Control methods, devices, electronic equipment, and readable storage media for computer rooms
By acquiring data on external temperature and server load in the data center, and using predictive models to optimize the data center's operating mode, the shortcomings of existing data center energy-saving algorithms are addressed, achieving more efficient energy-saving results.
Patent Information
- Application Number
- CN202410109012.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-25
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2044-01-25
AI Technical Summary
In existing technologies, energy-saving algorithms for data center computer rooms are based solely on temperature sensing data within the computer room, which cannot achieve better energy-saving effects and fails to fully consider the impact of external ambient temperature and server load on cooling power consumption.
By acquiring current and future temperature data of the area where the data center is located, the system uses a trained load rate prediction model to predict server power consumption. Combining outdoor temperature and load rate scores, the system determines the data center score range and switches operating modes, including security and energy-saving modes, to optimize the data center's operating strategy.
It enables more accurate perception of changes in the external environment of the data center and its own load requirements, improving the energy efficiency of the data center and reducing cooling energy consumption.
Smart Images

Figure CN118258113B_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present invention relate to the field of energy-saving technology, and in particular to a control method for a computer room, a control device for a computer room, an electronic device, and a computer-readable storage medium. Background Technology
[0002] With the emergence of technologies such as cloud computing, big data, and virtualization, and the enhancement of computing power, AI (Artificial Intelligence) technology is maturing, and various industries are undergoing an intelligent transformation. Among them, data centers, with their massive data volume and high-performance computing capabilities, have become an important application area supporting the realization of AI.
[0003] The most pressing issue facing data centers today is energy consumption (such as the electricity required for air conditioning in server rooms). The enormous electricity costs have become a bottleneck for the rapid development of data centers. It is predicted that by 2030, the annual electricity consumption of telecommunications systems will account for more than 20% of the global annual electricity consumption, of which the energy consumption of data centers will reach 3000 TWh (terawatt-hour), accounting for 8% of the global annual electricity consumption. Therefore, research on energy-saving operation of data center server rooms has significant social and economic value.
[0004] Currently, most energy-saving algorithms for data centers are based on temperature sensing data and air conditioning data in the computer room to adjust the air conditioning. However, this method of adjusting the air conditioning in the computer room based solely on temperature sensing data cannot achieve better energy-saving results. Summary of the Invention
[0005] The present invention provides a control method, device, electronic equipment, and computer-readable storage medium for a data center to solve the problem that data center data centers cannot achieve better energy-saving effects.
[0006] This invention discloses a control method for a computer room, wherein the computer room is configured with several operating modes, and different operating modes have corresponding score ranges. The method includes:
[0007] Obtain current temperature data and future predicted temperature data for the area where the computer room is located;
[0008] The future power consumption of the server in the data center is predicted using a trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center.
[0009] The outdoor temperature score of the computer room is determined based on the current temperature data, the predicted future temperature data, and the preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room.
[0010] The load rate score of the data center is determined based on the future power consumption of the data center servers and the preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers.
[0011] The computer room score is determined based on the outdoor temperature score and the load rate score. The computer room score range into which the computer room score falls is taken as the target computer room score range. The computer room operation mode corresponding to the target computer room score range is taken as the target computer room operation mode, so as to control the computer room to operate according to the target computer room operation mode.
[0012] Optionally, obtaining the current temperature data and future predicted temperature data of the area where the computer room is located includes:
[0013] Obtain input information; the input information includes the current date, data center identifier, and data center address;
[0014] The corresponding data center is determined based on the data center identifier and the data center address, and data center data matching the current date is obtained. The data center data includes current temperature data, future predicted temperature data, preset temperature range, preset load rate range, historical temperature data, and historical data center server power consumption.
[0015] Optionally, before determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature range, the method further includes:
[0016] The historical temperature data is sorted according to a preset order to obtain the sorted historical temperature data;
[0017] The sorted historical temperature data is divided into several preset temperature intervals according to preset temperature quantiles, and the temperature state corresponding to each preset temperature interval is determined.
[0018] Optionally, before determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature range, the method further includes:
[0019] The historical power consumption of the data center servers is sorted according to a preset order to obtain the sorted historical power consumption of the data center servers.
[0020] The power consumption of the historical data center servers is divided into several preset load rate intervals according to the preset load rate quantiles, and the load rate status corresponding to each preset historical data center server power consumption interval is determined.
[0021] Optionally, before predicting the future power consumption of the data center servers using a preset load rate prediction model, the method further includes:
[0022] Collect training sample data for the data center; the training sample data includes historical temperature data of the area where the data center is located, historical cooling power consumption of the data center, and historical power consumption of the data center servers;
[0023] The training sample data is used to train the autoregressive moving average model to be trained, and the trained load rate prediction model is obtained when the autoregressive moving average model meets the preset convergence condition.
[0024] Optionally, each of the preset temperature ranges has a corresponding temperature state, and determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature range includes:
[0025] The temperature state corresponding to the preset temperature range that matches the current temperature data is obtained as the current temperature state, and the historical temperature score is calculated based on the current temperature state and the preset historical temperature weighting coefficient.
[0026] The temperature state corresponding to the preset temperature range that matches the predicted future temperature data is obtained as the future temperature state, and the future temperature score is calculated based on the future temperature state and the preset future temperature weighting coefficient.
[0027] The outdoor temperature score of the computer room is calculated based on the historical temperature score and the future temperature score.
[0028] Optionally, each of the preset load rate intervals has a corresponding load rate status, and determining the load rate score of the data center based on the future power consumption of the data center servers and the preset load rate intervals includes:
[0029] The load rate status corresponding to the preset load rate range that matches the future power consumption of the server in the data center is obtained as the future load rate status. The load rate score of the data center is calculated based on the future load rate status and the preset load weight coefficient.
[0030] Optionally, after determining the data center score based on the outdoor temperature score and the load rate score, taking the data center score range into which the data center score falls as the target data center score range, and taking the data center operation mode corresponding to the target data center score range as the target data center operation mode of the data center, so as to control the data center to operate according to the target data center operation mode, the method further includes:
[0031] Obtain the cooling power consumption of the computer room;
[0032] The energy-saving effect of the computer room is evaluated based on the cooling power consumption of the computer room.
[0033] Optionally, the data center operation mode includes a secure data center operation mode for ensuring the safety of the data center and an energy-saving data center operation mode for saving the power consumption of the data center. The energy-saving data center operation mode includes several energy-saving data center operation modes with different energy-saving levels.
[0034] This invention also discloses a control device for a computer room, wherein the computer room is configured with several computer room operation modes, and different computer room operation modes have corresponding computer room score ranges. The device includes:
[0035] The temperature data acquisition module is used to acquire the current temperature data and the predicted future temperature data of the area where the computer room is located;
[0036] The power consumption prediction module is used to predict the future power consumption of the server in the data center using a trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center.
[0037] The outdoor temperature score determination module is used to determine the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and a preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room.
[0038] The load rate score determination module is used to determine the load rate score of the data center based on the future power consumption of the data center servers and a preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers.
[0039] The data center operation mode determination module is used to determine the data center score range into which the data center score falls based on the outdoor temperature score and the load rate score, and to determine the data center operation mode corresponding to the target data center score range as the target data center operation mode of the data center, so as to control the data center to operate according to the target data center operation mode.
[0040] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;
[0041] The memory is used to store computer programs;
[0042] When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.
[0043] This invention also discloses a computer program product stored in a storage medium, which is executed by at least one processor to implement the method described in this invention.
[0044] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.
[0045] The embodiments of the present invention have the following advantages:
[0046] In this embodiment of the invention, the data center is configured with several data center operation modes. Different data center operation modes have corresponding data center score ranges. The current temperature data and future predicted temperature data of the area where the data center is located are obtained. The future power consumption of the data center servers is predicted using a trained load rate prediction model. The outdoor temperature score of the data center is determined based on the current temperature data, the future predicted temperature data, and the preset temperature range. The load rate score of the data center is determined based on the future power consumption of the data center servers and the preset load rate range. Then, the data center score is determined based on the outdoor temperature score and the load rate score. The data center score range into which the data center score falls is then taken as the target data center score range. The data center operation mode corresponding to the target data center score range is taken as the target data center operation mode. The data center is controlled to operate according to the target data center operation mode to achieve energy saving in the data center. This invention combines two key factors—the weather conditions in the area where the data center is located and the load of the servers in the data center—as the basis for deciding the data center's operating mode. Therefore, it can fully perceive changes in the external environment of the data center and the data center's own load requirements, making the switching of the data center's operating mode more accurate and intelligent, thereby enabling the data center to achieve better energy-saving effects. Attached Figure Description
[0047] Figure 1 This is a flowchart of the steps of a computer room control method provided in an embodiment of the present invention;
[0048] Figure 2 This is a schematic diagram of a temperature range division rule for a computer room provided in an embodiment of the present invention;
[0049] Figure 3 This is a schematic diagram of a load range division rule for a computer room provided in an embodiment of the present invention;
[0050] Figure 4 This is a schematic diagram of the overall structure for switching computer room operation modes provided in an embodiment of the present invention;
[0051] Figure 5 This is a schematic diagram of a data acquisition method for a computer room provided in an embodiment of the present invention;
[0052] Figure 6 This is a schematic diagram of an ARIMA (Autoregressive Moving Average) model construction provided in an embodiment of the present invention;
[0053] Figure 7 This is a structural block diagram of a control device for a computer room provided in an embodiment of the present invention;
[0054] Figure 8 This is a schematic diagram of the structure of a device provided in an embodiment of this application. Detailed Implementation
[0055] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0056] The relevant technical solutions do not consider the impact of the ambient temperature outside the data center and the server load on the cooling power consumption of the air conditioning system inside the data center. This leads to two problems: First, changes in outdoor temperature affect the indoor temperature of the data center. When the outdoor temperature rises, the indoor temperature also rises, which poses a certain risk. When the outdoor temperature falls, the indoor temperature also falls, resulting in a certain amount of redundant cooling capacity, thus failing to achieve the ideal energy-saving effect. Second, regarding server load, when the server load rate in a data center increases, the temperature inside the data center will also change accordingly. Failure to capture this change makes it impossible to predict risks and achieve better energy savings.
[0057] To address the aforementioned issues, this invention proposes a data center control method. This method sets up several data center operating modes with different energy-saving effects, determines different temperature ranges based on the historical temperature distribution of the data center's location, determines the server load rate (server load) within the data center based on the power consumption of the servers in the data center, and determines the current data center operating mode based on the temperature of the area where the data center is located and the predicted server load rate. This achieves better energy-saving results by switching the data center operating mode according to temperature changes.
[0058] Reference Figure 1The diagram illustrates a flowchart of a data center control method provided in an embodiment of the present invention. The data center is configured with several data center operating modes, and different data center operating modes have corresponding data center score ranges. The method may specifically include the following steps:
[0059] Step 101: Obtain the current temperature data and future predicted temperature data of the area where the computer room is located.
[0060] In this embodiment of the invention, the data center's server room can be configured with several server room operation modes, each with a corresponding server room score range. Specifically, the server room operation modes can include a secure server room operation mode to ensure server room security and an energy-saving server room operation mode to save server room power consumption. The energy-saving server room operation mode can include several energy-saving server room operation modes with different energy-saving levels. As an example of the invention, the server room operation modes can be configured as follows: ① Energy-saving mode: Allows the issuance of energy-saving strategies such as shutdown and cooling, with an issuance cycle of 30 minutes; ② Extreme energy-saving mode: Adjusts the energy-saving cycle to 15 minutes based on the energy-saving mode; ③ Security mode: Only allows the issuance of cooling energy-saving strategies, disallows shutdown, with an issuance cycle of 30 minutes; ④ Extreme security mode: Adjusts the energy-saving cycle to 60 minutes based on the security mode; ⑤ Default mode: Does not use algorithms to control server room energy saving. Of course, the above server room operation modes are merely examples. In practical applications, other server room operation modes can be set according to specific needs, and this embodiment of the invention does not impose any limitations on this.
[0061] For example, different data center operation modes each have corresponding data center score ranges. Assuming `score` represents the data center's score, the score range for the extreme energy-saving mode can be `score < -0.8`, the score range for the energy-saving mode can be `-0.8 ≤ score ≤ 0`, the score range for the security mode can be `0 < score ≤ 0.8`, and the score range for the extreme security mode can be `score > 0.8`.
[0062]
[0063] In this embodiment of the invention, the actual temperature data (current temperature data) for the current hour can be obtained hourly from a designated interface for the data center. Furthermore, the cooling energy consumption (cooling power consumption) and server energy consumption (server power consumption) of the data center can also be obtained. For example, the cooling energy consumption and server energy consumption for the day can be obtained at 23:59 every night. In addition, predicted future temperature data for the area where the data center is located can also be obtained. For example, weather forecast information for the area where the data center is located for the next fifteen days can be obtained starting at 00:00 every day, and the weather forecast information may include the predicted future temperature data.
[0064] Step 102: Predict the future power consumption of the server in the data center using the trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center.
[0065] In this embodiment of the invention, the historical power consumption of the server in the data center is collected in advance to train the load rate prediction model. Then, the trained load rate prediction model can be used to predict the future power consumption of the server in the data center. For example, the load rate prediction model can be used to predict the power consumption of the server in the data center in the next seven days, which can be used as the load rate of the data center.
[0066] Step 103: Determine the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and the preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room.
[0067] In this embodiment of the invention, historical temperature data of the computer room can be obtained in advance, such as three years of historical temperature data. Based on this historical data, the temperature can be divided into several temperature ranges, such as extremely cold, cold, warm, hot, and extremely hot temperature ranges. Then, after obtaining the current temperature data and the predicted future temperature data of the computer room, the outdoor temperature score of the computer room can be determined based on the temperature ranges that the current temperature data and the predicted future temperature data fall into.
[0068] Step 104: Determine the load rate score of the data center based on the future power consumption of the data center servers and the preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers.
[0069] In this embodiment of the invention, the historical power consumption of the data center servers can be obtained in advance, for example, the historical power consumption of the data center servers over the past year. Based on this historical power consumption, the server load rate can be divided into high load, balanced, and low load rate ranges. Then, by obtaining the future power consumption of the data center servers, the load rate score of the data center can be determined based on the load rate range into which the future power consumption falls.
[0070] Step 105: Determine the computer room score based on the outdoor temperature score and the load rate score, take the computer room score range into which the computer room score falls as the target computer room score range, and take the computer room operation mode corresponding to the target computer room score range as the target computer room operation mode, so as to control the computer room to operate according to the target computer room operation mode.
[0071] In this embodiment of the invention, after obtaining the outdoor temperature score and the server room load rate score, the outdoor temperature score and the server room load rate score can be processed by weighted averaging to obtain the final server room score. A lower server room score indicates lower temperature and lower load, and vice versa. The server room score range into which the server room score falls is taken as the target server room score range. Then, the server room operation mode corresponding to the target server room score range can be taken as the target server room operation mode to control the server room to operate according to the target server room operation mode. In this embodiment of the invention, the target server room operation mode and other related information can be written into a server room status table, and the server room operation mode can be adjusted according to this table. For example, assuming -2, -1, 0, 1, and 2 are used to represent extreme energy-saving mode, energy-saving mode, default mode, security mode, and extreme security mode respectively, if the target server room operation mode written in the server room status table is 0, then the target server room operation mode for this server room is the default mode.
[0072] As an example of the present invention, the computer room status table may include the following information:
[0073]
[0074] In this embodiment of the invention, the data center is configured with several data center operation modes. Different data center operation modes have corresponding data center score ranges. The current temperature data and future predicted temperature data of the area where the data center is located are obtained. The future power consumption of the data center servers is predicted using a trained load rate prediction model. The outdoor temperature score of the data center is determined based on the current temperature data, the future predicted temperature data, and the preset temperature range. The load rate score of the data center is determined based on the future power consumption of the data center servers and the preset load rate range. Then, the data center score is determined based on the outdoor temperature score and the load rate score. The data center score range into which the data center score falls is then taken as the target data center score range. The data center operation mode corresponding to the target data center score range is taken as the target data center operation mode. The data center is controlled to operate according to the target data center operation mode to achieve energy saving in the data center. This invention combines two key factors—the weather conditions in the area where the data center is located and the load of the servers in the data center—as the basis for deciding the data center's operating mode. Therefore, it can fully perceive changes in the external environment of the data center and the data center's own load requirements, making the switching of the data center's operating mode more accurate and intelligent, thereby enabling the data center to achieve better energy-saving effects.
[0075] In one embodiment of the present invention, step 101, obtaining the current temperature data and future predicted temperature data of the area where the computer room is located, includes:
[0076] Obtain input information; the input information includes the current date, data center identifier, and data center address;
[0077] The corresponding data center is determined based on the data center identifier and the data center address, and data center data matching the current date is obtained. The data center data includes current temperature data, future predicted temperature data, preset temperature range, preset load rate range, historical temperature data, and historical data center server power consumption.
[0078] In this embodiment of the invention, when it is necessary to control the operating mode of the data center, the user can input information. The server can then obtain the corresponding data center data, such as the current temperature data, future predicted temperature data, preset temperature range, preset load rate range, historical temperature data, historical power consumption of the data center servers, and power consumption of the data center cooling, based on the input current date, data center identifier (data center name), data center address, etc. The server can then determine the operating mode of the data center based on the data center data and control the data center to operate according to the operating mode, so as to achieve the effect of energy saving or safety of the data center.
[0079] In one embodiment of the present invention, before step 103, determining the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and the preset temperature range, the method further includes:
[0080] The historical temperature data is sorted according to a preset order to obtain the sorted historical temperature data;
[0081] The sorted historical temperature data is divided into several preset temperature intervals according to preset temperature quantiles, and the temperature state corresponding to each preset temperature interval is determined.
[0082] In this embodiment of the invention, historical temperature data of the computer room can be obtained in advance, such as three years of historical temperature data. The historical temperature data can then be divided into several temperature ranges based on preset temperature quantiles. Figure 2 This is a schematic diagram of a temperature range division rule for a computer room provided in an embodiment of the present invention. Temperature quantiles may include 92%, 71%, 29%, and 8%. Historical temperature data can be sorted according to a preset order (e.g., from high to low, i.e., descending order) to obtain sorted historical temperature data. Then, the historical temperature data is divided into multiple temperature ranges using temperature quantiles. For example, after obtaining the historical temperature data at the day level for the past three years, the historical temperature data can be sorted in descending order. Then, the temperature range with a temperature above the 92% quantile is considered extremely hot, the temperature range with a temperature above the 71% quantile but below the 92% quantile is considered hot, the temperature range with a temperature above the 29% quantile but below the 71% quantile is considered warm, the temperature range with a temperature above the 8% quantile but below the 29% quantile is considered cold, and the temperature range with a temperature below the 8% quantile is considered extremely cold.
[0083] Each temperature range has a corresponding temperature status, which can be used to score the outdoor temperature of the computer room. For example, the temperature range can be from -2 to 2. If the temperature falls into the extremely cold temperature range, the temperature status is -2; if the temperature falls into the cold temperature range, the temperature status is -1; if the temperature falls into the warm temperature range, the temperature status is 0; if the temperature falls into the hot temperature range, the temperature status is 1; and if the temperature falls into the extremely hot temperature range, the temperature status is 2.
[0084] In one embodiment of the present invention, before step 103, determining the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and the preset temperature range, the method further includes:
[0085] The historical power consumption of the data center servers is sorted according to a preset order to obtain the sorted historical power consumption of the data center servers.
[0086] The power consumption of the historical data center servers is divided into several preset load rate intervals according to the preset load rate quantiles, and the load rate status corresponding to each preset historical data center server power consumption interval is determined.
[0087] In this embodiment of the invention, the historical power consumption of the data center servers can be obtained in advance, for example, the historical power consumption of the data center servers over the past year. Then, based on a preset load rate percentile, the historical power consumption of the data center servers can be divided into several load rate intervals, with reference to... Figure 3 This is a schematic diagram of a load rate interval division rule for a data center provided in an embodiment of the present invention. The load rate percentile can include the 74% percentile and the 26% percentile. Then, the historical power consumption of the data center servers can be sorted in a preset order (e.g., from high to low, i.e., descending order) to obtain the sorted historical power consumption of the data center servers. Then, the historical power consumption of the data center servers can be divided into multiple load rate intervals using the load rate percentile. For example, after obtaining the historical power consumption of the data center servers over the past year, the historical power consumption of the data center servers can be sorted in descending order. Then, the historical power consumption of the data center servers with power consumption higher than the 26% percentile and lower than the 74% percentile can be identified as a balanced load rate interval, and the historical power consumption of the data center servers with power consumption lower than the 26% percentile can be identified as a low-load load rate interval.
[0088] Each load rate range has a corresponding load rate status, which can be used to score the computer room's load rate. For example, the load rate range can be from 1 to 0. If the server's power consumption falls into the high load rate range, the load rate status is 1; if the server's power consumption falls into the balanced load rate range, the load rate status is 0; and if the server's power consumption falls into the low load rate range, the load rate status is -1.
[0089] In one embodiment of the present invention, before step 102, which involves predicting the future power consumption of the data center servers using a preset load rate prediction model, the method further includes:
[0090] Collect training sample data for the data center; the training sample data includes historical temperature data of the area where the data center is located, historical cooling power consumption of the data center, and historical power consumption of the data center servers;
[0091] The training sample data is used to train the autoregressive moving average model to be trained, and the trained load rate prediction model is obtained when the autoregressive moving average model meets the preset convergence condition.
[0092] The Autoregressive Integrated Moving Average (ARIMA) model is a classic time series forecasting model used to analyze and predict time series data with autocorrelation and trend. The ARIMA model combines the concepts of autoregression (AR) and moving average (MA), and uses differencing (I) to handle non-stationary time series. Autoregression (AR) represents the relationship between the current observation and previous observations. The AR model uses observations from past time steps as input variables to predict the current value. The order (p) of the AR model indicates how many past time steps of observations need to be considered. Difference (I) is used to handle non-stationary time series. By differencing the original time series, a non-stationary series can be transformed into a stationary series. The order (d) of the differencing indicates how many differencing operations are required. Moving average (MA) represents the relationship between the current observation and previous prediction errors (residuals). The MA model uses previous prediction errors as input variables to predict the current value. The order (q) of the MA model indicates how many previous prediction errors need to be considered.
[0093] In this embodiment of the invention, historical server energy consumption data (historical server power consumption in the data center) for one year is obtained as training sample data. Data preprocessing (missing value and outlier handling) is performed, and the data is processed by differencing and moving average until the data is smooth (ADF test) and not white noise data (randomness test). The model order is determined according to the information criterion, and the AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) are quantitatively analyzed. The smaller the AIC and BIC, the better the model. After order determination, the ARIMA model can be constructed by combining the data to perform model prediction. The standard deviation between the predicted value and the true value is calculated to evaluate the model. When the model is determined to have reached the convergence condition after evaluation, the load rate prediction mode of this embodiment of the invention, which can be used to predict the future server power consumption of the data center, is obtained.
[0094] In one embodiment of the present invention, the preset temperature ranges each have corresponding temperature states, and determining the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and the preset temperature ranges includes:
[0095] The temperature state corresponding to the preset temperature range that matches the current temperature data is obtained as the current temperature state, and the historical temperature score is calculated based on the current temperature state and the preset historical temperature weighting coefficient.
[0096] The temperature state corresponding to the preset temperature range that matches the predicted future temperature data is obtained as the future temperature state, and the future temperature score is calculated based on the future temperature state and the preset future temperature weighting coefficient.
[0097] The outdoor temperature score of the computer room is calculated based on the historical temperature score and the future temperature score.
[0098] In this embodiment of the invention, different temperature ranges have corresponding temperature states, and the current temperature state is [status]. -i Where i ranges from 1 to 7, and the current temperature status ranges from -2 to 2. If the current temperature falls within the extremely cold temperature range, the current temperature status is -2; if it falls within the cold temperature range, the current temperature status is -1; if it falls within the warm temperature range, the current temperature status is 0; if it falls within the hot temperature range, the current temperature status is 1; and if it falls within the extremely hot temperature range, the current temperature status is 2. The same operation is performed to determine the future temperature status for the next seven days. i .
[0099] Then, the current temperature state is multiplied by the historical temperature weighting coefficient [0.01, 0.1, 0.12, 0.15, 0.18, 0.20, 0.24] to obtain the historical temperature score. history The future temperature is multiplied by a future temperature weighting coefficient [0.24, 0.20, 0.18, 0.15, 0.12, 0.10, 0.01] to obtain the future temperature score. future Then, the historical temperature score is calculated. history and future temperature score future The outdoor temperature score is the weighted average of these two temperature scores. tem .
[0100] Specifically, historical temperature score history Future temperature score future outdoor temperature score tem The specific calculation formula is as follows:
[0101]
[0102]
[0103] score tem =0.6*score history +0.6*score future
[0104] Where, weight -i Represents the historical temperature weighting coefficient, weight i The future temperature weighting coefficient is indicated by 0.6 and 0.4, which are weighting values.
[0105] In one embodiment of the present invention, the preset load rate intervals each have corresponding load rate states, and determining the load rate score of the data center based on the future power consumption of the data center servers and the preset load rate intervals includes:
[0106] The load rate status corresponding to the preset load rate range that matches the future power consumption of the server in the data center is obtained as the future load rate status. The load rate score of the data center is calculated based on the future load rate status and the preset load weight coefficient.
[0107] In this embodiment of the invention, different load rate ranges have corresponding load rate states, and the future load rate state is status_power. i Where i ranges from 1 to 7, and the future load status ranges from 1 to 0. If the future server power consumption falls into the high load range, the future load status is 1; if the future server power consumption falls into the balanced load range, the future load status is 0; and if the future server power consumption falls into the low load range, the future load status is -1. The value of `status_power` is based on the predicted future load status. i Multiplying by the pre-audit load weighting coefficient [0.24, 0.20, 0.18, 0.15, 0.12, 0.10, 0.01] yields the predicted load rate score for the data center. power .
[0108] In one embodiment of the present invention, after setting the data center score range into which the data center score falls as the target data center score range, and setting the data center operation mode corresponding to the target data center score range as the target data center operation mode of the data center, so as to control the data center to operate according to the target data center operation mode, the method further includes:
[0109] Obtain the cooling power consumption of the computer room;
[0110] The energy-saving effect of the computer room is evaluated based on the cooling power consumption of the computer room.
[0111] In this embodiment of the invention, the target operating mode of the data center is determined by the data center score. After controlling the data center to operate according to the target operating mode, the cooling power consumption of the data center after a period of time (e.g., one year) can be obtained. The energy-saving effect of the data center can then be evaluated based on the cooling power consumption. If the energy-saving effect is determined to be better than the prior energy-saving scheme based on the cooling power consumption, the data center control method of this embodiment of the invention can be continued. Conversely, if the energy-saving effect is determined to be worse than the prior energy-saving scheme or the same as the prior energy-saving scheme based on the cooling power consumption, the data center control method of this embodiment of the invention can be continuously optimized. For example, the data center scoring method or the data center operating mode can be optimized to enable the data center to achieve a better energy-saving effect.
[0112] This invention addresses the shortcomings of existing technologies by proactively sensing changes in the external temperature and server load within the data center to adaptively adjust the data center's operating mode. It provides a method for switching data center operating modes based on changes in outdoor weather and server load. To facilitate a better understanding of this invention, specific examples are provided below to illustrate the method for switching data center operating modes.
[0113] Specifically, a method for switching data center operation modes based on weather and server load changes according to an embodiment of the present invention may include the following steps:
[0114] Step 1: Data collection; Collect three years of historical daily temperature data for the area where the data center is located, as well as one year of historical cooling power consumption and server power consumption for the data center. Continuously collect daily actual temperature data for the area where the data center is located, future 15-day predicted temperature data, cooling energy consumption of the data center, and server energy consumption.
[0115] Step 2: Load rate prediction; Server power consumption is used as an assessment of server load, simply replacing the server load rate. An autoregressive moving average model (load rate prediction model) is built using the server power consumption data of the past year to gradually predict the server power consumption for the next seven days.
[0116] Step 3: Range Division; Divide the historical temperature data into five temperature ranges: extremely cold, cold, warm, hot, and extremely hot, based on the three-year daily temperature quantiles; divide the server load into three load rate ranges: high load, balanced, and low load, based on the historical server power consumption quantiles over one year.
[0117] Step 4: Data Acquisition; Enter the current date, data center name, and data center address to obtain data such as the temperature range, load rate range, future temperature, historical temperature, and historical load rate of the data center;
[0118] Step 5: Outdoor temperature score calculation; Match the future temperature and historical temperature with the five temperature ranges to obtain the historical temperature score and the future temperature score, and then weight them to obtain the outdoor temperature score.
[0119] Step 6: Load rate score calculation; Based on the load rate prediction model in Step 2, combined with the server load rate prediction for the next seven days based on the historical seven days of server load rate, the predicted load rate for the next seven days is matched with three load rate intervals to obtain the server load rate score for the data center.
[0120] Step 7: Determine the data center operation mode; calculate the final score (data center score) by weighting the outdoor temperature score and the server load rate score, and write the relevant information into the data center status table, so that the data center operation mode can be adjusted according to the data center status table.
[0121] In summary, in a specific embodiment, referring to Figure 4 This is a schematic diagram of the overall structure of a data center operation mode switching provided in an embodiment of the present invention. The embodiment of the present invention can be mainly divided into 5 modules: data acquisition, load rate prediction model, outdoor temperature score calculation, load rate score module, and operation mode determination.
[0122] Specifically, in step 1, combined Figure 5 The specific data collection operation is as follows:
[0123] Historical data collection: Collect actual temperature data for three years from the internet, and obtain cooling energy consumption and server energy consumption data for one year from the historical power consumption of the designated data center;
[0124] Actual data collection: Obtain the actual temperature data for the current hour from the specified interface every hour, and obtain the cooling energy consumption and server energy consumption of the computer room at 23:59 every night.
[0125] Forecast data collection: Starting at 00:00 every day, we acquire the weather forecast information for the next 15 days for the area where the computer room is located;
[0126] In step 2, combine Figure 6 The specific operation for load rate prediction is as follows:
[0127] Obtain historical server energy consumption data for one year and perform data preprocessing (handling missing values and outliers);
[0128] Perform differencing and moving average processing on the data until the data is smooth (ADF (Augmented Dickey-Fuller) test, a commonly used unit root test method, used to detect whether time series data has a unit root (i.e., non-stationarity)) and non-white noise data (randomness test);
[0129] The model order is determined based on the information criterion, and the AIC and BIC are quantitatively analyzed. The smaller the AIC and BIC, the better the model.
[0130] After determining the order, an ARIMA model can be constructed by combining the data, the model can be used for prediction, and the standard deviation between the predicted value and the true value can be calculated to evaluate the model.
[0131] Predict server energy consumption for the next seven days using models and historical data as load.
[0132] In step 5, the outdoor temperature score is calculated as follows:
[0133] Based on the historical temperature data (i.e., the current temperature data) falling within a specific temperature range determined in step 1, a list of historical temperature statuses is obtained. -i i takes values from 1 to 7, and the states are -2 to 2 depending on the temperature range: -2 for extreme cold, -1 for cold, 0 for warm, 1 for hot, and 2 for extreme heat.
[0134] The same procedure was performed to determine the future temperature status for the next seven days. i ;
[0135] The historical temperature score is obtained by multiplying the historical temperature status by the historical temperature weighting coefficient [0.01, 0.1, 0.12, 0.15, 0.18, 0.20, 0.24]. history The future temperature score is obtained by multiplying the future temperature by a future temperature weighting coefficient [0.24, 0.20, 0.18, 0.15, 0.12, 0.10, 0.01]. future The outdoor temperature score is obtained by weighting the two temperature scores together. tem ;
[0136] In step 6, the load factor score is calculated as follows:
[0137] Based on the predicted power load for the next seven days in step 2 i Based on the range determined in step 1, a value of 1 is assigned to high load, -1 to low load, and 0 to balanced load, thus determining the load status (status_power) for the next seven days. i ;
[0138] The predicted load score is obtained by multiplying the predicted load status by the predicted load weighting coefficient [0.24, 0.20, 0.18, 0.15, 0.12, 0.10, 0.01]. power .
[0139] In step 7, the final score (data center score) is obtained based on the outdoor temperature score and the load score. resultAccording to the score result The data center operation mode will be adjusted accordingly for the next seven days.
[0140] score result =0.37*score tem +0.63*score power
[0141] After obtaining accurate historical load information and relatively accurate weather forecast information, the future load of the data center can be predicted. Then, combined with the outdoor temperature, this helps guide the selection of subsequent energy-saving modes, thereby further achieving the energy-saving goals of the data center while meeting the requirements of safe production.
[0142] In summary, applying the embodiments of this invention brings at least the following beneficial effects: 1. Constructing an autoregressive moving average model for predicting server load rate in the data center, based on historical real server load data, makes the predicted data more reliable; 2. Combining historical real outdoor temperature and future forecast outdoor temperature as an outdoor temperature assessment, incorporating historical real temperature reduces the uncertainty of forecast temperature; 3. Combining two key factors—weather factors and data center server load—and using these two key factors as one of the bases for decision-making. The embodiments of this invention comprehensively consider these two key factors, fully sensing changes in the data center environment and its own load requirements, making the switching of data center operation modes more accurate and intelligent, achieving better data center energy-saving effects.
[0143] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.
[0144] Reference Figure 7 The diagram illustrates a structural block diagram of a control device for a computer room provided in an embodiment of the present invention. The computer room is equipped with several computer room operation modes, and different computer room operation modes have corresponding computer room score ranges. The device may specifically include the following modules:
[0145] Temperature data acquisition module 701 is used to acquire current temperature data and future predicted temperature data of the area where the computer room is located.
[0146] The power consumption prediction module 702 is used to predict the future power consumption of the server in the data center using a trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center.
[0147] The outdoor temperature score determination module 703 is used to determine the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and a preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room.
[0148] The load rate score determination module 704 is used to determine the load rate score of the data center based on the future power consumption of the data center servers and a preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers.
[0149] The data center operation mode determination module 705 is used to determine the data center score range into which the data center score falls based on the outdoor temperature score and the load rate score, and to determine the data center operation mode corresponding to the target data center score range as the target data center operation mode of the data center, so as to control the data center to operate according to the target data center operation mode.
[0150] In one embodiment of the present invention, the temperature data acquisition module 701 is used for:
[0151] Obtain input information; the input information includes the current date, data center identifier, and data center address;
[0152] The corresponding data center is determined based on the data center identifier and the data center address, and data center data matching the current date is obtained. The data center data includes current temperature data, future predicted temperature data, preset temperature range, preset load rate range, historical temperature data, and historical data center server power consumption.
[0153] In one embodiment of the present invention, the device further includes: a temperature range division module, used for:
[0154] The historical temperature data is sorted according to a preset order to obtain the sorted historical temperature data;
[0155] The sorted historical temperature data is divided into several preset temperature intervals according to preset temperature quantiles, and the temperature state corresponding to each preset temperature interval is determined.
[0156] In one embodiment of the present invention, the device further includes: a load rate interval division module, used for:
[0157] The historical power consumption of the data center servers is sorted according to a preset order to obtain the sorted historical power consumption of the data center servers.
[0158] The power consumption of the historical data center servers is divided into several preset load rate intervals according to the preset load rate quantiles, and the load rate status corresponding to each preset historical data center server power consumption interval is determined.
[0159] In one embodiment of the present invention, the apparatus further includes: a model training module, used for:
[0160] Collect training sample data for the data center; the training sample data includes historical temperature data of the area where the data center is located, historical cooling power consumption of the data center, and historical power consumption of the data center servers;
[0161] The training sample data is used to train the autoregressive moving average model to be trained, and the trained load rate prediction model is obtained when the autoregressive moving average model meets the preset convergence condition.
[0162] In one embodiment of the present invention, the outdoor temperature score determination module 703 is used for:
[0163] The temperature state corresponding to the preset temperature range that matches the current temperature data is obtained as the current temperature state, and the historical temperature score is calculated based on the current temperature state and the preset historical temperature weighting coefficient.
[0164] The temperature state corresponding to the preset temperature range that matches the predicted future temperature data is obtained as the future temperature state, and the future temperature score is calculated based on the future temperature state and the preset future temperature weighting coefficient.
[0165] The outdoor temperature score of the computer room is calculated based on the historical temperature score and the future temperature score.
[0166] In one embodiment of the present invention, the load rate score determination module 704 is used for:
[0167] The load rate status corresponding to the preset load rate range that matches the future power consumption of the server in the data center is obtained as the future load rate status. The load rate score of the data center is calculated based on the future load rate status and the preset load weight coefficient.
[0168] In one embodiment of the present invention, the device further includes: an energy-saving assessment module, used for:
[0169] Obtain the cooling power consumption of the computer room;
[0170] The energy-saving effect of the computer room is evaluated based on the cooling power consumption of the computer room.
[0171] In one embodiment of the present invention, the data center operation mode includes a secure data center operation mode for ensuring the safety of the data center and an energy-saving data center operation mode for saving the power consumption of the data center. The energy-saving data center operation mode includes several energy-saving data center operation modes with different energy-saving levels.
[0172] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.
[0173] In addition, this invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described computer room control method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.
[0174] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described computer room control method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.
[0175] This invention also provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described computer room control method embodiments, and can achieve the same technical effects. To avoid repetition, it will not be described again here.
[0176] The embodiments of this disclosure can be implemented as an apparatus configured as desired using any suitable hardware, firmware, software, or any combination thereof, including electronic devices such as terminal devices, servers (clusters), etc. Figure 8 An exemplary apparatus 800 is schematically shown that can be used to implement the various embodiments described in this application.
[0177] In one embodiment, Figure 8An exemplary device 800 is shown, which includes one or more processors 802, a control module (chipset) 804 coupled to at least one of the processors 802, a memory 806 coupled to the control module 804, a non-volatile memory (NVM) / storage device 808 coupled to the control module 804, one or more input / output devices 810 coupled to the control module 804, and a network interface 812 coupled to the control module 804.
[0178] Processor 802 may include one or more single-core or multi-core processors, and processor 802 may include any combination of general-purpose processors or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In some embodiments, device 800 can serve as a terminal device, server (cluster), or other device as described in the embodiments of this application.
[0179] In some embodiments, apparatus 800 may include one or more computer-readable media (e.g., memory 806 or NVM / storage device 808) having instructions 814 and one or more processors 802 that are combined with the one or more computer-readable media and configured to execute the instructions 814 to implement the module and thus perform the actions described in this disclosure.
[0180] In one embodiment, the control module 804 may include any suitable interface controller to provide any suitable interface to at least one of the processors 802 and / or any suitable device or component communicating with the control module 804.
[0181] The control module 804 may include a memory controller module to provide an interface to the memory 806. The memory controller module may be a hardware module, a software module, and / or a firmware module.
[0182] Memory 806 may be used, for example, to load and store data and / or instructions 814 for device 800. In one embodiment, memory 806 may include any suitable volatile memory, such as suitable DRAM. In some embodiments, memory 806 may include double data rate type quad synchronous dynamic random access memory (DDR4 SDRAM).
[0183] In one embodiment, the control module 804 may include one or more input / output controllers to provide an interface to the NVM / storage device 808 and (one or more) input / output devices 810.
[0184] For example, NVM / storage device 808 may be used to store data and / or instructions 814. NVM / storage device 808 may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable (one or more) non-volatile storage devices (e.g., one or more hard disk drives (HDDs), one or more optical disc drives (CDs), and / or one or more digital universal optical disc (DVD) drives).
[0185] NVM / storage device 808 may include storage resources that are physically part of a device on which device 800 is mounted, or that are accessible to the device but do not necessarily have to be part of the device. For example, NVM / storage device 808 may be accessed via a network via one or more input / output devices 810.
[0186] One or more input / output devices 810 may provide an interface for device 800 to communicate with any other suitable device. Input / output devices 810 may include communication components, audio components, sensor components, etc. A network interface 812 may provide an interface for device 800 to communicate via one or more networks. Device 800 may wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, such as accessing a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G, 5G, etc., or a combination thereof.
[0187] In one embodiment, at least one of the processors 802 may be logically packaged with one or more controllers (e.g., memory controller modules) of the control module 804. In one embodiment, at least one of the processors 802 may be logically packaged with one or more controllers of the control module 804 to form a system-in-package (SiP). In one embodiment, at least one of the processors 802 may be integrated with the logic of one or more controllers of the control module 804 on the same die. In one embodiment, at least one of the processors 802 may be integrated with the logic of one or more controllers of the control module 804 on the same die to form a system-on-a-chip (SoC).
[0188] In various embodiments, device 800 may be, but is not limited to, a server, desktop computing device, or mobile computing device (e.g., laptop, handheld computing device, tablet, netbook, etc.). In various embodiments, device 800 may have more or fewer components and / or different architectures. For example, in some embodiments, device 800 includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including a touchscreen display), a non-volatile memory port, multiple antennas, a graphics chip, an application-specific integrated circuit (ASIC), and a speaker.
[0189] The detection device may use a main control chip as a processor or control module, and sensor data, position information, etc. may be stored in a memory or NVM / storage device. The sensor group may be used as an input / output device, and the communication interface may include a network interface.
[0190] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.
[0191] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0192] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable computer room control terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable computer room control terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0193] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable computer room control terminal equipment to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0194] These computer program instructions can also be loaded onto a computer or other programmable terminal equipment, causing a series of operational steps to be executed on the computer or other programmable terminal equipment to produce a computer-implemented process. Thus, the instructions executing on the computer or other programmable terminal equipment provide the means for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0195] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.
[0196] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.
[0197] The above provides a detailed description of a control method and apparatus for a computer room, an electronic device, and a storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A control method for a computer room, characterized in that, The computer room is configured with several operating modes, and different operating modes have corresponding score ranges. The method includes: Obtain current temperature data and future predicted temperature data for the area where the computer room is located; The future power consumption of the server in the data center is predicted using a trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center. The outdoor temperature score of the computer room is determined based on the current temperature data, the predicted future temperature data, and the preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room. The load rate score of the data center is determined based on the future power consumption of the data center servers and the preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers. The computer room score is determined based on the outdoor temperature score and the load rate score. The computer room score range into which the computer room score falls is taken as the target computer room score range. The computer room operation mode corresponding to the target computer room score range is taken as the target computer room operation mode of the computer room, so as to control the computer room to operate according to the target computer room operation mode. The preset temperature ranges each have corresponding temperature states. Determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature ranges includes: The historical temperature score is calculated based on the current temperature status and the preset historical temperature weighting coefficient. The future temperature score is calculated based on the future temperature state and the preset future temperature weighting coefficient. The outdoor temperature score of the computer room is calculated based on the historical temperature score and the future temperature score. The preset load rate intervals each have corresponding load rate states. Determining the load rate score of the data center based on the future power consumption of the data center servers and the preset load rate intervals includes: The load rate score of the data center is calculated based on the future load rate status and the preset load weight coefficient. The future load rate status is the load rate status corresponding to a preset load rate range that matches the future power consumption of the server in the data center.
2. The method according to claim 1, characterized in that, The acquisition of current temperature data and predicted future temperature data for the area where the computer room is located includes: Obtain input information; the input information includes the current date, data center identifier, and data center address; The corresponding data center is determined based on the data center identifier and the data center address, and data center data matching the current date is obtained. The data center data includes current temperature data, future predicted temperature data, preset temperature range, preset load rate range, historical temperature data, and historical data center server power consumption.
3. The method according to claim 1, characterized in that, Before determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature range, the method further includes: The historical temperature data is sorted according to a preset order to obtain the sorted historical temperature data; The sorted historical temperature data is divided into several preset temperature intervals according to preset temperature quantiles, and the temperature state corresponding to each preset temperature interval is determined.
4. The method according to claim 1, characterized in that, Before determining the outdoor temperature score of the computer room based on the current temperature data, the predicted future temperature data, and the preset temperature range, the method further includes: The historical power consumption of the data center servers is sorted according to a preset order to obtain the sorted historical power consumption of the data center servers. The historical data center server power consumption is divided into several preset load rate intervals according to the preset load rate quantile, and the load rate status corresponding to each of the historical data center server power consumption intervals is determined.
5. The method according to claim 3, characterized in that, Before using the trained load rate prediction model to predict the future power consumption of the data center servers, the method further includes: Collect training sample data for the data center; the training sample data includes historical temperature data of the area where the data center is located, historical cooling power consumption of the data center, and historical power consumption of the data center servers; The training sample data is used to train the autoregressive moving average model to be trained, and the trained load rate prediction model is obtained when the autoregressive moving average model meets the preset convergence condition.
6. The method according to claim 3, characterized in that, The method further includes: Obtain the temperature state corresponding to the preset temperature range that matches the current temperature data as the current temperature state; The temperature state corresponding to a preset temperature range that matches the predicted future temperature data is obtained as the future temperature state.
7. The method according to claim 1, characterized in that, After determining the data center score based on the outdoor temperature score and the load rate score, taking the data center score range into which the data center score falls as the target data center score range, and taking the data center operation mode corresponding to the target data center score range as the target data center operation mode, and controlling the data center to operate according to the target data center operation mode, the method further includes: Obtain the cooling power consumption of the computer room; The energy-saving effect of the computer room is evaluated based on the cooling power consumption of the computer room.
8. The method according to claim 1, characterized in that, The data center operation modes include a secure data center operation mode to ensure the safety of the data center and an energy-saving data center operation mode to save power consumption in the data center. The energy-saving data center operation mode includes several energy-saving data center operation modes with different energy-saving levels.
9. A control device for a computer room, characterized in that, The computer room is equipped with several operating modes, and different operating modes have corresponding score ranges. The device includes: The temperature data acquisition module is used to acquire the current temperature data and future predicted temperature data of the area where the computer room is located; The power consumption prediction module is used to predict the future power consumption of the server in the data center using a trained load rate prediction model; the load rate prediction model is trained using the historical power consumption of the server in the data center. The outdoor temperature score determination module is used to determine the outdoor temperature score of the computer room based on the current temperature data, the future predicted temperature data, and a preset temperature range; the preset temperature range is determined based on the historical temperature data of the computer room. The load rate score determination module is used to determine the load rate score of the data center based on the future power consumption of the data center servers and a preset load rate range; the preset load rate range is determined based on the historical power consumption of the data center servers. The data center operation mode determination module is used to determine the data center score range into which the data center score falls based on the outdoor temperature score and the load rate score, and to determine the data center operation mode corresponding to the target data center score range as the target data center operation mode of the data center, so as to control the data center to operate according to the target data center operation mode. The preset temperature ranges each have corresponding temperature states. The outdoor temperature score determination module is specifically used to calculate the historical temperature score based on the current temperature state and a preset historical temperature weighting coefficient; to calculate the future temperature score based on the future temperature state and a preset future temperature weighting coefficient; and to calculate the outdoor temperature score of the computer room based on the historical temperature score and the future temperature score. The preset load rate intervals each have corresponding load rate states. The data center operation mode determination module is specifically used to calculate the load rate score of the data center based on the future load rate state and the preset load weight coefficient. The future load rate state is the load rate state corresponding to the preset load rate interval that matches the future power consumption of the data center servers.
10. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-8.
11. A computer-readable storage medium, characterized in that, It stores instructions that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-8.
Citation Information
Patent Citations
Method for dynamically forecasting external air and load intelligent energy saving control
CN107036238A
Air-conditioning control method and device based on load prediction and computer storage medium
CN109945420A
