The rapid proliferation of intelligent services across Internet of Things (IoT), edge computing, and software-defined networking ecosystems has fundamentally transformed the computational landscape of modern digital infrastructure. This transformation has been driven by the convergence of deep learning, large language models, programmable networks, and heterogeneous hardware architectures. However, the exponential growth in model size, data traffic, and service heterogeneity has exposed critical challenges related to scalability, latency, energy efficiency, privacy, and deployability, particularly in resource-constrained environments. This research article presents an extensive theoretical and system-level investigation into the integration of efficient large language models, RISC-V-based hardware acceleration, and software-defined networking paradigms as a unified foundation for next-generation edge-intelligent systems.

Drawing strictly from the provided body of literature, this work synthesizes advances in model compression techniques such as pruning, quantization, and distillation for large language models; emerging instruction set architecture extensions for mixed-precision and packed-SIMD execution on RISC-V cores; and the evolution of software-defined networking from OpenFlow-based control to fully programmable data planes using P4. These strands are examined within the broader context of edge computing, mobile networks, and IoT service orchestration, highlighting how intelligent workloads can be dynamically deployed, optimized, and managed across distributed infrastructures.

The article develops a detailed methodological framework that conceptually integrates adaptive structured pruning, mixed-precision inference, and edge-aware orchestration with software-defined control planes. The results are presented as a descriptive synthesis of expected performance, efficiency, and scalability outcomes, emphasizing how such integration enables low-latency inference, energy-aware computation, and privacy-preserving data processing at the network edge. A deep discussion follows, critically examining theoretical implications, architectural trade-offs, regulatory considerations, and open research challenges. The study concludes by positioning edge-intelligent, software-defined, RISC-V-accelerated systems as a cornerstone for future IoT and networked intelligence, while identifying pathways for sustained innovation.

Edge Computing, Large Language Model, Model Compression

Agrawal, R., Kumar, H. and Lnu, S.R. (2025). Efficient LLMs for Edge Devices: Pruning, Quantization, and Distillation Techniques. 2025 International Conference on Machine Learning and Autonomous Systems, 1413–1418.

Ali, M., Aliagha, E., Elnashar, M. and Göhringer, D. (2025). P-CORE: Exploring RISC-V Packed-SIMD Extension for CNNs. IEEE Access, 13, 146603–146616.

An, Y., Zhao, X., Yu, T., Tang, M. and Wang, J. (2024). Fluctuation-Based Adaptive Structured Pruning for Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 38(10), 10865–10873.

Armeniakos, G., Maras, A., Xydis, S. and Soudris, D. (2025). Mixed-precision Neural Networks on RISC-V Cores: ISA extensions for Multi-Pumped Soft SIMD Operations. Proceedings of the IEEE/ACM International Conference on Computer-Aided Design.

Arriola, M., Gokaslan, A.K., Chiu, J.T., Yang, Z., Qi, Z., Han, J., Sahoo, S.S. and Kuleshov, V. (2025). Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models. International Conference on Learning Representations.

Amin, R., Rojas, E., Aqdus, A., Ramzan, S., Casillas-Perez, D. and Arco, J.M. (2021). A survey on machine learning techniques for routing optimization in SDN. IEEE Access, 9, 104582–104611.

Bengio, Y., Ducharme, R., Vincent, P. and Jauvin, C. (2003). A neural probabilistic language model. Journal of Machine Learning Research, 3, 1137–1155.

Cisco (2016). Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update (2017–2022).

Collobert, R. and Weston, J. (2008). A unified architecture for natural language processing: Deep neural networks with multitask learning. Proceedings of the International Conference on Machine Learning, 160–167.

Huang, W., Song, G., Hong, H. and Xie, K. (2014). Deep architecture for traffic flow prediction: deep belief networks with multitask learning. IEEE Transactions on Intelligent Transportation Systems, 15(5), 2191–2201.

Kaur, K., Mangat, V. and Kumar, K. (2020). A comprehensive survey of service function chain provisioning approaches in SDN and NFV architecture. Computer Science Review, 38, 100298.

Kendall, A. and Gal, Y. (2017). What uncertainties do we need in Bayesian deep learning for computer vision? Advances in Neural Information Processing Systems, 5574–5584.

Kreutz, D., Ramos, F.M.V., Veríssimo, P.E., Rothenberg, C.E., Azodolmolky, S. and Uhlig, S. (2015). Software-defined networking: A comprehensive survey. Proceedings of the IEEE, 103(1), 14–76.

Latif, Z., Sharif, K., Li, F., Karim, M.M., Biswas, S. and Wang, Y. (2020). A comprehensive survey of interface protocols for software defined networks. Journal of Network and Computer Applications, 156, 102563.

Liatifis, A., Sarigiannidis, P., Argyriou, V. and Lagkas, T. (2022). Advancing SDN: From OpenFlow to P4, a survey. ACM Computing Surveys.

Shi, W., Cao, J., Zhang, Q., Li, Y. and Xu, L. (2016). Edge computing: Vision and challenges. IEEE Internet of Things Journal, 3(5), 637–646.

Tran, T.X., Hajisami, A., Pandey, P. and Pompili, D. (2017). Collaborative mobile edge computing in 5G networks: New paradigms, scenarios, and challenges. IEEE Communications Magazine, 55(4), 54–61.

Voigt, P. and Von demBussche, A. (2017). The EU General Data Protection Regulation (GDPR). Springer International Publishing.