research. This survey also highlights the need for further research focusing on modern SDN controllers and developing comprehensive datasets to advance effective security solutions.
Software-defined networks (SDN) is an emerging paradigm that has been widely explored by the research community. At the same time, it has attracted a lot of attention from the industry. SDN breaks the integration between control and data plane and creates the concept of a network operating system (controller). The controller should be logically centralized, but it must comply with availability, reliability, and security requirements, which implies that it should be physically distributed in the network. In this context, two questions arise: How many controllers should be included? and Where should they be located? These questions comprise the controller placement problem (CPP). The scope of this study is to solve the CPP using the meta-heuristic Tabu search algorithm to optimize the cost of network operation, considering flow setup latency and inter-controller latency constraints. The network model presented considers both controllers and links as IT resources as a service, which allows focusing on operational cost.
The integration of IT services is a critical challenge for public organizations that seek to modernize their operational ecosystems and strengthen mission-oriented processes. In the field of fiscal oversight, supreme audit institutions (SAIs) increasingly require systematized and interoperable service architectures to ensure transparency, accountability, and effective public resource control. However, existing literature reveals persistent gaps concerning how service integration models can be deployed and validated within complex government environments. This study describes an enterprise architecture-driven service integration model designed and evaluated within the Office of the General Comptroller of the Republic of Colombia (Contraloría General de la República, CGR). The study tests the hypothesis that an Enterprise Architecture-driven integration model provides the necessary structural coupling to align technical IT performance with the legal requirements of fiscal oversight, which is an alignment that typically does not appear in generic governance frameworks. The methodological approach followed in this study combines an IT service management maturity assessment, process analysis, architecture repository review, and iterative validation sessions with institutional stakeholders. The model integrates ITILv4 (Information Technology Infrastructure Library), TOGAF (The Open Group Architecture Framework), COBIT (Control Objectives for Information and Related Technologies), and ISO20000 into a coherent framework tailored to the operational and regulatory requirements of an SAI. Results show that the proposed model reduces service fragmentation, improves process standardization, strengthens information governance, and enables a unified service catalog aligned with fiscal oversight functions. The empirical validation demonstrates measurable improvements in service delivery, transparency, and organizational responsiveness. The study contributes to the field of applied system innovation by: (i) providing an integration model, which is scientifically grounded and evidence-based, (ii) demonstrating how hybrid governance and architecture frameworks can be adapted to complex public-sector environments, and (iii) offering a replicable approach for SAIs that seek to modernize their technological service ecosystems through enterprise architecture principles. Future research directions are also discussed to provide guidelines to advance integrated governance and digital transformation in oversight institutions.
Software-Defined Networking (SDN) separates the control and data planes, introducing a logically centralized controller that is itself a high-value attack target. Despite growing interest in SDN intrusion detection, publicly available datasets either restrict evaluation to binary normal-vs-DDoS classification or lack control-plane telemetry, leaving multi-class detection of SDN-architectural attacks without a dedicated benchmark. This work presents LAN-SDN-NIDS, a publicly available, multi-class flow-level dataset of 1,125,059 records generated in a fully containerized Containernet/OpenDaylight testbed across five standard network topologies. Each flow record combines 29 traffic-level features with 11 control-plane-aware metrics—including Packet-In and Flow-Mod counts and first-seen delay. The dataset covers five attack classes in two categories: three that exploit SDN control-plane mechanisms (link fabrication, host injection, and port hijack) alongside DDoS and port scan, plus normal traffic. An XGBoost classifier trained on the full feature set achieved a macro F1 of 0.94; an ablation study showed that removing OpenFlow features causes link fabrication F1 to collapse from 0.97 to 0.19, indicating that control-plane telemetry is decisive for detecting SDN-architectural attacks under the conditions evaluated. A UMAP embedding is consistent with class separability, except for a structural overlap between host injection and normal traffic attributable to their shared ARP protocol.