ADAPTIVE CONTROL OF STREAM DATA PROCESSING IN DISTRIBUTED SYSTEMS BASED ON LOAD FORECASTING
Modeling Control and Information Technologies · 11.11.2026 · (9) · 33
DOI: 10.31713/mcit.2026.004 ↗Kurz gefasst
Fachartikel in der Zeitschrift Modeling Control and Information Technologies, online erschienen am 11.11.2026. Verfasst von Олександр Лісняк und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
To propose a reference architecture for closed-loop stream-processing adaptation and preliminarily evaluate predictive scaling policies under variable synthetic workloads. Materials and Methods. A closed-loop control architecture is proposed that integrates multi-level monitoring, short-term workload forecasting, a stabilized decision controller, controlled reconfiguration, and audit logging. The simulation compares a reactive threshold policy with two predictive approaches: EWMA with hysteresis and delayed reconfiguration, and a warm-pool approach with pre-initialized resources. A robustness analysis over 30 random seeds was additionally performed.
Results
For the main seed, the number of intervals with SLA-violation risk decreased from 26 to 3 and 0, while average estimated latency decreased from 201.23 ms to 107.62 ms and 92.17 ms, respectively.
Conclusions
The obtained results are preliminary and support further implementation and validation of the proposed approach on a real stream-processing platform.
Zitieren
APA
Лісняк, О., & Олещенко, Л. (2026). ADAPTIVE CONTROL OF STREAM DATA PROCESSING IN DISTRIBUTED SYSTEMS BASED ON LOAD FORECASTING. Modeling Control and Information Technologies, 33. https://doi.org/10.31713/mcit.2026.004
Vancouver
Лісняк О, Олещенко Л. ADAPTIVE CONTROL OF STREAM DATA PROCESSING IN DISTRIBUTED SYSTEMS BASED ON LOAD FORECASTING. Modeling Control and Information Technologies. 2026:33. doi:10.31713/mcit.2026.004
BibTeX
@article{2026,
author = {Лісняк, Олександр and Олещенко, Любов},
title = {ADAPTIVE CONTROL OF STREAM DATA PROCESSING IN DISTRIBUTED SYSTEMS BASED ON LOAD FORECASTING},
journal = {Modeling Control and Information Technologies},
year = {2026},
number = {9},
pages = {33},
doi = {10.31713/mcit.2026.004}
}
EUVIMED