KEY TRENDS IN THE HARDWARE AND SOFTWARE ORGANIZATION OF MODERN COMPUTER NETWORKS: FROM SPECIALIZED HARDWARE PLATFORMS TO SOFTWARE-COMPOSABLE ENVIRONMENTS
DOI:
https://doi.org/10.36773/1818-1112-2026-139-1-138-144Keywords:
hardware networking, software networking, SDN, NFV, P4, SmartNIC, DPU, hyperconverged infrastructure (HCI), IBN, decomposition, cloud native networks, orchestratorAbstract
Modern computer networks are currently undergoing a fundamental reorganization driven by the demands of cloud computing, the Internet of Things (IoT), big data, and the need for extreme flexibility and adaptability. This article explores and comprehensively analyzes the methods and tools for hardware and software network organization in an era of deep decomposition and convergence. The paper examines the evolution of hardware platforms from closed, application-specific integrated circuits (ASICs) of vendor-specific hardware (operating optimally only within the manufacturer's ecosystem) to open, programmable architectures, network processors (SmartNICs), and data processing units (DPUs), blurring the line between network and computing nodes. The paper also analyzes the transformation of network software: from monolithic operating systems (proprietary, closed-source, and subject to restrictions on use, modification, and distribution) to software-defined networking (SDN) and network functions virtualization (NFV) paradigms.
The paper systematizes modern networking tools, including data-plane programming languages such as P4. The main conclusion of the study is that modern networking is characterized by a transition from tightly coupled, vertically integrated hardware-software architectures to horizontal, software-composable, and decentralized models. In such models, network services and functions are dynamically deployed across multiple heterogeneous computing resources under the control of intelligent controllers and orchestrators (centralized tools that automate the management, coordination, deployment, and scaling of distributed containerized applications). Promising areas for future research are proposed, including integration platforms and tools for complex networking, in particular hyperconverged infrastructures; Cloud-native networking plugins for container orchestrators; cloud networking services.
References
Олифер, В. Г. Компьютерные сети. Принципы, технологии, протоколы : учебник для вузов / В. Г. Олифер, Н. А. Олифер. – СПб. : Питер, 2021. – 944 с.
Куроуз, Дж. Ф. Компьютерные сети. Нисходящий подход / Дж. Ф. Куроуз, К. В. Росс. – М. : Питер, 2021. – 789 с.
Таненбаум, Э. Компьютерные сети / Э. Таненбаум, Д. Уэзеролл. – СПб. : Питер, 2019. – 992 с.
Петерсон, Л. Л. Компьютерные сети: системный подход / Л. Л. Петерсон, Б. Дэви. – СПб. : БХВ-Петербург, 2018. – 944 с.
Бруно, Ф. Программирование FPGA для начинающих / Ф. Бруно. – М. : ДМК Пресс, 2022. – 304 с.
Dhar, S. FPGA Accelerated FPGA Placement / S. Dhar, L. Singhal, M. Iyer // 29th International Conference on Field Programmable Logic and Applications (FPL). – Barcelona, Spain, 2019. – P. 404–410.
P4: Programming Protocol-Independent Packet Processors / P. Bosshart, D. Daly, G. Bibb [et al.] // ACM SIGCOMM Computer Communication Review. – 2014. – Vol. 44, № 3. – P. 87–95.
A Comprehensive Survey of Network Function Virtualization / B. Yi, X. Wang, К. Li, M. Huang // Computer Networks. – 2018. – Vol. 133. – P. 212–262.
Mijumbi, R. Network Function Virtualization: State-of-the-art and Research Challenges / R. Mijumbi, J. Serrat, J. L.Gorricho // IEEE Communications Surveys and Tutorials. – 2015. – Vol. 18. – P. 236–262.
Callegati, F. Virtual Networking Performance in OpenStack Platform for Network Function Virtualization / F. Callegati, W. Cerroni, C. Contoli // Journal of Electrical and Computer Engineering. – 2016. – P. 1–15.
Найдхарт, Т. Программируемые сети: от идеи к реализации. Основы автоматизации сетей с помощью Linux, Python, Ansible и OpenDaylight / Т. Найдхарт. – М. : ДМК Пресс, 2019. – 264 с.
Kim, H. Improving Network Management with Software Defined Networking / H. Kim, N. Feamster. // IEEE Communications Magazine. – 2013. – Vol. 51, № 2. – P. 114–119.
Kreutz, D. Software-Defined Networking: A Comprehensive Survey / D. Kreutz, F. M. Ramos, P. E. Verissimo // Proceedings of the IEEE. – 2015. – Vol. 103, № 1. – P. 14–76.
SD-WAN ROI: How to Quantify Cost Savings and Operational Efficiency. – URL: https://www.scc.com/insights/it-services/networking/sd-wan-roi-how-to-quantify-cost-savings-and-operational-efficiency (date of access: 27.12.2025).
NFV Will Drive Enterprise SDN Adoption. – URL: https://www.sdxcentral.com/analysis/nfv-will-drive-enterprise-sdn-adoption (date of access: 27.12.2025).
Prakash, V. Intent-Based Networking for Modern Infrastructure / V. Prakash, R. Singh. – USA, Sebastopol : O’Reilly Media, 2023. – 310 p.
Моррис, К. Инфраструктура как код: управление серверами в облаке / К. Моррис ; пер. с англ. А. С. Киселева. – СПб. : Питер, 2021. – 384 с.
Halabi, S. Hyperconverged Infrastructure Data Centers: Demystifying HCI / S. Halabi. – USA, Indianapolis : Cisco Press, 2019. – 544 p.
Blokdyk, G. Hyperconverged Infrastructure: A Complete Guide / G. Blokdyk. – Toronto : 5STARCooks, 2021. – 312 p.
Project Calico. Calico documentation. – URL: https://projectcalico.docs.tigera.io (date of access: 15.12.2025).
Cilium – Cloud Native, eBPF-based Networking, Observability, and Security. – URL: https://cilium.io (date of access: 22.12.2025).
AWS. What is Cloud Networking? – URL: https://aws.amazon.com/ what-is/cloud-networking (date of access: 28.12.2025).
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The work is provided under the terms of Creative Commons public license Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). This license allows an unlimited number of persons to reproduce and share the Licensed Material in all media and formats. Any use of the Licensed Material shall contain an identification of its Creator(s) and must be for non-commercial purposes only. Users may not prevent other individuals from taking any actions allowed by the license.
