EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FlowSlice-P4: Service-Aware Data-Plane Decision Making for Traffic Slicing in Programmable SD-IoT Networks

Thamer Alrefai

Journal of Intelligent Decision Making and Information Science · 16.09.2026 · 3 (12s) · 1863-1890

DOI: 10.17762/jidmis.v3.3594 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 16.09.2026. Verfasst von Thamer Alrefai. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The increasing scale and service heterogeneity of Software-Defined Internet of Things (SD-IoT) networks create a critical decision-making challenge: incoming traffic must be rapidly identified, associated with an appropriate logical service context, and forwarded accordingly without imposing excessive processing on the centralized SDN controller. This paper proposes FlowSlice-P4, a lightweight service-aware data-plane decision-making framework for programmable SD-IoT traffic slicing. FlowSlice-P4 establishes an in-network decision process that transforms observable traffic information into an immediately enforceable forwarding decision through the sequence Traffic Signature, Service Context, Context-Aware Forwarding. The framework employs a Service Mapping Engine (SME) to evaluate protocol, transport, application, and device-related attributes using deterministic P4 match-action processing and select the corresponding Service Context (SC). The resulting context is then consumed by a Local Slice Forwarder (LSF) to determine the appropriate logical forwarding treatment. A local-first, controller-on-miss strategy enables established service decisions to be completed within the programmable switch, while the SDN controller retains responsibility for policy management, updates, and unresolved conditions. Experimental evaluation demonstrates an overall service-mapping accuracy of 99.08% and forwarding correctness of 99.14%. At the largest evaluated scale of 1,000 IoT devices and 5,000 concurrent flows, FlowSlice-P4 maintains 98.96% mapping accuracy, 99.02% forwarding correctness, and 0.55 ms decision latency. Compared with controller-based slicing, the framework reduces decision latency by approximately 94.7%, locally resolves 94.62% of traffic, and decreases controller CPU utilization by approximately 58.9%. These results demonstrate that coupling service identification with deterministic in-network decision making provides an effective approach to scalable SD-IoT traffic slicing, enabling accurate and low-latency forwarding decisions while substantially reducing routine control-plane dependence.

Zitieren

APA

Alrefai, T. (2026). FlowSlice-P4: Service-Aware Data-Plane Decision Making for Traffic Slicing in Programmable SD-IoT Networks. Journal of Intelligent Decision Making and Information Science, 3(12s), 1863-1890. https://doi.org/10.17762/jidmis.v3.3594

Vancouver

Alrefai T. FlowSlice-P4: Service-Aware Data-Plane Decision Making for Traffic Slicing in Programmable SD-IoT Networks. Journal of Intelligent Decision Making and Information Science. 2026;3(12s):1863-1890. doi:10.17762/jidmis.v3.3594

BibTeX
@article{Alrefai2026,
  author = {Alrefai, Thamer},
  title = {FlowSlice-P4: Service-Aware Data-Plane Decision Making for Traffic Slicing in Programmable SD-IoT Networks},
  journal = {Journal of Intelligent Decision Making and Information Science},
  year = {2026},
  volume = {3},
  number = {12s},
  pages = {1863--1890},
  doi = {10.17762/jidmis.v3.3594}
}
Kostenlos starten →