Yaqi Hou
School of Information Science and Engineering, Central South University, Changsha 410083, Hunan, China.
*Corresponding author: Yaqi Hou
Abstract
The heterogeneous server architecture, as an important infrastructure for high-performance computing, its upgrade strategies and resource scheduling logic have gradually become key issues in the field of green computing. In the face of the actual environment where computing unit types are diverse, and energy efficiency structures vary, the traditional upgrade path based on performance priority has obvious shortcomings in energy consumption control and resource utilization efficiency. This article, aiming at the green computing goal, reconfigures the identification and classification mechanism of heterogeneous devices, introduces task feature-oriented adaptation logic, links the scheduling model with the dynamic energy efficiency perception module, and builds a strategy adjustment channel with energy consumption feedback as the closed loop. This enables the upgrade process to balance adaptation accuracy and resource scheduling flexibility. On this basis, application tests are conducted in combination with typical heterogeneous systems in actual scenarios, comparing the changes in various energy efficiency indicators and resource utilization rates before and after the upgrade, to verify the feasibility of this strategy path in reducing unit energy consumption, balancing device loads, and optimizing the overall computing resource structure. It provides feasible support for the upgrade and operation system of data centers towards a green direction.
References
[1] Shaoying Y. Development of low-carbon energy trading and economic management based on cloud computing data optimization. Int J Syst Assur Eng Manag. 2025 Feb 18:1-15.
https://doi.org/10.1007/s13198-025-02330-z
[2] Yuan Z, Zhisheng Y, Kang C, et al. A cloud resource allocation strategy with entry control for multi-priority cloud requests. Arab J Sci Eng. 2023;48(8):10405-15.
https://doi.org/10.1007/s13369-022-07545-3
[3] Wang S, Li X, Ruiz R. Performance analysis for heterogeneous cloud servers using queueing theory. IEEE Trans Comput. 2020;69(4):563-76.
https://doi.org/10.1109/TC.2019.2954591
[4] Zhang H, Li J, Yang H. Cloud computing load prediction method based on CNN-BiLSTM model under low-carbon background. Sci Rep. 2024;14(1):18004.
https://doi.org/10.1038/s41598-024-69059-2
[5] Peng Y, Wang J, Ye X, et al. An intelligent resource allocation strategy with slicing and auction for private edge cloud systems. Future Gener Comput Syst. 2024;160:879-89.
https://doi.org/10.1016/j.future.2024.05.042
[6] Keqin L. Energy-constrained DAG scheduling on edge and cloud servers with overlapped communication and computation. J Grid Comput. 2024;22(3):60.
https://doi.org/10.1007/s10723-024-09771-5
[7] Sophiya S, Mohammad S, Manas S, et al. A load distribution based resource allocation strategy for bag of tasks (BoT) in computational grid environment. Wirel Pers Commun. 2024;135(1):47-80.
https://doi.org/10.1007/s11277-024-11110-6
[8] Kalyan B, Amlan C. A heterogeneous multi-core architectural model for video scheduling for transcoding in clouds. Int J Inf Technol. 2024;16(5):2831-45.
https://doi.org/10.1007/s41870-024-01919-6
[9] Yu H, Feipeng Q, Fei Z, et al. Multi-dimensional resource allocation strategy for LEO satellite communication uplinks based on deep reinforcement learning. J Cloud Comput. 2024;13(1):132.
https://doi.org/10.1186/s13677-024-00695-9
How to cite this paper
Research on Heterogeneous Server Upgrade Strategies and Resource Utilization Efficiency Oriented Toward Green Computing Objectives
How to cite this paper: Yaqi Hou. (2026) Research on Heterogeneous Server Upgrade Strategies and Resource Utilization Efficiency Oriented Toward Green Computing Objectives. Advances in Computer and Communication, 7(1), 47-51.
DOI: http://dx.doi.org/10.26855/acc.2026.03.007