Aug 2026· Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication· 0 citations· 7 references
Computer Science
TL;DR
This work proposes a new logically centralized controller powered entirely by an LLM, called Agentic-Defined Networking (ADN), a novel architecture that integrates LLMs as the reasoning core of an SDN control plane implemented in a real network controller.
Abstract
Software-Defined Networking (SDN) has revolutionized network management by decoupling control logic from data forwarding. However, limited by the traditional controller paradigm, existing SDN controllers remain inherently static, relying on predefined rules. This rigidity makes them ill-equipped to handle unforeseen traffic patterns or emerging threats, often defaulting to generic actions that fail to address nuanced scenarios. Large-Language Models (LLMs), a group of models with billions of parameters that are trained on diverse datasets, are known to excel at performing complex tasks that require human-level reasoning and prior knowledge. With such powerful models assumed to encapsulate the collective knowledge of network operations within their parameters, one question that this work asks is "Are static networking rules provided by humans or by heuristics still relevant?" To answer this question, we propose a new logically centralized controller powered entirely by an LLM, called Agentic-Defined Networking (ADN). ADN introduces a novel architecture that integrates LLMs as the reasoning core of an SDN control plane implemented in a real network controller. To support the main thesis, we present preliminary results on ADN's performance on dealing with unseen malicious traffic and congestion-aware routing.
The study provides initial evidence of feasibility while identifying the challenges that must be addressed before production deployment and formalize the ADN agent model and workflow and define an operational framework covering communication, lifecycle management, governance, and security.
F. Rossi, Paulo Silas S. De Souza, Diogo M. Monteiro et al.· IEEE Access· 0 citations
As Software-Defined Networking (SDN) and Network Function Virtualization (NFV) enabled networks scale in size and complexity, monitoring and managing Service Function Chains (SFCs) under stringent latency and resource constraints becomes increasingly challenging. Although Deep Reinforcement Learning (DRL) is widely applied to SFC provisioning and Virtual Network Function (VNF) placement, enhanced network state monitoring is crucial to capture unexpected network conditions and guide DRL agents toward more adaptive decisions. In this context, Language Models (LMs) enable flexible, natural-language (NL)–based, query-driven network monitoring; however, directly processing complex multi-metric NL queries is computationally expensive and error-prone. This paper proposes an end-to-end (E2E) edge-based query translation pipeline that decomposes multi-metric NL queries into simpler single-metric sub-queries. Query decomposition is performed using a retrieval-augmented language model (RAG-LLM) and compared with a lightweight rule-based decomposition baseline. The resulting sub-queries are translated into Structured Query Language (SQL) using FLAN-T5. A cloud-only baseline, which directly translates NL queries to SQL without decomposition, is also evaluated. The results show that the rule-based edge pipeline achieves the lowest latency, reducing E2E latency by up to 78% compared to RAG-LLM and 18% compared to cloud execution under high workloads. Under increasing arrival rates for the largest workload, the rule-based edge pipeline maintains superior performance over cloud, reducing total E2E latency by 57% at $\lambda = 0.8$ . While RAG-LLM provides greater flexibility for unseen query patterns, both edge-based approaches achieve 100% NL2SQL accuracy with zero decomposition failures, outperforming the cloud-only baseline (95% accuracy).
Parisa Fard Moshiri, Xinyu Zhu, Poonam Lohan et al.· IEEE Transactions on Network...· 0 citations
Findings affirm that the suggested scalable control plane is practical in supporting large scale SDN implementation and is therefore applicable in future carrier grade, data center and wide area network deployments at realistic workloads with varying topological setups in the modern programmable networks in the world.
A. Nagadeepan, Vishakha Abhay Gaidhani, Bhambare Rajesh et al.· Journal of Intelligent Decis...· 0 citations
Periplus, an in-band SDN control plane designed to address four challenges that this approach presents in such contexts: automatic bootstrapping, source-based routing, fast failure recovery, and multi-controller coordination is presented.
E. M. Castro Barbero, P. de las Heras Quirós, F. J. Simó Reigadas· arXiv.org· 1 citation· ⚡1
Data Centers that implement Software-Defined Networks (SDN) are not required to employ the Address Resolution Protocol (ARP), but network hosts do. Therefore, there is a need to support this protocol without modifying the intrinsic functionality of the SDN controller. In this work, four strategies for handling ARP are evaluated using an SDN and OpenFlow rules. The strategies include disabling ARP at the host level, using static MAC addresses, introducing a fake gateway, and generating ARP replies using OpenFlow flows. To our knowledge, nobody has tested and compared the main characteristics and advantages offered by these four strategies. Experimental evaluation was conducted on a real SDN network and complemented with similar experiments using Mininet. Performance was assessed using metrics such as ping response time, address resolution response time, jitter, and packet loss ratio. The results show that OpenFlow-based ARP replies provide a good balance in terms of scalability, performance, and configuration effort. This strategy achieved the lowest average ping response time (0.641 ms) and ARP response time (0.6188 ms), while avoiding the manual configuration requirements of static approaches.
Jose Neftali Limon-Ortiz, P. D. Arjona-Villicaña, A. G. Silva-Trujillo et al.· Network· 0 citations
A reference architecture is developed that separates proposal generation from governed execution, identifies recurring integration patterns for LLM-enabled operations, and derives a research agenda for higher network autonomy under explicit assurance, safety, and governance constraints.
Tianzhu Zhang, Changgang Zheng, Shanshan Wang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
