As the scale of AI training, online inference and cloud services expands continuously, the energy consumption of servers in large-scale data centers has changed from a traditional energy efficiency problem at the data center level to a collaborative optimisation problem involving servers, racks, clusters and power systems. This paper introduces a multi-granularity monitoring and adaptive control framework for server energy consumption, focusing on the technical ideas of "observable server energy consumption, predictable load status, closed-loop control actions, and constrained service quality". First, the growth trends of global and US data center power demand are analyzed, and then key problems at five levels are raised: sensor acquisition, power consumption modelling, thermal constraints, service quality constraints, and closed-loop control. Next, build server power consumption models for CPUs, GPUs, memory, fans and power supply units, and design an adaptive control strategy that integrates power capping, DVFS, task migration, sleep/wake-up and cooling coordination. Based on the above results, this method can reduce the peak power and energy consumption per unit of task while meeting SLA requirements, providing an engineering path for energy-saving operation of data centers in high-density AI computing scenarios.

Zhu Xu. Server Energy Consumption Monitoring and Adaptive Control in Large-Scale Data Center. International Journal of Multimedia Computing (2026), Vol. 7, Issue 2: 1-9. https://doi.org/10.38007/IJMC.2026.070201

[1] International Energy Agency. Energy and AI. Paris: IEA, 2025.

[2] Shehabi, A., Smith, S. J., Hubbard, A., Newkirk, A., Lei, N., Siddik, M. A. B., Holecek, B., Koomey, J., Masanet, E., & Sartor, D. 2024 United States Data Center Energy Usage Report. Lawrence Berkeley National Laboratory, LBNL-2001637, 2024.

[3] Liang, Q. (2026). Reconstruction of Commercial Building Space Reuse Mode Driven by Composite Business Types. International Journal of Engineering Advances, 3(1), 107-113.

[4] Gao, Y. (2026). Application of Delaware Corporate Law in Cross-Border M&A: Business Structures, Contractual Risk, and Case-Based Lessons. Economics and Management Innovation, 3(2), 128-136.

[5] Gao, Y. (2026). The Role and Practice of Delaware Law in Global Cross-border M&A Transactions. International Journal of Law, Policy & Society, 2(1), 38-46.

[6] Wang, N. (2026). Research on Evaluation and Optimization Models of Enterprise Resource Allocation Efficiency from a Data-driven Perspective. Advances in Computer and Communication, 7(1).

[7] Su, J. (2026). Research on Android Real-time Communication System Architecture and High-reliability Assurance Pathways Integrating AI-based Anomaly Detection Mechanisms. Engineering Advances, 6(2).

[8] Liang, Q. (2026). Research on the Renovation Design Path for Enhancing the Efficiency of Mixed-Use Office and Retail Spaces. European Journal of AI, Computing & Informatics, 2(2), 163-170.

[9] Liang, Q. (2026). How Architectural Design and Utility Infrastructure Impacts AI Supporting Campus and Drive Future Innovation, Operational Efficiency and Sustainable Advancement in Utility-Critical Environment. European Journal of Engineering and Technologies, 2(2), 71-77.

[10] Liu, Z. (2026). Research on Measuring User Behavior Response Differences Supported by Propensity Scoring Method. European Journal of AI, Computing & Informatics, 2(2), 47-53.

[11] Liu, Z. (2026). Research on Measuring User Behavior Response Differences Supported by Propensity Scoring Method. European Journal of AI, Computing & Informatics, 2(2), 47-53.

[12] Gao, Y. (2026). Governance Structures and Checks and Balances in Private Equity Funds: Practice, Problems, and Improvement Paths. Financial Economics Insights, 3(2), 82-91.

[13] Liang, Q. (2026). Research on Low-Energy Space Organization Methods in the Context of Building System Integration. Global Journal of Science & Innovation, 3(1), 9-15.

[14] Zhu, P. (2025, December). Construction of Multi-Scale Biostatistical Analysis Framework and its Application in Biomedical Signal Feature Recognition and Classification. In 2025 5th International Conference on Mobile Networks and Wireless Communications (ICMNWC) (pp. 1-7). IEEE.

[15] Chen, J. (2026). Elastic Scaling and Stability Assurance Mechanisms for Distributed Systems under High-Throughput Scenarios.

[16] Chang, C. W. (2026). Privacy Infrastructure Vulnerability Mining and Automated Framework Based on Multimodal AI. Procedia Computer Science, 279, 702-711.

[17] Wang, Y. (2026). Construction of Supply Chain Demand Forecasting Algorithm Based On Time Series Data Analysis and Improvement of Its Management Efficiency. Procedia Computer Science, 281, 484-493.

[18] Wang, Z. (2026). Mineral Commodity Time-Series Cycle Modeling and Feature Learning Algorithm Based On Improved Transformer. Procedia Computer Science, 281, 1347-1356.

[19] Chen, J. (2026). Construction of a Cloud-Native High-Performance Service Engineering System for Real-Time Decision-Making Platforms.

[20] Qian, X. (2026). Supply Chain Collaboration Mechanisms Driven by Demand Orchestration in Omni-Channel Retail Environments.