NEXUS: A Microservice-Based Network Emulation Platform with Topology Isolation, Flow-Based Analytics, and AI-Driven Control
Bildiriler > Tam Metin Bildiri
Yazarlar: Ege Erberk USLU, Orhan DAĞDEVİREN
2026 8th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA), , , Uluslararası, Tam Metin Bildiri, ss 1-10
DOI: 10.1109/ichora69329.2026.11536976
Özet
Network emulation platforms constitute fundamental infrastructure for prototyping and evaluating distributed systems, routing protocols, Software-Defined Networking (SDN) control logic, and network security mechanisms. However, existing emulation workflows encounter significant challenges when experiments require frequent runtime topology modifications, heterogeneous protocol stacks, flow-based analytics with robust observability, and integration with NFV Management and Orchestration (MANO) frameworks. This paper presents NEXUS (Network EXperimentation Unified System), a microservicebased network emulation platform that decouples the privileged emulation engine from a containerized control plane. The platform persists topologies as database-backed constructs, instantiates per-topology emulation containers on demand, and exposes unified control through an API gateway. Our architecture, comprising 21 microservices, supports concurrent execution of multiple isolated topologies, CRIU-based checkpoint/restore operations, and per-topology observability stacks. We systematically identify 19 functionalities for network emulation platforms and demonstrate through comparative analysis that NEXUS is, to the best of our knowledge, the first system to consolidate all of them, including nine advanced functionalities F11 through F19 that have not been consolidated in a single emulation platform in prior work. We additionally report a single-host experimental evaluation measuring topology instantiation up to 100 nodes in 32.06 s, runtime modification latency of 91.5 ms for link addition and 260.8 ms for node addition, and concurrent scalability limits that motivate future distributed deployment.