Full Article: PDF
Scientific Object Identifier: http://s-o-i.org/1.1/TAS-01-153-18
DOI: https://dx.doi.org/10.15863/TAS.2026.01.153.18
Language: Russian
Citation: Noskova, A.S., & Sergeev, A.V. (2026). Comparative performance analysis of docker containers for arithmetic operations under multitasking load in a windows environment. ISJ Theoretical & Applied Science, 01 (153), 157-170. Soi: https://s-o-i.org/1.1/TAS-01-153-18 Doi: https://dx.doi.org/10.15863/TAS.2026.01.153.18 |
Pages: 157-170
Published: 30.01.2026
Abstract: This article presents a study of the performance of Docker containers when performing arithmetic operations with integers and floating-point numbers, as well as the impact of multitasking on performance compared to a native host in a Windows 10 environment using the WSL2 subsystem. The relevance of the work is due to the widespread use of containerization both in development and in industry for complex technical tasks, as well as the contradictory results of existing Docker performance studies. The aim of the work was a quantitative assessment of the performance degradation when performing operations with integers and real numbers in a multitasking environment on a modern software stack (Docker Desktop 28.0.4, Python 3.13.7). The methodology is based on running Python code that performs arithmetic operations and measures execution time and CPU utilization. The results showed that the host consistently outperforms Docker in terms of performance for single instances. However, with an increase in the number of simultaneously running instances, the performance on the host degrades more significantly than in Docker containers under the same conditions. At the same time, Docker demonstrates higher stability and predictability of results. The obtained data update the results of earlier studies on this issue, highlighting the impact of WSL2 on the final CPU performance.
Key words: Docker, Windows, performance, containerization, WSL2, virtualization, CPU, Python, multitasking.
|
