A server configured with a double carbon management mode

Abstract

This invention discloses a server configured with a dual-carbon management mode, comprising: a governance module, a management module, a carbon emission accounting module, a security module, an energy-saving module, and a data analysis module. The governance module is used for dynamically analyzing and evaluating Power Usage Effectiveness (PUE). The management module is used for collecting energy and material flows, performing intelligent analysis, refined management, and system optimization. The carbon emission accounting module is used for performing carbon emission accounting and generating carbon asset management solutions based on the accounting results. The security module is used to improve the underlying IT / OT application infrastructure. The energy-saving module is used to acquire power consumption data of server nodes and generate corresponding energy-saving strategies based on the power consumption data. The data analysis module is used to provide intelligent decision support for the planning, deployment, and operation and maintenance of data center clusters. This addresses carbon emission compliance requirements, ensures improved lifecycles for data center equipment, and meets data center requirements such as fundamentality, continuity, reliability, stability, security, and energy efficiency.

Description

Technical Field

[0001] This invention relates to the field of server technology, and in particular to a server configured with a dual-carbon management mode. Background Technology

[0002] Currently, dual-carbon management is a complex and long-term task. Specifically, the dual-carbon work has a complex value chain, a long life cycle, and involves many business application systems. The application services involved in dual-carbon work vary at different stages. This includes carbon emission servers, which are dedicated carbon management servers for various industries, implementing carbon data verification and management based on industry, objectives, and targets, including carbon labeling, carbon footprint, carbon accounting, and carbon settlement. With the increasing socialization of national carbon emission policies, dedicated industrial control servers are needed to ensure the implementation of industry policies across various systems, enabling rapid verification and policy subsidy applications. However, current carbon emission servers cannot perform effective carbon emission statistics, resulting in low energy efficiency, poor server stability and security, and an inability to guarantee the life cycle of data center equipment. Summary of the Invention

[0003] This invention aims to at least partially solve one of the technical problems in the aforementioned technologies. Therefore, the purpose of this invention is to propose a server configured with a dual-carbon management mode to address carbon emission compliance requirements, ensure the extended lifespan of data center equipment, meet data center requirements such as basic infrastructure, continuity, reliability, stability, security, and energy efficiency, perform effective carbon emission statistics, and improve energy efficiency.

[0004] To achieve the above objectives, this invention proposes a server configured with a dual-carbon management mode, comprising: a governance module, a management module, a carbon emission accounting module, a security module, an energy-saving module, and a data analysis module; wherein, The governance module is used to dynamically analyze and evaluate PUE based on data collected from the microgrid sensing layer, and to perform harmonic suppression and reactive power compensation through active power filter (APF) and static var generator (SVG). The management module is used to collect energy and material flows through the industrial-grade edge gateway of the edge collaboration layer for intelligent analysis, fine management, and system optimization. The carbon emission accounting module is used to perform carbon emission accounting based on the emission range and generate a carbon asset management solution by combining regional power grid emission factors and green electricity certificate data. The safety module is used to address voltage dips in DVR devices through dynamic voltage regulation. Combined with the overheat prevention module and the ACOPS system, it strengthens the underlying application foundation of IT / OT. The energy-saving module is used to acquire power consumption data and energy storage BMS data of server nodes, and generate corresponding energy-saving strategies by combining peak and off-peak electricity prices and cooling system water temperature regulation requirements. The data analysis module, relying on real-time databases and dual-carbon thematic libraries, provides intelligent decision support for the planning, deployment, and operation and maintenance of data center clusters.

[0005] According to some embodiments of the present invention, the governance module includes: The first acquisition module is used to acquire the energy consumption of IT equipment through the IT load power acquisition device of the microgrid sensing layer. The second acquisition module is used to collect energy consumption data of the data center cluster through the precision air conditioning sensors and chiller energy consumption acquisition modules of the microgrid sensing layer; the energy consumption data of the data center cluster includes the energy consumption of IT equipment, the energy consumption of the cooling system and the energy consumption of the power supply and distribution system. The calculation module is used to calculate the ratio of energy consumption data to the energy consumption of IT equipment, which is used as the PUE value. By comparing the PUE values ​​at different time periods, the energy efficiency change trend of the data center cluster can be evaluated. The power quality management module is used for active power filter (APF) to filter out harmonics and reactive power compensation based on static var generator (SVG).

[0006] According to some embodiments of the present invention, the carbon emission accounting module includes: The statistics module is used to correct carbon emissions based on the carbon emissions of each module in the emission range statistics data center cluster during the operation of the data within a preset time period, combined with regional power grid emission factors and green electricity certificate data. The comparison module is used to compare the carbon emissions of each module during its operation within a preset time period with the corresponding preset carbon emission standards, obtain several comparison results, generate adjustment plans for each module based on the comparison results, and thus obtain a carbon asset management solution.

[0007] According to some embodiments of the present invention, the security module includes: The frequency management module is used to monitor and adjust the system frequency to ensure that it operates within a safe range; The overheat prevention module is used to monitor the temperature of electronic chips and links in real time. When the temperature is determined to be greater than the preset overheat threshold, the cooling mechanism is activated or the load is adjusted. The ACOPS collaboration module is used to integrate with the ACOPS system to jointly monitor CPU temperature. Power quality protection module, used to address voltage dips in DVR equipment via dynamic voltage regulation; The optimization module is used to optimize power distribution and load balancing.

[0008] According to some embodiments of the present invention, the energy-saving module includes: The first generation module is used for: The server nodes are classified according to the type of business requirements, resulting in several sets of server nodes; Each server node set is monitored to obtain power consumption data and energy storage BMS data for the server node set, and the time corresponding to the power consumption data and energy storage BMS data is recorded. Based on the power consumption data of the server node set and the energy storage BMS data obtained from monitoring within a preset time period, and combined with the peak and off-peak electricity prices and the water temperature regulation requirements of the cooling system, a power consumption curve of the server node set is generated. The first determining module is used to determine, based on the power consumption curve, the set of server nodes that need to be adjusted, as the target set of server nodes; and to determine the target time period for which the target set of server nodes needs to be adjusted. The second generation module is used to determine the operational requirements of the target server node set during the target time period and generate energy-saving strategies based on the operational requirements.

[0009] According to some embodiments of the present invention, the energy-saving module further includes: The second determining module is used to determine, during the execution of energy-saving strategies on the target server node set, when the operating requirements of the target server nodes change in the future time period, to stop the energy-saving strategy and generate a new operating strategy.

[0010] According to some embodiments of the present invention, the second generation module includes: The third determination module is used to obtain the running scenario of the target server node set in the target time period and determine the running requirements based on the running scenario; The adjustment module is used to adjust the target server node set according to operational requirements and generate energy-saving strategies.

[0011] According to some embodiments of the present invention, the third determining module includes: The acquisition module is used to collect runtime environment data and extract feature vectors; The fusion module is used for: The feature vector is divided into several sub-feature vectors based on the quadtree algorithm; Determine the data type and scene processing requirements corresponding to each sub-feature vector, and assign each sub-feature vector and scene processing requirements to the corresponding scene processing node; based on the scene processing node, extract several scene labels according to the sub-feature vector and scene processing requirements, each scene label corresponds to a scene feature, match the scene features with the preset scene features in the preset scene feature database, and determine the preset scene feature corresponding to the maximum matching degree as the target preset scene feature; Based on several scene labels, several target preset scene features are obtained. Target preset scene features with a maximum matching degree less than a preset threshold are removed to determine the target set. Based on the fusion relationship of each target preset scene feature in the target set, the running scene of the target server node set in the target time period is obtained.

[0012] According to some embodiments of the present invention, the adjustment module includes: The fourth determining module is used for: Analyze the operational requirements to determine the value of the operational tasks that need to be completed within time t; Number the server nodes in the target server node set, denoted as i, i = 1, 2, 3, ..., n, where n is the number of server nodes in the target server node set; let P i P is the power consumption of the i-th server node during operation. is To enable the instantaneous power of the i-th server node, P is >P i Let F i F represents the calculated value when the i-th server node is running normally; t This represents the value of the running task that needs to be completed within time t; The fifth determining module is used for: When it is detected that the target server node set needs to be adjusted, the value F of the running task that needs to be completed within time t is... t At that time, determine the group S of server nodes that need to be in a running state from the target server node set. t for: ; in, This represents the inversion and transpose of a matrix; This indicates the running status of the i-th server node. This indicates that the i-th server node is in a closed state. This indicates that the i-th server node is active. The function represents array operations, where e is a natural constant; This indicates the instantaneous power consumption coefficient of some server nodes: ; The factor representing the cost-saving effect when some server nodes are shut down: ; in, To enable the instantaneous computation value of the i-th server node; Energy-saving strategies are generated based on the group of server nodes that need to be running in the target server node set.

[0013] This invention proposes a server configured with a dual carbon management mode to address carbon emission compliance requirements, ensure the extended lifespan of data center equipment, meet data center requirements such as basic infrastructure, continuity, reliability, stability, security, and energy efficiency, conduct effective carbon emission statistics, and improve energy saving rates.

[0014] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a block diagram of a server configured with a dual-carbon management mode according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a data center cluster according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a governance module according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a management module according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a carbon emission accounting module according to an embodiment of the present invention. Detailed Implementation

[0017] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0018] like Figure 1 As shown in the figure, this embodiment of the invention proposes a server configured with a dual-carbon management mode, including: a governance module, a management module, a carbon emission accounting module, a security module, an energy-saving module, and a data analysis module; wherein, The governance module is used to collect data from the microgrid sensing layer, dynamically analyze and evaluate PUE, and perform harmonic suppression and reactive power compensation through active power filter (APF) and static var generator (SVG). The management module is used to collect energy and material flows through the industrial-grade edge gateway of the edge collaboration layer for intelligent analysis, fine management, and system optimization. The carbon emission accounting module is used to perform carbon emission accounting based on the emission range and generate a carbon asset management solution by combining regional power grid emission factors and green electricity certificate data. The safety module is used to address voltage dips in DVR devices through dynamic voltage regulation. Combined with the overheat prevention module and the ACOPS system, it strengthens the underlying application foundation of IT / OT. The energy-saving module is used to acquire power consumption data and energy storage BMS data of server nodes, and generate corresponding energy-saving strategies by combining peak and off-peak electricity prices and cooling system water temperature regulation requirements. The data analysis module, relying on real-time databases and dual-carbon thematic libraries, provides intelligent decision support for the planning, deployment, and operation and maintenance of data center clusters.

[0019] The working principle of the above technical solution is as follows: Figure 2 As shown, the data center cluster includes field control and data management. Field control includes CPU overheat protection, field control units, process management units, link management units, and self-protection systems; data management includes a comprehensive database, an ITOT database, a system database, a PUE database, and an operations and maintenance database.

[0020] like Figure 3 As shown, the governance module generates sampling, display, and compensation upload commands. Based on voltage and current sampling analysis, the analysis results are displayed on a touch screen, and fault alarms are triggered when anomalies are detected. Reactive power compensation and harmonic filtering are implemented during the analysis and processing. The processing results are uploaded to cloud platforms, mobile devices, and PCs. Dynamic analysis and evaluation of PUE avoids the problems of scattered, data-less applications with multiple mechanized power equipment. It addresses the issues of frequent faults, high maintenance intensity, and scarce data center space requirements. By integrating multiple power equipment units into one, the footprint can be reduced by more than 30% compared to traditional solutions. Energy flow includes power supply and distribution, and photovoltaic-energy storage; material flow includes refrigerant for refrigeration systems and IT loads.

[0021] In this embodiment, such as Figures 4-5 As shown, the management module utilizes next-generation information technologies such as 5G, industrial internet, artificial intelligence, and big data to collect, monitor, intelligently analyze, finely manage, and optimize information such as energy flow and material flow, thereby improving the level of energy resource management in industrial manufacturing processes.

[0022] In this embodiment, based on the requirements of full lifecycle accounting, data center carbon emissions are divided into three categories, clearly defining the accounting objects and boundaries of each category: the emission scope includes a first emission scope, a second emission scope, and a third emission scope; wherein, the first emission scope is scope 1, representing direct emissions: emergency generator fuel consumption; accounting for carbon emissions directly generated within the data center, with the core object being the fuel consumption of emergency generators; the second emission scope is scope 2, representing indirect emissions: purchased electricity; accounting for carbon emissions generated from purchased electricity in the data center, with the core object being the amount of purchased electricity consumed; the third emission scope is scope 3, representing other indirect emissions: equipment transportation. Indirect carbon emissions related to data center operation are accounted for, with core objects including energy consumption for transporting servers / cooling equipment, energy consumption for waste disposal, etc. Multi-dimensional data collection and integration are performed using hardware sensing and data analysis modules. Core data is collected through dedicated equipment in the microgrid sensing layer, and after conversion by the edge collaboration layer, the data is uploaded to the data analysis module. Key parameters are retrieved from the data analysis module: regional grid emission factor (e.g., 0.6tCO2 / MWh), green electricity consumption corresponding to green electricity certificates, and default emission factors for fuel / transportation. Based on the calculation logic of the carbon accounting engine, carbon emission calculations are performed in different ranges, and a green electricity deduction mechanism is introduced to correct the results: Range 1 calculation: The total direct emissions are the product of emergency generator fuel consumption and fuel emission factor; Range 2 calculation: The total basic indirect emissions are the product of purchased electricity consumption and regional power grid emission factor; After green electricity deduction, the total emissions in Range 2 are the total basic indirect emissions minus the product of green electricity consumption and regional power grid emission factor; Range 3 calculation: The total other indirect emissions are the product of computing equipment transportation fuel consumption and transportation emission factor plus the product of waste treatment energy consumption and corresponding emission factor; Total carbon emission calculation: Total carbon emissions of the data center = Range 1 emissions + Range 2 emissions after deduction + Range 3 emissions. The calculation results of the total carbon emission calculation provide intelligent decision support for the planning, deployment, and operation and maintenance of data center clusters.

[0023] In this embodiment, the security module strengthens the underlying IT / OT application foundation, preventing system malfunctions caused by frequency-induced energy consumption overheating or link overheating, ensuring the health and lifespan of electronic chips, and preventing ACOPS-triggered shutdowns. The power quality monitoring device in the microgrid sensing layer collects voltage data from the power supply and distribution system in real time, uploading the data to the power quality monitoring and analysis module in the business logic layer via the edge collaboration layer. When the module detects a voltage dip, it immediately generates a DVR start command and quickly sends it to the DVR device in the power quality governance layer via the edge gateway. After receiving the command, the DVR device quickly adjusts the output voltage through its internal power electronic devices, compensating for the voltage dip to 90%-105% of the rated value. The overheat prevention module uses high-precision temperature sensors deployed on the server motherboard, CPU link, and cooling system heat exchanger to collect real-time electronic chip temperature, link interface temperature, and ambient temperature. The ACOPS system acquires CPU core temperature data in real time; simultaneously, through the protocol conversion function of the edge collaboration layer, it connects the temperature data from the ACOPS system with the temperature data from the overheat prevention module, avoiding monitoring deviations caused by single data sources. Security modules can improve the stability of IT / OT links, reduce the risk of failure, extend the life cycle of data center equipment, reduce replacement costs, reduce maintenance intervention, and lower operating costs.

[0024] In this embodiment, the energy-saving module collects the real-time operating power of each server node through intelligent PDUs deployed in the IT cabinet; simultaneously, it binds the power consumption data to the service type of the node. Real-time status data of the energy storage system, including individual battery voltage, charging / discharging current, SOC, and charging / discharging power limits, is acquired through the energy storage BMS collector in the microgrid sensing layer. Two types of key auxiliary data are collected simultaneously: peak-valley electricity price data and cooling system status data. A time-power curve is generated based on the power consumption data for a preset time period; then, a load analysis algorithm identifies the set of nodes with high power consumption and low load in the curve (e.g., non-core storage nodes with power consumption >100W but load rate <20% between 2-4 AM), and these are determined as the target server node set. Combining peak-valley electricity price periods and cooling system load, a target time period is determined and a scenario is matched. If the target time period is a valley period and the energy storage SOC <30%, a scenario of energy storage charging + low-load node frequency reduction is matched; if the target time period is a peak period and the cooling system outlet water temperature <14℃, a scenario of energy storage discharging + cooling water temperature increase is matched. Server node adjustment strategies are generated based on scenario requirements. The cooling system's water temperature regulation strategy is generated based on the server node's heat dissipation requirements. The energy-saving strategy is executed via a link-based command structure: the intelligent PDU executes node switching / frequency reduction commands, the energy storage controller executes charging / discharging commands, and the cooling inverter executes water temperature regulation commands. This facilitates improved energy utilization efficiency, achieves core energy-saving goals, reduces operating costs, and balances energy conservation and economic efficiency.

[0025] In this embodiment, the data analysis module provides an intuitive visual experience through a graphical interface, enabling administrators to monitor and manage the data center cluster in real time. It achieves automated management and optimization of the data center cluster through intelligent means, automating report generation and reducing manual intervention costs. Utilizing data analysis, machine learning, and other technologies, it provides intelligent decision support for the planning, deployment, and operation and maintenance of the data center cluster.

[0026] The beneficial effects of the above technical solutions are: solving carbon emission compliance requirements, ensuring the improvement of data center equipment life cycle, meeting data center requirements such as basic, continuous, reliable, stable, secure, and energy-efficient performance, conducting effective carbon emission statistics, and improving energy efficiency.

[0027] According to some embodiments of the present invention, the governance module includes: The first acquisition module is used to acquire the energy consumption of IT equipment through the IT load power acquisition device of the microgrid sensing layer. The second acquisition module is used to collect energy consumption data of the data center cluster through the precision air conditioning sensors and chiller energy consumption acquisition modules of the microgrid sensing layer; the energy consumption data of the data center cluster includes the energy consumption of IT equipment, the energy consumption of the cooling system and the energy consumption of the power supply and distribution system. The calculation module is used to calculate the ratio of energy consumption data to the energy consumption of IT equipment, which is used as the PUE value. By comparing the PUE values ​​at different time periods, the energy efficiency change trend of the data center cluster can be evaluated. The power quality management module is used for active power filter (APF) to filter out harmonics and reactive power compensation based on static var generator (SVG).

[0028] The working principle and beneficial effects of the above technical solution are as follows: The first acquisition module is specifically used to collect energy consumption data from IT equipment to understand the energy consumption of individual devices or groups of devices, thereby enabling targeted energy efficiency optimization. The second acquisition module comprehensively collects energy consumption data from the data center cluster. By comparing PUE values ​​over different time periods, the energy efficiency trend of the data center cluster can be intuitively assessed. The lower the PUE value, the less energy the data center consumes to provide the same IT services, and the higher its energy efficiency. Real-time acquisition and calculation of PUE values ​​allow data center managers to monitor energy efficiency at any time; by comparing PUE values ​​over different time periods, the trend of energy efficiency changes can be analyzed, problems of declining energy efficiency can be identified in a timely manner, and measures can be taken to improve them. Based on the collected data and calculated PUE values, data center managers can formulate targeted energy efficiency optimization strategies, such as adjusting cooling system parameters and optimizing power distribution.

[0029] According to some embodiments of the present invention, the carbon emission accounting module includes: The statistics module is used to correct carbon emissions based on the carbon emissions of each module in the emission range statistics data center cluster during the operation of the data within a preset time period, combined with regional power grid emission factors and green electricity certificate data. The comparison module is used to compare the carbon emissions of each module during its operation within a preset time period with the corresponding preset carbon emission standards, obtain several comparison results, generate adjustment plans for each module based on the comparison results, and thus obtain a carbon asset management solution.

[0030] The working principle and beneficial effects of the above technical solution are as follows: Based on the carbon emissions of each module in the data center cluster during operation within a preset time period, the carbon emissions are corrected by combining regional power grid emission factors and green electricity certificate data to calculate the carbon emissions of each module within the preset time period. Reasonable carbon emission standards are set for each module according to industry standards, policies, regulations, or internal corporate carbon emission targets. The carbon emissions obtained from the statistical module are compared with the preset carbon emission standards to generate comparison results. Based on the comparison results, targeted adjustment plans are proposed for each module. For example, for modules whose carbon emissions exceed the standard, suggestions such as energy conservation and consumption reduction, optimized equipment configuration, and adoption of clean energy can be proposed. The adjustment plans of each module are integrated to form an overall carbon asset management solution. This solution should include specific implementation steps, timelines, responsible persons, and expected carbon emission reduction effects. Through the statistical module, the data center cluster can monitor the carbon emissions of each module in real time, providing data support for carbon emission management. Through the comparison module, modules with high carbon emissions can be identified in a timely manner, and corresponding adjustment measures can be taken to achieve carbon emission reduction targets. The resulting carbon asset management solution helps data center clusters optimize resource allocation, improve energy efficiency, reduce carbon emission costs, and enhance enterprises' carbon asset management capabilities.

[0031] According to some embodiments of the present invention, the security module includes: The frequency management module is used to monitor and adjust the system frequency to ensure that it operates within a safe range; The overheat prevention module is used to monitor the temperature of electronic chips and links in real time. When the temperature is determined to be greater than the preset overheat threshold, the cooling mechanism is activated or the load is adjusted. The ACOPS collaboration module is used to integrate with the ACOPS system to jointly monitor CPU temperature. Power quality protection module, used to address voltage dips in DVR equipment via dynamic voltage regulation; The optimization module is used to optimize power distribution and load balancing.

[0032] The working principle and beneficial effects of the above technical solutions are as follows: The frequency management module monitors the system's frequency changes in real time and compares them with a preset safe frequency range. If the system frequency exceeds the safe range, it automatically adjusts system parameters (such as voltage, current, etc.) or triggers an alarm mechanism to ensure stable system operation. The overheat prevention module collects temperature data of electronic chips and links in real time through built-in temperature sensors or external temperature monitoring devices. When the temperature exceeds a preset overheat threshold, it automatically activates cooling mechanisms (such as fans, heat sinks, etc.) or adjusts the load to reduce the temperature and prevent overheating from damaging the system. The ACOPS system is typically used to monitor and optimize data center cooling and power management. Through the ACOPS collaboration module, the security module can share data and work collaboratively with the ACOPS system to jointly monitor CPU temperature changes and adjust cooling strategies and power distribution as needed to ensure the CPU operates at a safe temperature. The power quality protection module is used to address voltage dips by dynamically regulating DVR equipment. The optimization module analyzes the system's power requirements and load conditions, and can dynamically adjust power distribution to ensure that each component receives sufficient power while avoiding energy waste caused by over-supply. Furthermore, the optimization module can dynamically adjust load balancing strategies based on load changes to ensure stable system operation and optimal performance. Through measures such as frequency management, overheat prevention, and power optimization, the security module can significantly improve system stability and reduce system failure rates. Real-time monitoring and early warning mechanisms can help promptly detect and address potential security risks, such as overheating and frequency anomalies, thereby preventing security incidents. By optimizing power distribution and load balancing, the security module can improve system energy efficiency and reduce energy costs. Through the collaborative work of all modules, the stable operation of critical systems in data centers or similar environments can be ensured, and potential security risks can be prevented.

[0033] According to some embodiments of the present invention, the energy-saving module includes: The first generation module is used for: The server nodes are classified according to the type of business requirements, resulting in several sets of server nodes; Each server node set is monitored to obtain power consumption data and energy storage BMS data for the server node set, and the time corresponding to the power consumption data and energy storage BMS data is recorded. Based on the power consumption data of the server node set and the energy storage BMS data obtained from monitoring within a preset time period, and combined with the peak and off-peak electricity prices and the water temperature regulation requirements of the cooling system, a power consumption curve of the server node set is generated. The first determining module is used to determine, based on the power consumption curve, the set of server nodes that need to be adjusted, as the target set of server nodes; and to determine the target time period for which the target set of server nodes needs to be adjusted. The second generation module is used to determine the operational requirements of the target server node set during the target time period and generate energy-saving strategies based on the operational requirements.

[0034] The working principle of the above technical solution is as follows: Server nodes are divided into different sets based on business needs. This facilitates the implementation of more precise energy-saving measures tailored to different business characteristics. Real-time monitoring is performed on each server node set, collecting power consumption data and energy storage BMS data, and recording the corresponding timestamps. Based on the power consumption data and energy storage BMS data collected within a preset time period, combined with peak-valley electricity pricing and cooling system water temperature regulation requirements, a power consumption curve is generated for each server node set. The power consumption curve can intuitively reflect the energy consumption trend of the server nodes. By analyzing the power consumption curve, server node sets with abnormal or above-average energy consumption are identified as target server node sets. Further analysis of the power consumption curve determines the time periods during which the target server node sets have higher energy consumption; these time periods are the target time periods requiring adjustment. A thorough understanding of the business load and data processing requirements of the target server node sets during the target time periods ensures that energy-saving strategies do not affect the normal operation of the business. Energy-saving strategies are generated based on operational needs, aiming to reduce energy consumption while maintaining business performance and stability. Monitor each server node set, including monitoring the load of each server node in the server node set; specifically, monitor the motherboard load, processor power consumption, processor frequency, memory frequency, network bandwidth, and overall power consumption of the server nodes.

[0035] The beneficial effects of the above technical solution are as follows: Through refined management and optimization strategies, the energy-saving module can significantly reduce the energy consumption of server nodes, improving overall energy efficiency and environmental protection. The implementation of energy-saving strategies needs to be carried out while ensuring business stability; therefore, the energy-saving module can also improve the reliability and stability of business operations to a certain extent.

[0036] According to some embodiments of the present invention, the energy-saving module further includes: The second determining module is used to determine, during the execution of energy-saving strategies on the target server node set, when the operating requirements of the target server nodes change in the future time period, to stop the energy-saving strategy and generate a new operating strategy.

[0037] The working principle and beneficial effects of the above technical solution are as follows: The second determination module continuously monitors the operational status of the target server node set, including key indicators such as business load and data processing requirements. By comparing the current operational requirements with preset thresholds or historical data, the second determination module can identify whether there have been significant changes in operational requirements. Once a change in operational requirements is detected, the second determination module immediately stops the currently executed energy-saving strategy to avoid performance degradation or resource waste caused by strategy mismatch. Simultaneously, it generates and deploys new operational strategies based on the new operational requirements. This ability to dynamically adjust strategies according to changes in actual operational requirements improves the system's flexibility and adaptability. Through continuous monitoring and strategy adjustment, the energy-saving module ensures that server nodes always operate in optimal energy efficiency, further reducing energy consumption. During strategy adjustment, the second determination module ensures that business performance and stability are not affected, thereby guaranteeing business continuity. The entire energy-saving module, through intelligent monitoring, analysis, and strategy generation, achieves refined management of server node energy consumption, improving management efficiency.

[0038] According to some embodiments of the present invention, the second generation module includes: The third determination module is used to obtain the running scenario of the target server node set in the target time period and determine the running requirements based on the running scenario; The adjustment module is used to adjust the target server node set according to operational requirements and generate energy-saving strategies.

[0039] The working principle and beneficial effects of the above technical solution are as follows: It acquires operational environment data of the target server node set within a target time period, such as business type, data traffic, and processing task type, to construct a complete operational scenario. Based on the acquired operational scenario, it infers the operational needs of the target server node set in future time periods, such as required computing power. The adjustment module generates a series of energy-saving strategies based on operational needs, combined with server hardware configuration, current energy efficiency status, and the feasibility of energy-saving technologies. This includes adjusting the number and status of server nodes that need to be operational within the target server node set. Through intelligent operational scenario identification, demand prediction, and strategy adjustment, it achieves refined management and optimization of the server node set's energy consumption.

[0040] According to some embodiments of the present invention, the third determining module includes: The acquisition module is used to collect runtime environment data and extract feature vectors; The fusion module is used for: The feature vector is divided into several sub-feature vectors based on the quadtree algorithm; Determine the data type and scene processing requirements corresponding to each sub-feature vector, and assign each sub-feature vector and scene processing requirements to the corresponding scene processing node; based on the scene processing node, extract several scene labels according to the sub-feature vector and scene processing requirements, each scene label corresponds to a scene feature, match the scene features with the preset scene features in the preset scene feature database, and determine the preset scene feature corresponding to the maximum matching degree as the target preset scene feature; Based on several scene labels, several target preset scene features are obtained. Target preset scene features with a maximum matching degree less than a preset threshold are removed to determine the target set. Based on the fusion relationship of each target preset scene feature in the target set, the running scene of the target server node set in the target time period is obtained.

[0041] The working principle of the above technical solution is as follows: The acquisition module collects runtime environment data, such as server hardware status, network traffic, and task queue length. It then uses a feature extraction algorithm to extract representative feature vectors from this raw data. The fusion module first uses a quadtree algorithm to partition the feature vectors into several sub-feature vectors. The quadtree algorithm is an effective spatial partitioning method capable of efficiently processing multi-dimensional data. The fusion module assigns each sub-feature vector to a corresponding scene processing node based on its data type and scene processing requirements. These scene processing nodes are specifically designed to process specific types of data and can efficiently extract scene labels. At each scene processing node, the fusion module extracts several scene labels based on the sub-feature vectors and scene processing requirements. Each scene label corresponds to a scene feature, which describes the running status of the server node set within a specific time period. Then, the fusion module matches these scene features with preset scene features in a preset scene feature database to determine the preset scene feature with the highest matching degree, which is then used as the target preset scene feature. Based on several scene labels, the fusion module obtains several target preset scene features. Then, it removes target preset scene features with a maximum matching degree less than a preset threshold to ensure that the final determined running scene is accurate and reliable. The fusion module will synthesize the running scene of the target server node set in the target time period based on the fusion relationship of each target preset scene feature in the target set.

[0042] The beneficial effects of the above technical solution are as follows: Through the collaborative work of the acquisition module and the fusion module, the third determination module can accurately identify and extract the operating scenarios of the target server node set within the target time period. Utilizing the quadtree algorithm and the allocation of scene processing nodes, the fusion module can efficiently process a large number of feature vectors and scene labels, improving the system's processing speed and accuracy. By setting a preset scene feature database and matching thresholds, the third determination module can flexibly adapt to different server node sets and operating environments, improving the system's versatility and scalability.

[0043] According to some embodiments of the present invention, the adjustment module includes: The fourth determining module is used for: Analyze the operational requirements to determine the value of the operational tasks that need to be completed within time t; Number the server nodes in the target server node set, denoted as i, i = 1, 2, 3, ..., n, where n is the number of server nodes in the target server node set; let P i P is the power consumption of the i-th server node during operation. is To enable the instantaneous power of the i-th server node, P is >P i Let F i F represents the calculated value when the i-th server node is running normally; t This represents the value of the running task that needs to be completed within time t; The fifth determining module is used for: When it is detected that the target server node set needs to be adjusted, the value F of the running task that needs to be completed within time t is... t At that time, determine the group S of server nodes that need to be in a running state from the target server node set. t for: ; in, This represents the inversion and transpose of a matrix; This indicates the running status of the i-th server node. This indicates that the i-th server node is in a closed state. This indicates that the i-th server node is active. The function represents array operations, where e is a natural constant; This indicates the instantaneous power consumption coefficient of some server nodes: ; The factor representing the cost-saving effect when some server nodes are shut down: ; in, To enable the instantaneous computation value of the i-th server node; Energy-saving strategies are generated based on the group of server nodes that need to be running in the target server node set.

[0044] The working principle and beneficial effects of the above technical solution are as follows: The fourth determination module analyzes the operational requirements and determines the value of the operational tasks that need to be completed within time t; it numbers each server node in the target server node set and sets P... i P is the power consumption of the i-th server node during operation. is To enable the instantaneous power of the i-th server node, P is >P i Let F i F represents the calculated value when the i-th server node is running normally; t The value of the running tasks that need to be completed within time t; the fifth determining module is responsible for determining which server nodes should be in running state based on the running task value Ft and the server node settings when an adjustment is detected, and generating an energy-saving strategy accordingly. If all server nodes are currently in a shutdown state, then while ensuring Ft... t At the same time, only the fewest possible combinations of server nodes with the lowest instantaneous power consumption should be activated. This avoids energy waste caused by activating too many server nodes or activating a moderate number of server nodes but activating server nodes with high instantaneous power consumption. If some server nodes are currently running and some are shut down, the processing power can be increased or decreased accordingly. tSimultaneously, the system enables or disables server nodes with low instantaneous power consumption. When increased processing capacity is needed, enabling server nodes with low instantaneous power consumption reduces energy consumption; when reduced processing capacity is needed, disabling server nodes with low instantaneous power consumption (resulting in lower energy consumption upon subsequent enabling). If all server nodes are currently enabled, disabling server nodes with low instantaneous power consumption reduces energy consumption while ensuring processing capacity. Instantaneous power consumption coefficient α: represents the ratio of the instantaneous power consumption of some server nodes when enabled to their normal operating power consumption. Saving effect factor β: represents the proportion of power saved when some server nodes are disabled compared to when all are enabled. The fifth determination module determines a server node group St, in which the server nodes should be in an operational state within time t. The fifth determination module generates an energy-saving strategy based on the determined server node group St, including which nodes should be enabled, which nodes should be disabled, and corresponding power adjustments. The energy-saving strategy should aim to maximize server energy efficiency, i.e., minimize energy consumption while meeting operational requirements. While saving energy, it is essential to ensure that business performance is not affected; that is, the computing power of the server nodes should be sufficient to meet the needs of the operational tasks. By precisely controlling the operating status of server nodes, energy consumption can be significantly reduced, achieving the goals of energy conservation and emission reduction. Reducing energy consumption means lowering operating costs, and optimizing the use of server nodes can improve resource utilization and avoid resource waste.

[0045] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A server configured with a dual-carbon management mode, characterized in that, include: The system includes modules for governance, management, carbon emission accounting, safety, energy conservation, and data analysis; among which, The governance module is used to dynamically analyze and evaluate PUE based on data collected from the microgrid sensing layer, and to perform harmonic suppression and reactive power compensation through active power filter (APF) and static var generator (SVG). The management module is used to collect energy and material flows through the industrial-grade edge gateway of the edge collaboration layer for intelligent analysis, fine management, and system optimization. The carbon emission accounting module is used to perform carbon emission accounting based on the emission range and generate a carbon asset management solution by combining regional power grid emission factors and green electricity certificate data. The safety module is used to address voltage dips in DVR devices through dynamic voltage regulation. Combined with the overheat prevention module and the ACOPS system, it strengthens the underlying application foundation of IT / OT. The energy-saving module is used to acquire power consumption data and energy storage BMS data of server nodes, and generate corresponding energy-saving strategies by combining peak and off-peak electricity prices and cooling system water temperature regulation requirements. The data analysis module, relying on real-time databases and dual-carbon thematic libraries, provides intelligent decision support for the planning, deployment, and operation and maintenance of data center clusters.

2. The server configured with dual-carbon management mode as described in claim 1, characterized in that, The governance module includes: The first acquisition module is used to acquire the energy consumption of IT equipment through the IT load power acquisition device of the microgrid sensing layer. The second acquisition module is used to collect energy consumption data of the data center cluster through the precision air conditioning sensors and chiller energy consumption acquisition modules of the microgrid sensing layer; the energy consumption data of the data center cluster includes the energy consumption of IT equipment, the energy consumption of the cooling system and the energy consumption of the power supply and distribution system. The calculation module is used to calculate the ratio of energy consumption data to the energy consumption of IT equipment, which is used as the PUE value. By comparing the PUE values ​​at different time periods, the energy efficiency change trend of the data center cluster can be evaluated. The power quality management module is used for active power filter (APF) to filter out harmonics and reactive power compensation based on static var generator (SVG).

3. The server configured with dual-carbon management mode as described in claim 1, characterized in that, The carbon emission accounting module includes: The statistics module is used to correct carbon emissions based on the carbon emissions of each module in the emission range statistics data center cluster during the operation of the data within a preset time period, combined with regional power grid emission factors and green electricity certificate data. The comparison module is used to compare the carbon emissions of each module during its operation within a preset time period with the corresponding preset carbon emission standards, obtain several comparison results, generate adjustment plans for each module based on the comparison results, and thus obtain a carbon asset management solution.

4. The server configured with dual-carbon management mode as described in claim 1, characterized in that, The security module includes: The frequency management module is used to monitor and adjust the system frequency to ensure that it operates within a safe range; The overheat prevention module is used to monitor the temperature of electronic chips and links in real time. When the temperature is determined to be greater than the preset overheat threshold, the cooling mechanism is activated or the load is adjusted. The ACOPS collaboration module is used to integrate with the ACOPS system to jointly monitor CPU temperature. Power quality protection module, used to address voltage dips in DVR equipment via dynamic voltage regulation; The optimization module is used to optimize power distribution and load balancing.

5. The server configured with dual-carbon management mode as described in claim 1, characterized in that, The energy-saving module includes: The first generation module is used for: The server nodes are classified according to the type of business requirements, resulting in several sets of server nodes; Each server node set is monitored to obtain power consumption data and energy storage BMS data for the server node set, and the time corresponding to the power consumption data and energy storage BMS data is recorded. Based on the power consumption data of the server node set and the energy storage BMS data obtained from monitoring within a preset time period, and combined with the peak and off-peak electricity prices and the water temperature regulation requirements of the cooling system, a power consumption curve of the server node set is generated. The first determining module is used to determine, based on the power consumption curve, the set of server nodes that need to be adjusted, as the target set of server nodes; and to determine the target time period for which the target set of server nodes needs to be adjusted. The second generation module is used to determine the operational requirements of the target server node set during the target time period and generate energy-saving strategies based on the operational requirements.

6. The server configured with dual-carbon management mode as described in claim 5, characterized in that, The energy-saving module also includes: The second determining module is used to determine, during the execution of energy-saving strategies on the target server node set, when the operating requirements of the target server nodes change in the future time period, to stop the energy-saving strategy and generate a new operating strategy.

7. The server configured with dual-carbon management mode as described in claim 5, characterized in that, The second generation module includes: The third determination module is used to obtain the running scenario of the target server node set in the target time period and determine the running requirements based on the running scenario; The adjustment module is used to adjust the target server node set according to operational requirements and generate energy-saving strategies.

8. The server configured with dual-carbon management mode as described in claim 7, characterized in that, The third determining module includes: The acquisition module is used to collect runtime environment data and extract feature vectors; The fusion module is used for: The feature vector is divided into several sub-feature vectors based on the quadtree algorithm; Determine the data type and scene processing requirements corresponding to each sub-feature vector, and assign each sub-feature vector and scene processing requirements to the corresponding scene processing node; based on the scene processing node, extract several scene labels according to the sub-feature vector and scene processing requirements, each scene label corresponds to a scene feature, match the scene features with the preset scene features in the preset scene feature database, and determine the preset scene feature corresponding to the maximum matching degree as the target preset scene feature; Based on several scene labels, several target preset scene features are obtained. Target preset scene features with a maximum matching degree less than a preset threshold are removed to determine the target set. Based on the fusion relationship of each target preset scene feature in the target set, the running scene of the target server node set in the target time period is obtained.

9. The server configured with dual-carbon management mode as described in claim 7, characterized in that, The adjustment module includes: The fourth determining module is used for: Analyze the operational requirements to determine the value of the operational tasks that need to be completed within time t; Number the server nodes in the target server node set, denoted as i, i = 1, 2, 3, ..., n, where n is the number of server nodes in the target server node set; let P i P is the power consumption of the i-th server node during operation. is To enable the instantaneous power of the i-th server node, P is >P i Let F i F represents the calculated value when the i-th server node is running normally; t This represents the value of the running task that needs to be completed within time t; The fifth determining module is used for: When it is detected that the target server node set needs to be adjusted, the value F of the running task that needs to be completed within time t is... t At that time, determine the group S of server nodes that need to be in a running state from the target server node set. t for: ； in, This represents the inversion and transpose of a matrix; This indicates the running status of the i-th server node. This indicates that the i-th server node is in a closed state. This indicates that the i-th server node is active. The function represents array operations, where e is a natural constant; This indicates the instantaneous power consumption coefficient of some server nodes: ； The factor representing the cost-saving effect when some server nodes are shut down: ； in, To enable the instantaneous computation value of the i-th server node; Energy-saving strategies are generated based on the group of server nodes that need to be running in the target server node set.

Images