A monitoring method and monitoring system for energy consumption in a data center

Abstract

The present invention relates to the field of data processing, and discloses a monitoring method and a monitoring system for the energy consumption of a data center. The method includes: Step 1, collecting the basic data of the data center; Step 2, periodically collecting the energy consumption data of the data center; Step 3, using a box plot to analyze each key energy consumption data in the energy consumption data, checking whether there are outliers, and if there are outliers, entering Step 5; Step 4, taking the area of the box plot where the key energy consumption data is located as the area value of the corresponding region, and the area values of each key energy consumption data form a control energy consumption vector, which is compared with the standard control energy consumption vector, and it is marked whether it is consistent with the standard control energy consumption vector; Step 5, calculating the energy consumption statistics according to the energy consumption data; Step 6, displaying the energy consumption statistics, the outliers, and the mark indicating whether the control energy consumption vector is consistent with the standard control energy consumption vector. The purpose of energy consumption monitoring is achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a monitoring method and a monitoring system for the energy consumption of a data center. Background Art

[0002] The annual electricity consumption of data centers in China has accounted for about 2% of the total electricity consumption of the whole society, and the data volume of data centers is still growing rapidly. To achieve the goal of reducing carbon emissions, it is necessary to further explore the energy-saving potential in aspects such as the construction mode of data centers, the technologies adopted, and the utilization of renewable energy.

[0003] At present, data centers mainly rely on the form of data reporting for energy consumption information. The main energy consumption data has no relevance, no data analysis tools, and no warning prompt information, which affects the monitoring personnel's mastery of the energy consumption information of data centers. Summary of the Invention

[0004] The purpose of the present invention is to provide a monitoring method and a monitoring system for the energy consumption of a data center, which uses the box plot algorithm to construct an analysis model for monitoring energy consumption, analyzes the key energy consumption data in the energy consumption data, prompts the monitoring personnel to process anomalies in a timely manner, and improves the degree of minimizing the consumed energy through the prompted energy consumption information.

[0005] To achieve the above object, an embodiment of the present invention provides a monitoring method for the energy consumption of a data center, including:

[0006] Step 1: Collect the basic data of the data center;

[0007] Step 2: Periodically collect the energy consumption data of the data center;

[0008] Step 3: Use the box plot to analyze each key energy consumption data in the energy consumption data, check whether there are outliers. If there are outliers, go to Step 5; if there are no outliers, go to Step 4;

[0009] Step 4: Take the area of the box plot where the key energy consumption data is located as the area value of the region. The area values of each key energy consumption data form a control energy consumption vector, which is compared with the standard control energy consumption vector, and it is marked whether it is consistent with the standard control energy consumption vector;

[0010] Step 5: Calculate the energy consumption statistics according to the energy consumption data;

[0011] Step 6: Display the energy consumption statistics, the outliers, the control energy consumption vector, and the mark indicating whether it is consistent with the standard control energy consumption vector.

[0012] An embodiment of the present invention further provides a monitoring system for the energy consumption of a data center, characterized by comprising: at least one processor; and,

[0013] a memory communicatively connected to the at least one processor; wherein,

[0014] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute a monitoring method for the energy consumption of a data center as described in any one of the above.

[0015] In the embodiment of the present invention, first, the basic data of the data center and the energy consumption data of the data center are periodically collected, then the box plot algorithm is used to analyze each key energy consumption data of the energy consumption data to find out the outliers and the control energy consumption vectors inconsistent with the standard control energy consumption vector, then the energy consumption data is statistically analyzed, and finally, the energy consumption data is displayed and a warning prompt message is given. That is, a judgment model for each key energy consumption data is constructed based on the box plot algorithm, and the energy consumption data can find outliers and judge whether the consumed energy is the minimum through the judgment model.

[0016] Preferably, the basic data of the data center in step 1 includes: the total designed number of racks and the building area.

[0017] Preferably, the energy consumption data in step 2 includes: cycle information, mains power, diesel power generation, gas power generation, renewable energy power generation, utilization of industrial waste energy, recovered and externally supplied energy, UPS power consumption including IT equipment load, refrigeration power consumption including cabinet refrigeration and UPS power supply refrigeration, power consumption of other supporting systems, non-data center power consumption, peak-valley power discharge, and off-peak cold storage and cold release.

[0018] That is to say, all information related to energy consumption is collected to statistically analyze and judge whether the consumed energy is the minimum consumed energy.

[0019] Preferably, the box plot in step 3 is a sequence sorted from small to large, and the upper limit value ULV, the upper quartile Q3, the median Q2, the lower quartile Q1, the lower limit value LLV, and the outlier EV of the sequence are respectively calculated.

[0020] wherein, Q1 = 0.25 × sequence item [round down the position of Q1 in the sequence] + 0.75 × sequence item [the position of Q1 in the sequence + 1],

[0021] Q2 = 0.5 × sequence item [round down the position of Q2 in the sequence] + 0.5 × sequence item [the position of Q2 in the sequence + 1],

[0022] Q3 = 0.75 × [rounded value of the sequence item at the Q3 sequence position] + 0.25 × [sequence item at the Q3 sequence position + 1],

[0023] Among them, the interquartile range IQR, IQR = Q3 - Q1,

[0024] ULV = Q3 + 1.5 × IQR,

[0025] LLV = Q1 - 1.5 × IQR,

[0026] Among them, EV < LLV, or, EV > ULV.

[0027] That is to say, the position of the key energy consumption data in the box plot can be calculated through the box plot algorithm.

[0028] Preferably, in step 3, the box plot is used to analyze the key energy consumption data of the energy consumption data, check whether there are outliers, and if the outliers are detected, a warning prompt message is displayed.

[0029] Preferably, in step 4, if the controlled energy consumption vector is inconsistent with the standard controlled energy consumption vector, a warning prompt message is displayed.

[0030] That is to say, if there are outliers in the key energy consumption data, and if the controlled energy consumption vector is inconsistent with the standard controlled energy consumption vector, a warning prompt message will be displayed.

[0031] Preferably, the standard controlled energy consumption vector is the vector that consumes the least energy.

[0032] Preferably, the energy consumption data in step 5 includes IT equipment load utilization rate, total energy equivalent value, total energy equivalent value, power usage effectiveness PUE, data center power utilization efficiency during the statistical period, renewable energy adjustment factor, peak-valley electricity adjustment factor, off-peak cold storage factor, IT equipment load utilization rate factor, data center comprehensive power utilization efficiency CPUE during the statistical period.

[0033] Preferably, the energy consumption data further includes the summary information of each data center, and the summary information includes the name, longitude and latitude of each data center, and the data center comprehensive power utilization efficiency CPUE during the statistical period marked with different colors. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. In the drawings:

[0035] Figure 1 It is the power distribution diagram of the data center energy consumption in the embodiment of the present invention;

[0036] Figure 2 It is the flowchart of the monitoring method for the energy consumption of the data center in the embodiment of the present invention;

[0037] Figure 3a It is the example diagram of the IT equipment load of the data center energy consumption in the embodiment of the present invention;

[0038] Figure 3b It is the example diagram of the cabinet refrigeration of the data center energy consumption in the embodiment of the present invention. Detailed implementation manners

[0039] The present invention provides many applicable creative concepts, and these creative concepts can be embodied in a large number in specific contexts. The specific embodiments described in the following embodiments of the present invention are only illustrative descriptions of the specific implementation manners of the present invention, and do not constitute a limitation on the scope of the present invention.

[0040] Figure 1 It is the power distribution diagram of the data center energy consumption, and the specific components are as follows:

[0041] The data center includes an energy supply side and an energy consumption side.

[0042] The energy supply side includes commercial power, diesel power generation, gas power generation, renewable energy power generation, power generation using industrial waste energy, and the energy recovered and supplied externally.

[0043] The energy consumption side includes UPS power consumption, refrigeration power consumption, power consumption of other supporting systems, and non-data center power consumption. Among them, the UPS power consumption includes IT equipment load, and the refrigeration power consumption includes UPS-powered refrigeration and cabinet refrigeration.

[0044] Collect the information of the above energy supply side and energy consumption side, form the energy consumption statistics of each data center, then collect the energy consumption statistics of each data center, and display the name, longitude and latitude of the data center, and the comprehensive power utilization efficiency CPUE of the data center during the statistical period marked with different colors on the map, so as to display the energy consumption situation of the regional data center.

[0045] Figure 2 It is the flowchart of the monitoring method for the energy consumption of the data center, and each step is as follows:

[0046] Step 1: The computer device collects the basic data of the data center.

[0047] Among them, the basic data of the data center includes: the total number of designed racks, building area.

[0048] Step 2: The computer device periodically collects the energy consumption data of the data center.

[0049] Among them, the energy consumption data includes: cycle information, mains electricity, diesel power generation, gas power generation, renewable energy power generation, utilization of industrial waste energy, recovered and externally supplied energy, UPS power consumption including IT equipment load, refrigeration power consumption including cabinet refrigeration and UPS power supply refrigeration, power consumption of other supporting systems, non-data center power consumption, peak-valley power discharge, and cold storage discharge during off-peak periods.

[0050] Step 3: The computer device uses a box plot to analyze each key energy consumption data in the energy consumption data, checks whether there are outliers. If there are such outliers, it enters Step 5; if there are no outliers, it enters Step 4.

[0051] Among them, the box plot is a sequence sorted from small to large, and the upper limit value ULV, the upper quartile Q3, the median Q2, the lower quartile Q1, the lower limit value LLV, and the outlier EV of the sequence are calculated respectively.

[0052] Among them, Q1 = 0.25 × sequence item [round down the position of Q1 in the sequence] + 0.75 × sequence item [round down the position of Q1 in the sequence + 1],

[0053] Q2 = 0.5 × sequence item [round down the position of Q2 in the sequence] + 0.5 × sequence item [round down the position of Q2 in the sequence + 1],

[0054] Q3 = 0.75 × sequence item [round down the position of Q3 in the sequence] + 0.25 × sequence item [round down the position of Q3 in the sequence + 1],

[0055] Among them, the interquartile range IQR, IQR = Q3 - Q1,

[0056] ULV = Q3 + 1.5 × IQR,

[0057] LLV = Q1 - 1.5 × IQR,

[0058] Among them, EV < LLV, or, EV > ULV.

[0059] According to the above box plot formula, check whether there are outliers in the key energy consumption data of the energy consumption data. If the outliers are detected, a warning prompt message needs to be displayed.

[0060] Step 4: The computer device takes the area of the box plot where the key energy consumption data is located as the area value of the location. The area values of each key energy consumption data form a control energy consumption vector, which is compared with the standard control energy consumption vector, and it is identified whether it is consistent with the standard control energy consumption vector.

[0061] If it is inconsistent with the standard control energy consumption vector, a warning prompt message is displayed.

[0062] Step 5: The computer device calculates the energy consumption statistics based on the energy consumption data.

[0063] Among them, the energy consumption data includes the IT equipment load utilization rate, the total equivalent value of energy, the total equivalent value of energy, the power usage effectiveness (PUE), the data center power utilization efficiency during the statistical period, the renewable energy adjustment factor, the peak-valley electricity adjustment factor, the off-peak cold storage factor, the IT equipment load utilization rate factor, and the data center comprehensive power utilization efficiency (CPUE) during the statistical period.

[0064] Step 6: Display the energy consumption statistics, the outliers, the control energy consumption vector, and the identifier indicating whether it is consistent with the standard control energy consumption vector.

[0065] Figure 3a It is an example diagram of the IT equipment load for the energy consumption of the data center.

[0066] The electricity consumption (kilowatts) consumed by the IT equipment load within a 2-week period is sorted from largest to smallest, and the data is: 12, 15, 17, 19, 20, 21, 23, 24, 26, 28, 29, 33, 36, 37; the upper limit value obtained according to the 2-week box plot algorithm: 47.25, the upper quartile: 30, the median: 23.5, the lower quartile: 18.5, the lower limit value: 1.25.

[0067] Among them, if there is an outlier: 49, it exceeds the upper limit value.

[0068] Figure 3b It is an example diagram of the cabinet cooling for the energy consumption of the data center.

[0069] The cabinet cooling (kilowatts) consumed by the IT equipment load within a 2-week period is sorted from largest to smallest, and the data is: 24, 30, 34, 40, 42, 44, 46, 48, 50, 56, 58, 66, 72, 76; the upper limit value obtained according to the 2-week box plot algorithm: 92.25, the upper quartile: 60, the median: 47, the lower quartile: 38.5, the lower limit value: 6.25.

[0070] And, Figure 3a the value of the IT equipment load in Figure 3b is consistent with the interval in the box plot of the value consumed by the cabinet cooling in

[0071] The positions of the box plots where the key energy consumption data, including IT equipment load and cabinet cooling, are located form a vector. When this vector is compared with the standard control energy consumption vector, if they are consistent, the consumed energy is the minimum energy. If they are inconsistent, a warning message is displayed to optimize the consumed energy further.

[0072] In summary, using box plots to analyze key energy consumption data, checking for outliers, and verifying whether the control energy consumption vectors corresponding to the key energy consumption data are consistent with the standard control energy consumption vector, and performing energy consumption statistics can help identify the reasons for abnormal or high energy consumption in the data center and quantify the degree of energy consumption through energy consumption statistics.

[0073] Another embodiment of the present application relates to a system, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method for intelligent automatic design of building wiring in the above embodiment.

[0074] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0075] That is, those skilled in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. This program is stored in a storage medium and includes several instructions to enable a device (such as a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, external hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0076] Those of ordinary skill in the art can understand that the above embodiments illustrate rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims.

Claims

1. A method for monitoring the energy consumption of a data center, characterized in that: Including the following steps: Step 1: Collect the basic data of the data center; Step 2: Periodically collect the energy consumption data of the data center; Step 3: Analyze each key energy consumption data in the energy consumption data using a box plot to check for outliers. If there are outliers, go to Step 5; if there are no outliers, go to Step 4; Step 4: Take the area of the box plot where the key energy consumption data is located as the area value of the location. The area values of each key energy consumption data form a control energy consumption vector, which is compared with the standard control energy consumption vector, and it is identified whether it is consistent with the standard control energy consumption vector; Step 5: Calculate the energy consumption statistics based on the energy consumption data; Step 6: Display the energy consumption statistics, the outliers, the control energy consumption vector, and the identification of whether it is consistent with the standard control energy consumption vector; Among them, in Step 3, the box plot is a sequence sorted from small to large. Calculate the upper limit value ULV, the upper quartile Q3, the median Q2, the lower quartile Q1, the lower limit value LLV, and the outlier EV of the sequence respectively; Among them, Q1 = 0.25×sequence item [round down the sequence position where Q1 is located] + 0.75×sequence item [the sequence position where Q1 is located + 1], Q2 = 0.5×sequence item [round down the sequence position where Q2 is located] + 0.5×sequence item [the sequence position where Q2 is located + 1], Q3 = 0.75×sequence item [round down the sequence position where Q3 is located] + 0.25×sequence item [the sequence position where Q3 is located + 1], Among them, the interquartile range IQR, IQR = Q3 - Q1, ULV = Q3 + 1.5×IQR, LLV = Q1 - 1.5×IQR, Among them, EV < LLV, or, EV > ULV; Among them, in Step 3, use the box plot to analyze the key energy consumption data in the energy consumption data to check for outliers. If the outliers are detected, a warning prompt message is displayed; Among them, if the control energy consumption vector in Step 4 is not consistent with the standard control energy consumption vector, a warning prompt message is displayed; Among them, the standard control energy consumption vector is the vector that controls the minimum energy consumed.

2. The monitoring method according to claim 1, characterized in that: The basic data of the data center in Step 1 includes: the total designed number of racks and the building area.

3. The monitoring method according to claim 1, characterized in that: The energy consumption data in Step 2 includes: cycle information, commercial power, diesel power generation, gas power generation, renewable energy power generation, utilization of industrial waste energy, recovered and externally supplied energy, UPS power consumption including IT equipment load, refrigeration power consumption including cabinet refrigeration and UPS power supply refrigeration, power consumption of other supporting systems, non-data center power consumption, peak-valley power discharge, and off-peak cold storage and cold release.

4. The monitoring method according to claim 1, characterized in that: The energy consumption data described in step 5 includes IT equipment load utilization rate, total equivalent value of energy, total equivalent value of energy, power usage effectiveness (PUE), data center power utilization efficiency during the statistical period, renewable energy adjustment factor, peak-valley electricity adjustment factor, off-peak cold storage factor, IT equipment load utilization rate factor, and data center comprehensive power utilization efficiency (CPUE) during the statistical period.

5. The monitoring method according to claim 4, characterized in that: The energy consumption data further includes summary information of each data center, and the summary information includes the name, location longitude and latitude of each data center, and the data center comprehensive power utilization efficiency (CPUE) during the statistical period marked with different colors.

6. A monitoring system for the energy consumption of a data center, characterized in that, Comprising: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute a method for monitoring energy consumption of a data center as described in any one of claims 1 to 5.

Images