Multi-Cloud ML Systems: Performance Benchmarking and Latency Optimization
Advances in Cyber-Physical Systems · 29.05.2026 · 11 (1) · 6-11
DOI: 10.23939/acps2026.01.006 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advances in Cyber-Physical Systems, online erschienen am 25.05.2026. Verfasst von Mykhailo Bavdys und 1 weiteren Autor:innen. Die Arbeit nennt 12 Literaturangaben.
Abstract (Originalsprache)
The increasing adoption of multi-cloud architectures for machine learning (ML) workloads presents significant challenges in performance optimization and latency management across heterogeneous cloud environments. Contemporary organizations are increasingly leveraging multiple cloud service providers to optimize costs, enhance reliability, ensure vendor independence, and access specialized services. However, this strategic approach introduces complex performance management challenges that traditional single-cloud optimization techniques cannot adequately address. This research presents a comprehensive benchmarking framework and optimization methodology specifically designed for ML-driven multi-cloud systems achieving an average latency reduction of 39.8% and cost optimization of 24.0% while maintaining high availability exceeding 99.87% uptime.
Zitieren
APA
Bavdys, M., & Kushnir, O. (2026). Multi-Cloud ML Systems: Performance Benchmarking and Latency Optimization. Advances in Cyber-Physical Systems, 11(1), 6-11. https://doi.org/10.23939/acps2026.01.006
Vancouver
Bavdys M, Kushnir O. Multi-Cloud ML Systems: Performance Benchmarking and Latency Optimization. Advances in Cyber-Physical Systems. 2026;11(1):6-11. doi:10.23939/acps2026.01.006
BibTeX
@article{Bavdys2026,
author = {Bavdys, Mykhailo and Kushnir, Oleksii},
title = {Multi-Cloud ML Systems: Performance Benchmarking and Latency Optimization},
journal = {Advances in Cyber-Physical Systems},
year = {2026},
volume = {11},
number = {1},
pages = {6--11},
doi = {10.23939/acps2026.01.006}
}
EUVIMED