The Effectiveness of Combining VXLAN, BGP, and EVPN in Large-Scale Networks for Multi-Tenant Data Centers

Read full text
Authors: Aghasi Gevorgyan
Cited by: 0
References: 5

Abstract

The article examines the effectiveness of combining VXLAN, BGP, and EVPN technologies in designing large-scale, multi-tenant data centers with predominantly east–west traffic. The objective of the study is to provide a conceptual and analytical substantiation of a reference data-center fabric architecture that separates the L3 underlay from the overlay plane, is capable of scaling segmentation to millions of logical domains, ensures strict tenant isolation, and supports automatable workload mobility. The relevance of the research is driven by the growth of microservice and cloud workloads, which turn classical L2 extension and VLAN-based segmentation into a source of systemic unpredictability due to flooding and the limited identifier space. The scientific novelty of the work lies in interpreting the VXLAN–EVPN–BGP combination as an integral, controllably scalable system in which the 24-bit VNI, the distributed gateway, and the protocol-formalized dissemination of reachability information form a stable invariant of multi-tenant segmentation, as well as in introducing a diagnostic matrix that operationalizes underlay/overlay separation when identifying the causes of degradations. It is shown that such an architecture reduces the share of parasitic broadcast and unknown unicast traffic, stabilizes latency for inter-server exchange, localizes failure domains in inter-site connectivity, and increases infrastructure readiness for template-driven automation of configurations and policies. The article will be useful to data-center network architects, operations engineers, and developers of cloud-based educational platforms in a multi-tenant environment.

Fields of Study & Subjects

Software-Defined Networks and 5G Cloud Computing and Resource Management Advanced Optical Network Technologies

Publication details

Journal / source
Universal library of engineering technology.
Volume
3
Issue
1
Pages
22–28
Publication date
30 January 2026
Language
EN