Russian Federation
Russian Federation
Russian Federation
Russian Federation
The article provides a comprehensive analysis of energy consumption optimization methods in high-performance computing (HPC) and cloud platforms. The research covers a multi-level approach to energy efficiency, including the development of energy-optimized algorithms, the application of dynamic scaling policies in Kubernetes, the use of low-level processor power management mechanisms, and the implementation of specialized metrics to assess energy efficiency. A detailed analysis of the energy consumption of various types of processor operations and memory access methods is presented. Based on the power management formulas of CMOS circuits and comparative tables of energy consumption of operations, the key factors influencing the energy efficiency of computing systems are demonstrated. The results of the study show that the integrated application of the considered methods makes it possible to achieve a significant reduction in energy consumption while maintaining the required level of productivity.
energy efficiency, HPC, cloud computing, green computing, energy consumption optimization, energy efficient algorithms, dynamic scaling, Kubernetes, processor power management, energy efficiency metrics
1. Antipko, A. V. Oblachnye vychisleniya. Modeli razvertyvaniya sistem oblachnyh vychisleniy / A. V. Antipko // Molodoy uchenyy. – 2023. – № 6(453). – S. 9-10. – EDN PHYRFQ.
2. Algoritm vychisleniya preobrazovaniya Laplasa-Stilt'esa dlya vremeni otklika sistemy oblachnyh vychisleniy s gisterezisnym upravleniem / K. E. Samuylov, Yu. V. Gaydamaka, E. S. Sopin, A. V. Gorbunova // Sovremennye informacionnye tehnologii i IT-obrazovanie. – 2015. – T. 11, № 2. – S. 172-177. – EDN WAQEZR.
3. Cherepenin, V. A. Integraciya i optimizaciya sistem oblachnyh, tumannyh i granichnyh vychisleniy: modelirovanie, zaderzhki i algoritmy / V. A. Cherepenin, S. P. Vorob'ev // Izvestiya vysshih uchebnyh zavedeniy. Severo-Kavkazskiy region. Tehnicheskie nauki. – 2024. – № 3(223). – S. 19-25. – DOIhttps://doi.org/10.17213/1560-3644-2024-3-19-25. – EDN LKVVKM.
