MODEL AND ALGORITHMS FOR QoS-ROUTING OF CORPORATE DATA FLOWS IN GLOBALLY DISTRIBUTED REAL-TIME CLOUD SYSTEMS

Titov Dmitry Andreevich, Romashkova Oxana Nikolaevna

The article examines the problem of ensuring Quality of Service (QoS) in the routing of corporate data flows in globally distributed real-time cloud systems. The relevance of the study is due to the need for migration of Russian enterprises to domestic cloud infrastructure (Yandex Cloud, VK Cloud, SberCloud) and compliance with requirements of RF legislation in the field of data sovereignty. A mathematical model of multi-criteria routing is proposed, taking into account the geographical extent of Russian communication networks and heterogeneity of data transmission channels. An adaptive routing algorithm has been developed that allows dynamic redistribution of traffic under conditions of peak loads and network node failures. A comparative analysis of the proposed algorithm with existing solutions has been conducted, including simulation on real-world topologies of Russian data centers. The results demonstrated the effectiveness of the algorithm in reducing latency by 40%, jitter by 70%, and increasing bandwidth utilization by 34%.

Keywords:QoS-routing, real-time cloud systems, corporate data, import substitution, distributed systems, Yandex Cloud, VK Cloud, optimization algorithms, multi-criteria optimization.