Early warning method and early warning device for monitoring environmental parameters of data center

Abstract

The invention discloses an early warning method and an early warning device for monitoring environmental parameters of a data center. The early warning method comprises the following steps: acquiring the environmental parameters of the data center detected by a sensor; acquiring environmental parameters of the data center according to the sliding time window and performing edge computing node storage, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters; analyzing and judging whether the numerical value in the first environment parameter is in a rising trend or not according to a preset algorithm; a dynamic early warning threshold value is calculated and determined according to the minimum value in the first environment parameters, a preset threshold value and a preset coefficient, the minimum value in the first environment parameters is in positive correlation with the dynamic early warning threshold value and the preset coefficient, and the dynamic early warning threshold value and the minimum value in the first environment parameters are smaller than the preset threshold value; and monitoring whether a numerical value exceeding a dynamic early warning threshold exists in the first environment parameter, and if yes, sending out early warning information. The environmental parameters of the data center can be accurately and flexibly judged, and the effectiveness and reliability of early warning are ensured.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an early warning method and an early warning device for monitoring environmental parameters of a data center. Background Art

[0002] With the acceleration of digital transformation, data centers have become one of the core infrastructures supporting the operation of modern society. Data centers are not only responsible for the storage, processing and transmission of massive amounts of data, but also carry various important business applications. In order to ensure the stable operation of high-performance computer equipment in data centers, data centers have extremely high requirements for environmental parameters, especially regarding temperature control.

[0003] Therefore, environmental parameter monitoring in data centers is crucial because it ensures that servers and other computer equipment can operate within the optimal temperature range. Effective environmental parameters can not only reduce the overall energy consumption of the data center, but also improve the reliability and service life of computer equipment. Abnormal environmental parameters can affect the performance of computer equipment and even trigger automatic shutdown mechanisms to prevent damage, which may cause business suspension. Summary of the invention

[0004] The embodiments of the present application provide an early warning method, device and electronic equipment for monitoring environmental parameters of a data center, which are used to accurately and flexibly judge the environmental parameters of the data center to ensure the effectiveness and reliability of the early warning.

[0005] In a first aspect, an embodiment of the present application provides an early warning method for monitoring environmental parameters of a data center, the method comprising:

[0006] Collecting environmental parameters of the data center detected by sensors;

[0007] Acquire the environmental parameters of the data center according to the sliding time window and store them in the edge computing node, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values;

[0008] Analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm;

[0009] The dynamic warning threshold is determined by calculating the minimum value of the first environmental parameter, the preset threshold value and the preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value;

[0010] Monitor whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, and if so, issue a warning message.

[0011] In the above method, by collecting the environmental parameters of the data center detected by the sensor, obtaining and storing the environmental parameters of the data center according to the sliding time window, and analyzing and judging whether the value of the first environmental parameter is on an upward trend according to the preset algorithm, that is, by collecting and analyzing the environmental parameters in real time, it is possible to capture subtle changes in the business environment in a timely manner, thereby achieving rapid and accurate early warning of environmental parameters such as temperature, and by judging the trend of the change of the numerical trajectory of the first environmental parameter within the sliding time window, intelligent support is provided for business decision-making, helping operation and maintenance personnel to take measures in advance to avoid potential risks;

[0012] Moreover, by calculating and determining the dynamic warning threshold based on the minimum value of the first environmental parameter in the sliding time window, the preset threshold and the preset coefficient, it can be ensured that the dynamic warning threshold is based on historical data and real-time data, and can be dynamically adjusted, so that the dynamic warning threshold is more in line with actual business needs and reduces the occurrence of false alarms and missed alarms.

[0013] By monitoring whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, if so, a warning message is issued. That is, accurate warning helps to timely discover and solve potential equipment failures, thereby improving the stability and reliability of the entire business system and ensuring the continuous operation of the business.

[0014] Optionally, the calculating and determining the dynamic warning threshold according to the minimum value of the first environmental parameter, the preset threshold and the preset coefficient specifically includes:

[0015] The preset threshold and the preset coefficient are input, the difference between the preset threshold and the minimum value of the first environmental parameter is calculated, the difference is multiplied by the preset coefficient and added to the minimum value of the first environmental parameter to obtain the dynamic warning threshold.

[0016] Optionally, the preset coefficient is proportional to the time when the warning information is issued.

[0017] Optionally, when the data center is in a peak operating period, the preset coefficient is a first preset coefficient;

[0018] When the data center is in a low-peak period, the preset coefficient is a second preset coefficient;

[0019] The first preset coefficient is greater than the second preset coefficient.

[0020] Optionally, when it is determined that the value of the first environmental parameter is on an upward trend, the value of the first environmental parameter is monitored in advance to determine whether to issue a warning message.

[0021] Optionally, before analyzing and determining whether the first environmental parameter is in an upward trend according to a preset algorithm, the step further includes:

[0022] A predicted time length is input, the first environmental parameter is obtained according to the predicted time length, and abnormal values ​​and invalid values ​​in the first environmental parameter are removed.

[0023] Optionally, before analyzing and determining whether the first environmental parameter is in an upward trend according to a preset algorithm, and after inputting a predicted time length, obtaining the first environmental parameter according to the predicted time length, and removing abnormal values ​​and invalid values ​​in the first environmental parameter, the method further includes:

[0024] A smoothing window length is input, and the first environmental parameter after removing abnormal values ​​and invalid values ​​is smoothed according to the smoothing window length.

[0025] Optionally, the preset algorithm includes the Mann-Kendall algorithm.

[0026] In a second aspect, an embodiment of the present application provides an early warning device for monitoring environmental parameters of a data center, including:

[0027] A collection module, used for collecting environmental parameters of the data center detected by sensors;

[0028] An edge computing node storage module is used to obtain the environmental parameters of the data center according to the sliding time window and store them in the edge computing node, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values;

[0029] An analysis module, used to analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm;

[0030] a calculation module, configured to calculate and determine a dynamic warning threshold value according to a minimum value of the first environmental parameter, a preset threshold value, and a preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value;

[0031] The early warning module is used to monitor whether there is a value in the first environmental parameter that exceeds the dynamic early warning threshold, and if so, issue an early warning message.

[0032] In a third aspect, a computer-readable storage medium stores a computer program, wherein the computer program is used to enable the computer to execute the above-mentioned early warning method.

[0033] The technical effects brought about by any one of the implementation methods in the second aspect and the third aspect can refer to the technical effects brought about by the corresponding implementation method in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of an application scenario of the early warning method provided in an embodiment of the present application;

[0035] Figure 2 A flowchart of an early warning method for monitoring data center environmental parameters provided in an embodiment of the present application;

[0036] Figure 3 A flowchart of an early warning method for monitoring data center environmental parameters provided in an embodiment of the present application;

[0037] Figure 4 A schematic diagram of the structure of an early warning device for monitoring environmental parameters of a data center provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be clearly and completely described below in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. And in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. Based on the described embodiments of the present application, all other embodiments obtained by ordinary technicians in the field without creative work belong to the scope of protection of this application.

[0039] Unless otherwise defined, the technical terms or scientific terms used in this application should be understood by people with ordinary skills in the field to which this application belongs. "First", "second" and similar words used in this application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and other similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and other similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0040] With the acceleration of digital transformation, data centers have become one of the core infrastructures supporting the operation of modern society. Data centers are not only responsible for the storage, processing and transmission of massive amounts of data, but also carry various important business applications. In order to ensure the stable operation of high-performance computer equipment in data centers, data centers have extremely high requirements for environmental parameters, especially regarding temperature control.

[0041] Therefore, environmental parameter monitoring in data centers is crucial because it ensures that servers and other computer equipment can operate within the optimal temperature range. Effective environmental parameters can not only reduce the overall energy consumption of the data center, but also improve the reliability and service life of computer equipment. Abnormal environmental parameters can affect the performance of computer equipment and even trigger automatic shutdown mechanisms to prevent damage, which may cause business suspension.

[0042] For example, the environmental parameter in the data center can be temperature. In the related art, the cold channel closure technology is usually used inside the data center, that is, by isolating the cold and hot air flow paths, ensuring that the cold air is directly delivered to key equipment such as servers and computers to achieve the best cooling effect. However, abnormal fluctuations in the temperature of the cold channel may cause the performance of servers and computers to decline, or even cause failures.

[0043] The establishment of the cold channel temperature early warning system is a profound innovation in the operation and maintenance management of data centers. The cold channel temperature early warning system in related technologies can detect temperature anomalies in a timely manner and issue early warnings by real-time monitoring of temperature data and combining historical trend analysis, providing decision support for operation and maintenance personnel, so that they can quickly take measures to adjust the cooling system to avoid potential equipment damage and business interruption. This early warning mechanism is not only to ensure the stable operation of the data center, but also to promote the development of the data center in a green, intelligent and efficient direction. And through precise temperature control, it can significantly reduce energy consumption, reduce carbon emissions, and achieve sustainable development. At the same time, the cold channel temperature early warning system can also improve the automation level of the data center, reduce the burden of manual inspections, improve operation and maintenance efficiency, and lay a solid foundation for the long-term and healthy development of the data center.

[0044] In the related art, data centers generally use the following methods to perform temperature warning:

[0045] For example, a fixed threshold warning method is used: that is, a fixed upper or lower temperature limit is set. When the monitored temperature exceeds or falls below this threshold, the system will issue a warning. This method is simple and direct, and easy to implement; but it lacks flexibility and cannot adapt to temperature changes under different equipment or environmental conditions. The setting of fixed thresholds may lead to frequent false alarms or missed alarms, which has a great impact on the work and operations of operation and maintenance personnel.

[0046] Exemplary, early warning method based on machine learning: using machine learning models, through a large amount of historical data training, predicting future temperature change trends, and issuing early warnings accordingly. This method has a certain degree of foresight and can better adapt to dynamically changing environments, but it requires a lot of data support and a professional technical team to maintain and optimize the model, and the training and maintenance costs of the model are high. In addition, the temperature data of the cold channel in the data center is relatively volatile and easily affected by on-site operation and maintenance, and the data trajectory prediction may be difficult.

[0047] For example, a real-time monitoring method based on the Internet of Things: by deploying a large number of sensors, temperature data is collected in real time, and data is transmitted and processed through the Internet of Things technology. This method can achieve real-time monitoring and rapid response to temperature changes, but it requires high hardware investment and maintenance costs, and when the sensor fails or the network is unstable, signal interruption or equipment shutdown may cause monitoring failure. In the event of downtime or serious situations, the operation and maintenance personnel cannot discover the problem in time, which will eventually cause great financial losses to the data center.

[0048] In summary, choosing the right warning method is crucial to ensure the stable operation of the data center. Fixed threshold warning is suitable for simpler environments, while machine learning-based warning and IoT-based real-time monitoring are more suitable for complex scenarios that require high accuracy and real-time response. The final choice should be based on the specific needs, budget, and technical capabilities of the data center.

[0049] Therefore, an embodiment of the present application provides an early warning method for monitoring environmental parameters of a data center, the method specifically comprising: collecting environmental parameters of the data center detected by sensors; obtaining the environmental parameters of the data center according to a sliding time window and storing them in an edge computing node, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values; analyzing and determining whether the values ​​in the first environmental parameters are on an upward trend according to a preset algorithm; determining a dynamic early warning threshold based on a minimum value, a preset threshold, and a preset coefficient in the first environmental parameter, wherein the minimum value in the first environmental parameter is positively correlated with the dynamic early warning threshold and the preset coefficient, and the dynamic early warning threshold and the minimum value in the first environmental parameter are less than the preset threshold; monitoring whether there are values ​​in the first environmental parameter that exceed the dynamic early warning threshold, and if so, issuing an early warning message.

[0050] In the embodiment of the present application, by collecting the environmental parameters of the data center detected by the sensor, obtaining the environmental parameters of the data center according to the sliding time window and storing them in the edge computing node, and judging whether the value of the first environmental parameter is on an upward trend according to the preset algorithm analysis, that is, by collecting and analyzing the environmental parameters in real time, it is possible to capture subtle changes in the business environment in a timely manner, thereby achieving rapid and accurate early warning of environmental parameters such as temperature, and by judging the change trend of the trajectory of the first environmental parameter within the sliding time window, intelligent support is provided for business decision-making, helping operation and maintenance personnel to take measures in advance to avoid potential risks;

[0051] Moreover, by calculating and determining the dynamic warning threshold based on the minimum value of the first environmental parameter in the sliding time window, the preset threshold and the preset coefficient, it can be ensured that the dynamic warning threshold is based on historical data and real-time data, and can be dynamically adjusted, so that the dynamic warning threshold is more in line with actual business needs and reduces the occurrence of false alarms and missed alarms.

[0052] By monitoring whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, if so, a warning message is issued. That is, accurate warning helps to timely discover and solve potential equipment failures, thereby improving the stability and reliability of the entire business system and ensuring the continuous operation of the business.

[0053] The embodiments of the present application not only help to optimize energy efficiency and reduce unnecessary energy consumption, but also significantly reduce operation and maintenance costs, thereby maximizing economic benefits by performing refined management of key environmental parameters such as temperature. In addition, the traditional machine learning and deep learning algorithms are abandoned, and the cumbersome process of model training for each data track is avoided, which saves a lot of resource costs and improves the operating efficiency of the system. It also simplifies the operating process and greatly reduces the consumption of time and resources. Moreover, the use of dynamic warning thresholds instead of traditional fixed warning thresholds can make more flexible and accurate judgments on the environmental parameters of the data center to ensure the effectiveness and reliability of the warning.

[0054] Exemplarily, a schematic diagram of an application scenario of the early warning method provided in the embodiment of the present application is as follows: Figure 1 As shown, it includes a server 100 and a terminal 101. The server 100 and the terminal 101 can be communicatively connected through a network to implement the early warning method for monitoring data center environmental parameters in the present application.

[0055] The user can use the server 100 to interact with the terminal 101 through the network, such as receiving or sending messages, etc. Various client applications can be installed on the terminal 101, such as program writing applications, web browser applications, search applications, etc.

[0056] It is understandable that in the embodiment of the present application, the server 100 can be implemented as an independent server or a server cluster composed of multiple servers. The terminal 101 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, desktop computers, etc.

[0057] The server 100 can be used to collect environmental parameters of a data center detected by sensors; obtain the environmental parameters of the data center according to a sliding time window and store them in an edge computing node, where the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values; determine whether the values ​​in the first environmental parameters are on an upward trend based on a preset algorithm analysis; determine a dynamic warning threshold based on a minimum value, a preset threshold, and a preset coefficient in the first environmental parameter, where the minimum value in the first environmental parameter is positively correlated with the dynamic warning threshold and the preset coefficient, and the dynamic warning threshold and the minimum value in the first environmental parameter are less than the preset threshold; monitor whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, and if so, issue a warning message.

[0058] like Figure 2 As shown, a flowchart of an early warning method for monitoring data center environmental parameters provided by an embodiment of the present application may specifically include the following steps:

[0059] Step S1, collecting environmental parameters of the data center detected by sensors;

[0060] Exemplarily, the environmental parameter may be temperature data information, and a temperature sensor may be installed in a cold channel in the data center. The temperature sensor detects the temperature of the cold channel in the data center and obtains the temperature data information, that is, the acquisition sensor detects the environmental parameters of the data center; this can help the user detect the temperature data information of the observed object in real time, wherein the observed object may be the cold channel in the data center.

[0061] Exemplarily, this step can be performed by deploying edge computing nodes to collect temperature data information of cold channels in the data center detected by temperature sensors in real time. The edge computing nodes are located at the source of data generation and can quickly process and analyze temperature data information, reducing the delay in temperature data information transmission.

[0062] Step S2: Acquire the environmental parameters of the data center according to the sliding time window and store them in the edge computing node. The environmental parameters of the data center stored in the sliding time window are the first environmental parameters.

[0063] Exemplarily, the server obtains the environmental parameters of the data center according to the sliding time window set by the user, and stores the received environmental parameters of the data center in the edge computing node, that is, stores them in the database of the deployed edge computing node server for subsequent analysis.

[0064] Exemplarily, this step can be performed by transmitting the temperature data information of the cold channel in the data center that has been preliminarily processed by the edge computing node to the server's TDengine time series database (TSDB), where TDengine is a high-performance, clustered open source, cloud-native, minimalist time series database designed specifically for IoT data. By using TDengine, it can be ensured that the temperature data information of the cold channel in the data center can be efficiently stored and managed, thus providing a solid foundation for subsequent data analysis and application.

[0065] Exemplarily, the sliding time window may be 5 minutes, 10 minutes, 15 minutes, etc. For example, the temperature data information of the cold channel in the data center detected by the temperature sensor may be automatically acquired and stored at a time frequency of 5 minutes;

[0066] That is, the temperature data information of the cold channel in the data center detected by the temperature sensor is obtained and stored every 5 minutes. The temperature data information of the cold channel in the data center detected by the temperature sensor during this period is the first temperature data information; the first temperature data information is the first environmental parameter; wherein, the sliding time window can be set according to demand and is not limited here.

[0067] Exemplarily, data transmission can be performed by wired means, for example, using wired communication protocols such as RS-485, RS-232, Ethernet, etc. to transmit temperature data information to the server. The wired method requires laying cables and configuring corresponding communication interfaces and protocols;

[0068] Of course, data can also be transmitted wirelessly, for example, using wireless communication technologies such as Wi-Fi, Bluetooth, LoRa, NB-IoT, etc. to transmit data to the server; the wireless method requires the use of wireless communication modules (such as ESP8266, ESP32, nRF52, etc.) that can be integrated into the data acquisition module or used as an independent device; and the parameters of the wireless communication module (such as SSID, password, baud rate, etc.) need to be configured, and code needs to be written to send temperature data information. You can choose to use wired or wireless methods according to your needs, and there is no limitation here.

[0069] In some embodiments of the present application, it also includes: inputting a predicted time length, obtaining a first environmental parameter according to the predicted time length, and removing abnormal values ​​and invalid values ​​in the first environmental parameter.

[0070] Exemplarily, the user inputs the predicted time length in the terminal, and the server obtains the first temperature data information based on the predicted time length, removes abnormal values ​​and invalid values ​​in the first temperature data information, that is, performs data cleaning on the first temperature data information in the database of the server, removes abnormal data values ​​caused by, for example, temperature sensor interruption, and removes invalid values ​​or abnormal values ​​in the first temperature data information, thereby ensuring the precision and accuracy of subsequent warnings.

[0071] Exemplarily, this step can be implemented using Python language.

[0072] In some embodiments of the present application, it also includes: inputting a smoothing window length, and smoothing the first environmental parameter after removing abnormal values ​​and invalid values ​​according to the smoothing window length.

[0073] Exemplarily, the user inputs the smoothing window length in the terminal, and the server smoothes the first environmental parameter after removing abnormal values ​​and invalid values ​​according to the smoothing window length, that is, smoothes the first temperature data information in the database of the server to remove short-term fluctuation data in the first temperature data information, so as to avoid frequent false alarms in subsequent warnings.

[0074] Exemplarily, this step can be implemented using the Rolling function in Python language.

[0075] Step S3: Analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm.

[0076] In some embodiments of the present application, the preset algorithm includes the Mann-Kendall algorithm.

[0077] For example, the Mann-Kendall algorithm is an effective nonparametric statistical method for testing trend changes in time series data. The core idea of ​​the Mann-Kendall algorithm is to determine whether there is an upward or downward trend in the entire data set by comparing the sequential relationship between data points. It does not require the data to follow a specific distribution, nor does it require the trend to be linear, so it has wide applicability.

[0078] Exemplarily, the server calls the Mann-Kendall algorithm to analyze and judge the first temperature data information after data cleaning and smoothing; the Mann-Kendall algorithm is called to determine the trend direction of the data and evaluate the significance of the trend, providing a scientific basis for subsequent dynamic warning threshold calculation and warning judgment, thereby ensuring the accuracy and timeliness of the temperature warning system.

[0079] It should be noted that when calling the Mann-Kendall algorithm, you need to call the result set (Result Set) in the Mann-Kendall algorithm, that is, you call an object containing multiple statistical results. This object contains the following properties: trend: Indicates the trend direction of the data, which can be increasing, decreasing, or not trending. p: Indicates the probability value of the trend significance, used to determine whether the trend is significant. z: Standardized test statistic, used to calculate p-value. Tau: The rank correlation coefficient of the Mann-Kendall algorithm, indicating the correlation between data. s: The test statistic of the Mann-Kendall algorithm. var_s: The variance of the test statistic s.

[0080] Exemplarily, the result set (Result Set) in the Mann-Kendall algorithm is called mainly to determine the trend direction of the data; determine whether the temperature data information is rising, falling or has no obvious trend; evaluate the significance of the trend; determine whether the trend is significant through p-value to avoid false alarms; provide statistical basis: provide scientific statistical basis for subsequent dynamic warning threshold calculation and warning judgment.

[0081] Exemplarily, this step may use the following code:

[0082] First, read the data in the Excel file, that is, use the read_excel_data function to read the time and data columns in the specified Excel file and form, including:

[0083] def read_excel_data(file_path,sheet_name,time_column,data_column):

[0084] """Read the time and data columns in the Excel file"""

[0085] df=pd.read_excel(file_path,sheet_name=sheet_name)

[0086] return df

[0087] Secondly, perform the Mann-Kendall trend test, that is, use the mann_kendall_test function to call the mk.original_test method to obtain the trend direction of the data, including:

[0088] def mann_kendall_test(data):

[0089] """Perform Mann-Kendall trend test"""

[0090] result=mk.original_test(data)

[0091] return result.trend

[0092] It should be noted that result is an object containing multiple statistical results. trend is an attribute of the result object, indicating the trend direction of the data. The mann_kendall_test function calls mk.original_test(data) to get the result object. The trend attribute is extracted from the result object to return the trend direction of the data.

[0093] Then, draw the result graph, that is, use the plot_results function to draw the data and the Mann-Kendall trend test results, including: the trend direction of the data, the starting value, the ending value, the difference, and the mean of the last five values, specifically including:

[0094] def plot_results(df,time_column,data_column,result):

[0095] """Plot the data and the Mann-Kendall trend test results"""

[0096] fig,ax=plt.subplots(figsize=(10,6))

[0097] df.plot(x=time_column,y=data_column,ax=ax,marker='o',label='Data')

[0098] # Calculate the difference between the start and end of the data

[0099] start_value=df[data_column].iloc[0]

[0100] end_value = df[data_column].iloc[-1]

[0101] difference = end_value - start_value

[0102] last_n_values = 5 # For example, the last 5 values

[0103] mean_last_values = df[data_column].tail(last_n_values).mean()

[0104] ax.axhline(y = start_value, color = 'g', linestyle = '--', label = f'StartValue:{start_value:.2f}')

[0105] ax.axhline(y = end_value, color = 'r', linestyle = '--', label = f'EndValue:{end_value:.2f}')

[0106] ax.text(df[time_column].iloc[len(df) / / 2], max(df[data_column]) * 0.9, f'Difference:{difference:.2f}',

[0107] fontsize = 12, color = 'b')

[0108] ax.set_title(f'trend:{result},difference:{difference},mean_last_5_values:{mean_last_values}')

[0109] ax.legend()

[0110] plt.show()

[0111] Finally, the main function specifically includes:

[0112] def main(file_path, sheet_name, time_column, data_column):

[0113] df=read_excel_data(file_path,sheet_name,time_column,data_column)

[0114] data = df[data_column].values

[0115] result=mann_kendall_test(data)

[0116] print(f"Mann-Kendall Test Result:{result}")

[0117] plot_results(df,time_column,data_column,result).

[0118] In some embodiments of the present application, when it is determined that the first environmental parameter is on an upward trend, the first environmental parameter is monitored in advance to determine whether to issue a warning message.

[0119] This step is mainly to warn of potential problems in advance when determining whether the first environmental parameter is on an upward trend. For example, to warn of potential temperature problems in advance; that is, to detect the upward trend in the first environmental parameter in a timely manner, so as to issue a warning before the first environmental parameter reaches the critical value, to avoid the first environmental parameter being too high and causing equipment failure or performance degradation. In addition, early detection of the upward trend of the first environmental parameter can help operation and maintenance personnel take preventive measures, such as adjusting the cooling system, checking the heat dissipation of the equipment, etc., to avoid potential risks.

[0120] Moreover, trend analysis can avoid false alarms caused by accidental fluctuations. Short-term fluctuations often exist in the first environmental parameters, which may be caused by first environmental factors or sensor errors. Trend analysis can filter out these short-term fluctuations to avoid false alarms. And improve alarm accuracy: Traditional static threshold methods are prone to false alarms and missed alarms because they cannot adapt to the dynamic changes of data centers. The preset algorithm can identify long-term trends, thereby improving the accuracy of alarms.

[0121] Step S4, calculate and determine the dynamic warning threshold value based on the minimum value of the first environmental parameter, the preset threshold value and the preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value.

[0122] In some embodiments of the present application, it specifically includes: inputting a preset threshold and a preset coefficient, calculating the difference between the preset threshold and the minimum value in the first environmental parameter, multiplying the difference with the preset coefficient and adding the minimum value in the first environmental parameter to obtain a dynamic warning threshold.

[0123] Exemplarily, the user inputs a preset coefficient and a preset threshold value in the terminal, and the server calculates and determines the dynamic warning threshold value according to a predefined calculation formula of the dynamic warning threshold value. The calculation formula of the predefined dynamic warning threshold value is: Among them, T 阈值 represents the dynamic warning threshold, T Min Represents the minimum value of the first environmental parameter, T Max Represents the preset threshold, level T Represents the preset coefficient.

[0124] <![CDATA[[T Min ,T Max ]]]> <![CDATA[Preset coefficient Level T : 0.6]]> <![CDATA[Preset coefficient Level T : 0.8]]> [18,25] 7*0.6+18＝22.2 7*0.8+18＝23.6 [19,25] 6*0.6+19＝22.6 6*0.8+19＝23.8 [20,25] 5*0.6+20＝23 5*0.8+20＝24 [20.5,25] 4.5*0.6+20.5＝23.2 4.5*0.8+20.5＝24.1 [21,25] 4*0.6+21＝23.4 4*0.8+21＝24.2 [21.25,25] 3.75*0.6+21.25＝23.5 3.75*0.8+21.25＝24.25 [22,25] 3*0.6+22＝23.8 3*0.8+22＝24.4 [23,25] 2.5*0.6+22.5＝24 2.5*0.8+22.5＝24.5

[0125] Table 1

[0126] Exemplarily, the calculation formula of the dynamic warning threshold is described in conjunction with Table 1;

[0127] When the preset threshold is 25 degrees and the preset coefficient is 0.6, the minimum value of the first environmental parameter is 18 degrees, the dynamic warning threshold is 22.2 degrees; the minimum value of the first environmental parameter is 19 degrees, the dynamic warning threshold is 22.6 degrees; the minimum value of the first environmental parameter is 20 degrees, the dynamic warning threshold is 23 degrees; the minimum value of the first environmental parameter is 20.5 degrees, the dynamic warning threshold is 23.2 degrees; the minimum value of the first environmental parameter is 21 degrees, the dynamic warning threshold is 23.4 degrees; the minimum value of the first environmental parameter is 21.25 degrees, the dynamic warning threshold is 23.5 degrees; the minimum value of the first environmental parameter is 22 degrees, the dynamic warning threshold is 23.8 degrees; the minimum value of the first environmental parameter is 23 degrees, the dynamic warning threshold is 24 degrees.

[0128] When the preset threshold is 25 degrees and the preset coefficient is 0.8, the minimum value of the first environmental parameter is 18 degrees, the dynamic warning threshold is 23.6 degrees; the minimum value of the first environmental parameter is 19 degrees, the dynamic warning threshold is 23.8 degrees; the minimum value of the first environmental parameter is 20 degrees, the dynamic warning threshold is 24 degrees; the minimum value of the first environmental parameter is 20.5 degrees, the dynamic warning threshold is 24.1 degrees; the minimum value of the first environmental parameter is 21 degrees, the dynamic warning threshold is 24.2 degrees; the minimum value of the first environmental parameter is 21.25 degrees, the dynamic warning threshold is 24.25 degrees; the minimum value of the first environmental parameter is 22 degrees, the dynamic warning threshold is 24.4 degrees; the minimum value of the first environmental parameter is 23 degrees, the dynamic warning threshold is 24.5 degrees.

[0129] In some embodiments of the present application, the preset coefficient is proportional to the time when the warning information is issued.

[0130] For example, the smaller the preset coefficient is, the earlier the warning information is issued; the larger the preset coefficient is, the later the warning information is issued. That is, the smaller the preset coefficient is, the earlier the warning time is, and preventive measures can be taken earlier. The larger the preset coefficient is, the more accurate the warning is, and the closer the alarm time is to the warning.

[0131] Therefore, the preset coefficients in the embodiments of the present application can be flexibly adjusted according to the actual environmental parameter settings of the data center and the requirements of different time periods.

[0132] In some embodiments of the present application, when the data center is in peak operating period, the preset coefficient is a first preset coefficient; when the data center is in off-peak operating period, the preset coefficient is a second preset coefficient; the first preset coefficient is greater than the second preset coefficient.

[0133] For example, during the peak working period of the data center, the preset coefficient can be set high to ensure more accurate warnings and avoid false alarms. During the off-peak working period of the data center, the preset coefficient can be set low to detect potential problems earlier and take measures in advance.

[0134] Step S5: monitor whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, and if so, issue a warning message.

[0135] The embodiments of the present application can adapt to the dynamic changes of the data center, thereby calculating and adjusting the dynamic warning threshold according to the dynamic changes of the data center, and in combination with the Mann-Kendall algorithm, it can identify the long-term trend of environmental parameters and calculate and adjust the dynamic warning threshold, so as to better adapt to the dynamic changes of the data center. And it can flexibly respond to different scenarios, that is, different data centers and different time periods, the characteristics of environmental parameter changes may be different. The calculation formula combined with the Mann-Kendall algorithm and the dynamic warning threshold can be flexibly applied to various scenarios to ensure the effectiveness and reliability of the warning.

[0136] like Figure 3 As shown, a flowchart of an early warning method for monitoring data center environmental parameters provided by an embodiment of the present application may include the following steps:

[0137] Step S101, collecting environmental parameters of the data center detected by sensors;

[0138] Step S102: Acquire the environmental parameters of the data center according to the sliding time window and store them in the edge computing node. The environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values.

[0139] Step S103: input the predicted time length, obtain the first environmental parameter according to the predicted time length, and remove abnormal values ​​and invalid values ​​in the first environmental parameter;

[0140] Step S104, inputting a smoothing window length, and performing smoothing processing on the first environmental parameter after removing abnormal values ​​and invalid values ​​according to the smoothing window length;

[0141] Step S105: Analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm;

[0142] Step S106, calculating and determining a dynamic warning threshold value according to the minimum value of the first environmental parameter, a preset threshold value and a preset coefficient;

[0143] Step S107: monitor whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, and if so, issue a warning message.

[0144] In summary, the embodiments of the present application perform fine-grained management of key environmental parameters such as temperature. The methods in the embodiments of the present application not only help to optimize energy efficiency and reduce unnecessary energy consumption, but also significantly reduce operation and maintenance costs, thereby maximizing economic benefits.

[0145] In addition, the method in the embodiment of the present application abandons the traditional machine learning and deep learning algorithms, avoids the tedious process of model training for each data track, saves a lot of resource costs, improves the operating efficiency of the system, simplifies the operation process, and greatly reduces the consumption of time and resources.

[0146] This is a schematic diagram of the structure of an early warning device for monitoring data center environmental parameters provided in an embodiment of the present application. Figure 4 As shown, the device comprises:

[0147] A collection module 10 is used to collect environmental parameters of the data center detected by sensors;

[0148] The edge computing node storage module 20 obtains the environmental parameters of the data center according to the sliding time window and performs edge computing node storage. The environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values.

[0149] An analysis module 30 is used to analyze and determine whether the value of the first environmental parameter is in an upward trend according to a preset algorithm;

[0150] A calculation module 40 is used to calculate and determine a dynamic warning threshold value according to a minimum value of the first environmental parameter, a preset threshold value, and a preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value;

[0151] The early warning module 50 is used to monitor whether there is a value in the first environmental parameter that exceeds the dynamic early warning threshold, and if so, issue an early warning message.

[0152] Based on the same concept, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions. The computer-executable instructions are used to enable a computer to execute an early warning method for monitoring data center environmental parameters in the above embodiment.

[0153] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0154] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0155] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0156] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0157] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0158] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. An early warning method for monitoring environmental parameters of a data center, characterized in that: include: Collecting environmental parameters of the data center detected by sensors; Acquire the environmental parameters of the data center according to the sliding time window and store them in the edge computing node, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values; Analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm; The dynamic warning threshold is calculated and determined according to the minimum value of the first environmental parameter, the preset threshold value and the preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value; Monitor whether there is a value in the first environmental parameter that exceeds the dynamic warning threshold, and if so, issue a warning message.

2. The early warning method for monitoring data center environmental parameters according to claim 1, characterized in that: The step of calculating and determining the dynamic warning threshold value according to the minimum value of the first environmental parameter, the preset threshold value and the preset coefficient specifically includes: The preset threshold and the preset coefficient are input, the difference between the preset threshold and the minimum value of the first environmental parameter is calculated, the difference is multiplied by the preset coefficient and added to the minimum value of the first environmental parameter to obtain the dynamic warning threshold.

3. The early warning method for monitoring data center environmental parameters according to claim 1, characterized in that: The preset coefficient is proportional to the time when the warning information is issued.

4. The early warning method for monitoring data center environmental parameters according to claim 3, characterized in that: When the data center is in a peak working period, the preset coefficient is a first preset coefficient; When the data center is in a low-peak period, the preset coefficient is a second preset coefficient; The first preset coefficient is greater than the second preset coefficient.

5. The early warning method for monitoring data center environmental parameters according to claim 1, characterized in that: When it is determined that the value of the first environmental parameter is on an upward trend, the value of the first environmental parameter is monitored in advance to determine whether to issue an early warning message.

6. The early warning method for monitoring data center environmental parameters according to claim 1, characterized in that: Before analyzing and determining whether the first environmental parameter is in an upward trend according to a preset algorithm, the method further includes: A predicted time length is input, the first environmental parameter is obtained according to the predicted time length, and abnormal values ​​and invalid values ​​in the first environmental parameter are removed.

7. The early warning method for monitoring data center environmental parameters according to claim 6, characterized in that: Before analyzing and determining whether the first environmental parameter is in an upward trend according to a preset algorithm, and after inputting a predicted time length, obtaining the first environmental parameter according to the predicted time length, and removing abnormal values ​​and invalid values ​​in the first environmental parameter, the method further includes: A smoothing window length is input, and the first environmental parameter after removing abnormal values ​​and invalid values ​​is smoothed according to the smoothing window length.

8. The early warning method for monitoring data center environmental parameters according to any one of claims 1 to 7, characterized in that: The preset algorithm includes the Mann-Kendall algorithm.

9. An early warning device for monitoring environmental parameters of a data center, characterized in that: include: A collection module, used for collecting environmental parameters of the data center detected by sensors; An edge computing node storage module is used to obtain the environmental parameters of the data center according to the sliding time window and store them in the edge computing node, wherein the environmental parameters of the data center stored in the sliding time window are first environmental parameters, and the first environmental parameters include multiple values; An analysis module, used to analyze and determine whether the value of the first environmental parameter is on an upward trend according to a preset algorithm; a calculation module, configured to calculate and determine a dynamic warning threshold value according to a minimum value of the first environmental parameter, a preset threshold value, and a preset coefficient, wherein the minimum value of the first environmental parameter is positively correlated with the dynamic warning threshold value and the preset coefficient, and the dynamic warning threshold value and the minimum value of the first environmental parameter are less than the preset threshold value; The early warning module is used to monitor whether there is a value in the first environmental parameter that exceeds the dynamic early warning threshold, and if so, issue an early warning message.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is used to enable the computer to execute the early warning method according to any one of claims 1 to 8.

Images