Open AccessDistributed computing systems synchronization modeling for solving machine learning tasks
Author(s) -
Т. В. Азарнова,
P V Polukhin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1902/1/012050
Subject(s) - computer science , synchronization (alternating current) , distributed computing , queueing theory , consistency (knowledge bases) , resource allocation , resource (disambiguation) , artificial intelligence , machine learning , computer network , channel (broadcasting)
Distributed computing systems are an effective tool for solving complex problems related to processing large amounts of information and data, in particular, machine learning tasks. To improve the efficiency of these systems in solving complex tasks, special methods for synchronization processes modeling, evaluating and predicting processing delays and query execution, developed in the terms of this models. The approaches for the distributed computing systems organization based on the mathematical apparatus of queuing theory, considered in this paper, allow optimizing the requests processing mechanisms related to solving resource allocation problems, and increasing the consistency of computational processes related to machine learning problems.