A Predictive Method of Resource Allocation in Computing Systems Based on a Multicriteria Decision-Making Model
Informatics and Automation · 06.04.2026 · 25 (2) · 568-601
DOI: 10.15622/ia.25.2.10 ↗Kurz gefasst
Fachartikel in der Zeitschrift Informatics and Automation, online erschienen am 06.04.2026. Verfasst von Egor Trushkin und 1 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Modern computing systems typically operate under heterogeneous and variable load conditions. Efficient distribution of computing resources is an important tool for ensuring high performance (e.g., productivity, reliability, stability). Therefore, developing task distribution methods that can simultaneously improve several metrics is a pressing issue. The approach proposed in this paper is an extension of the predictive resource allocation method. This is achieved through the introduction of a multi-criteria decision-making model that includes the predicted execution time, the current node load, and the reliability of the forecast, assessed by the statistics of discrepancies between actual and predicted values. The object of this study is computing systems with heterogeneous nodes processing task flows of variable complexity. The subject of this study is the models and algorithms for predictive distribution of computing resources based on multi-criteria decision making. The objective of the study is to improve the efficiency and stability of heterogeneous computing systems by using a more substantiated mechanism for node selection based on a set of criteria. A review of existing dynamic and predictive distribution methods is provided, and their advantages, disadvantages, and limitations for effective application are identified. A multi-criteria decision-making model was developed that implements the construction of a set of Pareto-optimal solutions and an arbitration procedure. Software simulations were conducted under various system operating scenarios, including conditions with reduced statistical reliability. The results of the study showed that the proposed predictive method based on multi-criteria decision-making reduces the average task execution time and improves node load uniformity compared to known approaches. These results are proposed for use in the construction of heterogeneous computing systems with adaptive resource management systems.
Zitieren
APA
Trushkin, E., & Freyman, V. (2026). A Predictive Method of Resource Allocation in Computing Systems Based on a Multicriteria Decision-Making Model. Informatics and Automation, 25(2), 568-601. https://doi.org/10.15622/ia.25.2.10
Vancouver
Trushkin E, Freyman V. A Predictive Method of Resource Allocation in Computing Systems Based on a Multicriteria Decision-Making Model. Informatics and Automation. 2026;25(2):568-601. doi:10.15622/ia.25.2.10
BibTeX
@article{Trushkin2026,
author = {Trushkin, Egor and Freyman, Vladimir},
title = {A Predictive Method of Resource Allocation in Computing Systems Based on a Multicriteria Decision-Making Model},
journal = {Informatics and Automation},
year = {2026},
volume = {25},
number = {2},
pages = {568--601},
doi = {10.15622/ia.25.2.10}
}
EUVIMED