A power consumption adjusting method, system and device

Abstract

The application discloses a power consumption adjusting method, system and device, and belongs to the technical field of power consumption control. The ambient temperature and resource utilization of each to-be-tested device in a current period are acquired; temperature region division is performed according to the ambient temperature; the score of each to-be-tested device is calculated according to the resource utilization, and the absolute value of the difference between the resource utilization of the to-be-tested device and the average resource utilization of all to-be-tested devices in the temperature region to which the to-be-tested device belongs is positively correlated with the score; and the resource utilization of the to-be-tested device is adjusted according to the order of the scores from large to small so that the score is reduced. When the to-be-tested device has a large difference with the average resource utilization of the temperature region to which the to-be-tested device belongs, the overall temperature of the temperature region will be changed, resulting in an increase in power consumption required for heat dissipation, and at this time, the difference with the average resource utilization needs to be reduced, so that the resource utilization of the to-be-tested device in each temperature region tends to be average, the temperature is maintained in a stable state, and the overall power consumption of the computer room is reduced.

Description

Technical Field

[0001] This invention relates to the field of power consumption control technology, and in particular to a power consumption regulation method, system and device. Background Technology

[0002] With the development of the Internet, the demand for large data centers is gradually increasing, and the demand for physical equipment in these data centers, especially servers, is increasing. This leads to a surge in power consumption, and the huge power consumption brings enormous usage and maintenance costs.

[0003] In existing technologies, reducing power consumption in data centers primarily involves two approaches: firstly, improving equipment utilization efficiency to reduce ineffective power consumption; and secondly, improving equipment distribution and ventilation within the data center to enhance air circulation efficiency and reduce heat dissipation pressure, thereby lowering the power consumption required for cooling. However, these two aspects are interconnected to some extent. Increased equipment utilization efficiency inevitably leads to a rise in temperature within the area, increasing the heat dissipation pressure on the data center. Furthermore, this temperature variation is not regular and cannot be reduced by adjusting equipment positions or adding cooling equipment. Therefore, how to reduce the overall power consumption of a data center is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a power consumption regulation method, system, and apparatus, which are applied in the field of power consumption control technology. When the difference between the device under test and the average resource utilization rate of its temperature zone is large, it will change the overall temperature of the temperature zone, resulting in increased power consumption required for heat dissipation. At this time, it is necessary to reduce the difference from the average resource utilization rate, so that the resource utilization rate of the device under test in each temperature zone tends to be average, the temperature is maintained in a stable state, and thus the overall power consumption of the computer room is reduced.

[0005] To address the aforementioned technical problems, this application provides a power consumption regulation method, comprising:

[0006] Obtain the ambient temperature and resource utilization rate of each device under test in the computer room during the current period;

[0007] Temperature zones are divided based on the ambient temperature of each of the devices under test.

[0008] The score of each device under test is calculated based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of each device under test and the average resource utilization rate of all devices under test in the temperature range is positively correlated with the score.

[0009] The resource utilization rate of the device under test is adjusted to reduce its score according to the descending order of the scores.

[0010] Preferably, the ambient temperature and resource utilization rate of each device under test in the computer room during the current period are obtained, including:

[0011] The ambient temperature of each device under test in the computer room during the current period is obtained by temperature sensors installed on each device under test, and the resource utilization rate of each device under test in the computer room during the current period is also obtained.

[0012] Preferably, the temperature zone is divided according to the ambient temperature of each of the devices under test, including:

[0013] S21: Divide the computer room into a set of N rows * M columns. Indicates the position of the Nth row and the Mth column;

[0014] S22: Create a new set Gi corresponding to the i-th temperature region, and identify the location of the device under test with the highest ambient temperature within the location set L.

[0015] S23: Determine the highest ambient temperature level, which corresponds one-to-one with the temperature range;

[0016] S24: Position The devices under test are assigned to the set Gi, and the spatially associated locations are determined. The devices under test are assigned to the preparatory set RG.

[0017] S25: Determine whether there is a device under test with the temperature level in the pre-set set RG. If yes, proceed to S26; otherwise, proceed to S27.

[0018] S26: The device under test with the temperature level is reclassified into the set Gi, and the device under test at the spatially associated position of the device under test reclassified into the set Gi is reclassified into the preliminary set RG, and then proceeds to S25.

[0019] S27: Determine whether the device under test exists in the preparatory set RG. If yes, proceed to S28; otherwise, stop.

[0020] S28: i = i + 1 and update the position set L to the preparatory set RG, then proceed to S22.

[0021] Preferably, the highest ambient temperature level is determined, and the temperature level corresponds one-to-one with the temperature range, including:

[0022] Determine the maximum, minimum, and average historical ambient temperatures of all the devices under test within the computer room;

[0023] When the highest ambient temperature is less than (a+b) / 2, the temperature level is determined to be "excellent", where a is the average of the historical ambient temperatures of all the devices under test, and b is the minimum of the historical ambient temperatures of all the devices under test.

[0024] When the highest ambient temperature is greater than (a+b) / 2 and less than a, the temperature level is determined to be "good".

[0025] When the highest ambient temperature is greater than a and less than (a+c) / 2, the temperature level is determined to be "medium", where c is the maximum historical ambient temperature of all the devices under test;

[0026] When the highest ambient temperature is greater than (a+c) / 2, the temperature level is determined to be "poor".

[0027] Preferably, the computer room is divided into a set of N rows * M columns. Following that, it also includes:

[0028] Remove locations from the location set L where the device under test cannot be placed.

[0029] Preferably, before obtaining the ambient temperature and resource utilization rate of each device under test in the computer room during the current period, the process further includes:

[0030] The current period is adjusted according to the frequency of tasks issued by external devices. The tasks need to be executed by the device under test, and the frequency of the tasks is negatively correlated with the current period.

[0031] Preferably, adjusting the resource utilization rate of the device under test to lower its score according to the descending order of the scores includes:

[0032] Based on the scores in descending order, the following steps are performed on the device under test:

[0033] When the resource utilization rate of the device under test is greater than the average resource utilization rate of all devices under test in the temperature range, the power consumption of the device under test is limited and external devices are prohibited from sending tasks to the device under test until its score drops below a preset threshold.

[0034] When the resource utilization rate of the device under test is less than the average resource utilization rate of all devices under test in the temperature range, the external device is given priority to send a task to the device under test until its score drops below the preset threshold.

[0035] Preferably, the resource utilization includes CPU utilization, memory utilization, and network utilization;

[0036] A score is calculated for each device under test based on its resource utilization rate. The absolute value of the difference between the resource utilization rate of each device under test and the average resource utilization rate of all devices under test within the same temperature range is positively correlated with the score, including:

[0037] According to the formula Calculate the score for each of the devices under test, where m(T) is the average ambient temperature of all the devices under test in the computer room, G is the temperature zone to which the device under test belongs, and m(T) G ) represents the average ambient temperature of all the devices under test within the temperature region G, D represents the CPU utilization of the device under test, and m(D) represents the average ambient temperature of all the devices under test within the temperature region G. G ) represents the average CPU utilization of all the devices under test within the temperature range G, E represents the memory utilization of the devices under test, and m(E) represents the average CPU utilization of all the devices under test. G ) represents the average memory utilization of all the devices under test within the temperature range G, F represents the network utilization of the devices under test, and m(F) represents the average memory utilization of all the devices under test. G ) represents the average network utilization of all the devices under test within the temperature range G.

[0038] To address the aforementioned technical problems, this application also provides a power consumption regulation system, comprising:

[0039] The data acquisition unit is used to acquire the ambient temperature and resource utilization rate of each device under test in the computer room during the current period.

[0040] A temperature zone division unit is used to divide the temperature zone according to the ambient temperature of each of the devices under test.

[0041] The scoring calculation unit is used to calculate the score of each of the devices under test based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of the device under test and the average resource utilization rate of all the devices under test in the temperature range is positively correlated with the score.

[0042] An adjustment unit is used to adjust the resource utilization rate of the device under test according to the order of the scores from largest to smallest, so that the scores are reduced.

[0043] To address the aforementioned technical problems, this application also provides a power consumption regulation device, comprising:

[0044] Memory, used to store computer programs;

[0045] A processor for executing the computer program to implement the power consumption regulation method.

[0046] This application provides a power consumption adjustment method, system, and apparatus, which are applied in the field of power consumption control technology. The method involves acquiring the ambient temperature and resource utilization rate of each device under test (DUT) within the current period; dividing the environment into temperature zones based on the ambient temperature; calculating a score for each DUT based on its resource utilization rate, where the absolute value of the difference between the DUT's resource utilization rate and the average resource utilization rate of all DUTs within its temperature zone is positively correlated with the score; and adjusting the resource utilization rate of the DUTs in descending order of their scores to lower their scores. When the difference between the DUT's resource utilization rate and the average resource utilization rate of its temperature zone is significant, it alters the overall temperature of the zone, leading to increased power consumption for heat dissipation. Therefore, it is necessary to reduce the difference from the average resource utilization rate, making the resource utilization rate of the DUTs within each temperature zone more even and maintaining a stable temperature, thereby reducing the overall power consumption of the computer room. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the prior art and embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a power consumption regulation method provided in this application;

[0049] Figure 2 This application provides a schematic diagram of the structure of a power consumption regulation system;

[0050] Figure 3 This is a schematic diagram of a power consumption regulation device provided in this application. Detailed Implementation

[0051] The core of this application is to provide a power consumption regulation method, system, and device, which is applied in the field of power consumption control technology. When the difference between the device under test and the average resource utilization rate of its temperature zone is large, it will change the overall temperature of the temperature zone, resulting in increased power consumption required for heat dissipation. At this time, it is necessary to reduce the difference from the average resource utilization rate, so that the resource utilization rate of the device under test in each temperature zone tends to be average, the temperature is maintained in a stable state, and thus the overall power consumption of the computer room is reduced.

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] Figure 1 A flowchart illustrating a power consumption regulation method provided in this application includes:

[0054] S11: Obtain the ambient temperature and resource utilization rate of each device under test in the computer room during the current period;

[0055] S12: Divide the temperature zones according to the ambient temperature of each device under test;

[0056] S13: Calculate the score of each device under test based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of the device under test and the average resource utilization rate of all devices under test in the temperature range is positively correlated with the score.

[0057] S14: Adjust the resource utilization rate of the device under test according to the order of scores from highest to lowest to reduce its score.

[0058] With the development of the Internet, the demand for large data centers is gradually increasing. The demand for physical equipment in data centers, especially servers, is increasing day by day. This leads to an ever-increasing demand for power consumption, and the huge power consumption brings huge usage and maintenance costs.

[0059] In existing technologies, reducing power consumption in data centers primarily involves two approaches: firstly, improving equipment utilization efficiency by detecting underutilized "zombie" devices and reducing ineffective power consumption; secondly, improving equipment distribution and ventilation to enhance air circulation efficiency and reduce heat dissipation pressure, thereby lowering the power consumption required for cooling. However, these two aspects are interconnected to some extent. Increased equipment utilization efficiency inevitably leads to a rise in temperature within the area, increasing the heat dissipation pressure on the data center. Furthermore, this temperature variation is not regular and cannot be reduced simply by adjusting equipment locations or adding cooling equipment. Therefore, how to reduce the overall power consumption of a data center is a technical problem that urgently needs to be solved by those skilled in the art.

[0060] To address the aforementioned technical issues, this application first divides the computer room into different temperature zones, then balances the resource utilization of the devices under test within each temperature zone, thereby balancing the heat generated during the operation of the devices under test and ultimately reducing the overall power consumption of the computer room.

[0061] Specifically, in S11, the ambient temperature and resource utilization of each device under test are periodically acquired, and then the temperature zones are periodically divided. The resource utilization of the devices under test in each temperature zone is then balanced, and the heat generated by the devices under test during operation is balanced. Finally, the temperature of adjacent temperature zones is balanced, thereby reducing the overall power consumption.

[0062] Considering that if the device under test in the data center is a server, multiple servers are usually housed in a rack. In this case, the ambient temperature is obtained on a per-rack basis, specifically the average ambient temperature of all servers within the rack. However, resource utilization is measured on a per-server basis, meaning the resource utilization of each server is obtained. This resource utilization can include CPU utilization, memory utilization, and network utilization.

[0063] In S12, temperature zones are divided. For example, devices under test with ambient temperatures above a certain value are assigned to one temperature zone, and devices with ambient temperatures below a certain value are assigned to another. Balancing resource utilization based on temperature zones is more organized and efficient than directly balancing resource utilization for all devices under test in the computer room.

[0064] In S13, within a temperature range, if the difference between the resource utilization rate of a device under test and the average resource utilization rate of all devices under test within that temperature range is greater, it indicates that the device under test does not meet the operating benchmark that a device under test should have within that temperature range. Its resource utilization rate needs to be adjusted to avoid the waste of power consumption caused by excessively low resource utilization rate, or to avoid the increase in heat dissipation power consumption caused by excessively high resource utilization rate.

[0065] In S14, the resource utilization rate of the device under test is adjusted sequentially according to the score from largest to smallest. Specifically, the resource utilization rate can be controlled by controlling the tasks sent to the device under test (the more tasks sent, the greater the resource utilization rate), so as to reduce the difference between the device under test and the average resource utilization rate of all devices under test in the temperature range, thereby reducing the score and achieving the average resource utilization rate of the devices under test in the temperature range.

[0066] By continuously repeating the above process, temperature zones are divided based on real-time data from each device under test. By scoring the devices under test within each temperature zone, nodes to be optimized are selected, i.e., devices under test with high scores. This reduces the impact of the data center's own design, making the data center's operating status, including the operating load of the devices under test and the temperature of the temperature zones, closer to the ideal average value, thereby reducing the data center's operating costs and improving its effective utilization rate.

[0067] Furthermore, all of the above steps can be achieved through ISPIM (Inspur Physical Infrastructure Management), which provides users with functions such as asset management, intelligent monitoring, energy management, automatic inspection and maintenance, stateless management, and report statistical analysis. It enables unified management of servers, storage, network devices, and firewalls, effectively helping enterprise users improve operational efficiency, reduce operational costs, and ensure the stable operation of data centers.

[0068] In summary, this application provides a power consumption adjustment method, which is applied in the field of power consumption control technology. The method involves acquiring the ambient temperature and resource utilization rate of each device under test (DUT) within the current period; dividing the DUT into temperature zones based on the ambient temperature; calculating a score for each DUT based on its resource utilization rate, where the absolute value of the difference between the DUT's resource utilization rate and the average resource utilization rate of all DUTs within its temperature zone is positively correlated with the score; and adjusting the resource utilization rate of the DUTs in descending order of their scores to lower their scores. When the difference between the DUT's resource utilization rate and the average resource utilization rate of its temperature zone is large, it will change the overall temperature of the temperature zone, leading to increased power consumption for heat dissipation. Therefore, it is necessary to reduce the difference from the average resource utilization rate, making the resource utilization rate of the DUTs within each temperature zone more even, maintaining a stable temperature, and thus reducing the overall power consumption of the computer room.

[0069] Based on the above embodiments:

[0070] As a preferred embodiment, the ambient temperature and resource utilization rate of each device under test in the computer room during the current period are obtained, including:

[0071] The ambient temperature of each device under test in the computer room during the current period is obtained by temperature sensors installed on each device under test, and the resource utilization rate of each device under test in the computer room during the current period is also obtained.

[0072] In this embodiment, the ambient temperature can be obtained through temperature sensors on each device under test, reflecting the temperature conditions around each device under test, and is used to divide the computer room into temperature zones.

[0073] As a preferred embodiment, temperature zones are divided based on the ambient temperature of each device under test, including:

[0074] S21: Divide the computer room into a set of N rows * M columns. Indicates the position of the Nth row and the Mth column;

[0075] S22: Create a new set Gi corresponding to the i-th temperature region, and identify the location of the device under test with the highest ambient temperature within the location set L.

[0076] S23: Determine the temperature level of the highest ambient temperature, with each temperature level corresponding to a temperature range;

[0077] S24: Position The devices under test are grouped into set Gi and their spatially associated locations are determined. The devices under test are assigned to the preliminary set RG.

[0078] S25: Determine whether there is a device under test with a temperature level in the preparatory set RG. If yes, proceed to S26; otherwise, proceed to S27.

[0079] S26: Reclassify the devices under test with temperature levels into set Gi, and classify the devices under test at the spatial correlation positions of the devices under test reclassified into set Gi into set RG, and proceed to S25.

[0080] S27: Determine whether there is a device under test in the preparatory set RG. If yes, proceed to S28; otherwise, stop.

[0081] S28: i = i + 1 and update the position set L to the preparatory set RG, then proceed to S22.

[0082] In this embodiment, the specific process of dividing the temperature zones is defined as follows: First, a data center map is established. The data center map is created in a parameterized manner based on the actual data center, that is, a set of N rows * M columns of locations. Then, temperature zones are quickly divided using the constructed location set L and the real-time collected ambient temperature.

[0083] Specifically, in S21, a data center map is created. First, a coordinate index map corresponding to the actual data center is established. Using the device under test as the basic location unit, the data center is parameterized into a set of N rows * M columns of locations. This represents the position of the first row and second column with coordinates (1, 2). The actual location of each device to be monitored within the computer room can be manually marked.

[0084] In S22-S24, for example, a set G1 corresponding to the first temperature region is established, and the location of the device under test with the highest ambient temperature within the location set L is determined. Position The device under test is assigned to set G1 and used as the starting point of the set, and its position is... The devices under test in the surrounding locations are divided into a preliminary set RG. The surrounding locations of each location are eight locations within a 3x3 area centered on that location, excluding the location itself. These locations can be spatially associated using a location index L. To quickly find surrounding locations, the superscript of each L must be greater than or equal to 1 and less than or equal to N, and the subscript of each L must be greater than or equal to 1 and less than or equal to M. If these conditions cannot be met, it means that the location L does not exist. For example... Representative position The surrounding area is and

[0085] In S25-S26, the ambient temperature of the device under test (DUT) at the location of the device under test with the highest ambient temperature is also very likely to be high. Therefore, we continue to search for the DUT at the location of the device under test, i.e., the preliminary set RG, to form a set G1. At the same time, we update the preliminary set RG until there is no DUT at the same temperature level in the preliminary set RG.

[0086] In steps S27-S28, after confirming that there are no devices under test (DUTs) with the same temperature level in the preliminary set RG, it is determined whether all DUTs in the preliminary set RG have been divided. If not, a second temperature zone set G2 is created, and the location of the DUT with the highest ambient temperature in the preliminary set RG is determined. The temperature range of the second temperature zone is smaller than that of the first temperature zone. Ultimately, the computer room is divided into multiple temperature zones, each with the same temperature level. All DUTs within each zone are adjacent, and there are no isolated DUTs.

[0087] In summary, by finding the device under test with the highest ambient temperature and establishing a temperature region based on its surrounding location, the process of dividing the temperature region is accelerated.

[0088] As a preferred embodiment, a temperature level with the highest ambient temperature is determined, and the temperature levels correspond one-to-one with the temperature ranges, including:

[0089] Determine the historical maximum, minimum, and average ambient temperatures of all devices under test in the computer room;

[0090] When the highest ambient temperature is less than (a+b) / 2, the temperature level is judged as "excellent", where a is the average of the historical ambient temperatures of all devices under test, and b is the minimum of the historical ambient temperatures of all devices under test.

[0091] When the highest ambient temperature is greater than (a+b) / 2 but less than a, the temperature level is determined to be "good".

[0092] When the highest ambient temperature is greater than a and less than (a+c) / 2, the temperature level is determined to be "medium", where c is the maximum historical ambient temperature of all devices under test.

[0093] When the highest ambient temperature is greater than (a+c) / 2, the temperature level is determined to be "poor".

[0094] In this embodiment, the temperature levels can be divided according to empirical values. The maximum value c, minimum value b, and average value a of the historical temperature data of the equipment under test in the computer room are statistically analyzed. The ambient temperature is divided into four levels: "Excellent", "Good", "Medium" and "Poor". In the process of periodically dividing the temperature areas and adjusting the resource utilization rate according to the scores, the historical temperature data is continuously refreshed, thereby continuously optimizing the numerical range of each temperature area.

[0095] As a preferred embodiment, the computer room is divided into a set of locations in N rows * M columns. Following that, it also includes:

[0096] Remove locations in location set L where the device under test cannot be placed.

[0097] Considering that in reality, some locations in the computer room cannot be used to place the device under test, such as the locations of air conditioning cabinets, columns, partitions, etc., or some locations that are not planned as locations for the device under test, these locations need to be removed from the location set L to avoid interference with subsequent steps performed through set L.

[0098] As a preferred embodiment, before obtaining the ambient temperature and resource utilization rate of each device under test in the computer room during the current period, the method further includes:

[0099] The current cycle size is adjusted according to the frequency of tasks issued by external devices. The task issuance requires execution by the device under test, and the frequency of task issuance is negatively correlated with the current cycle.

[0100] Since the indicators used in the process of obtaining the evaluation of the device under test (DUT), namely ambient temperature and resource utilization rate, are all real-time operating data of the DUT, the impact of the evaluation on resource utilization rate can be adjusted by adjusting the frequency of data acquisition. The frequency of data acquisition can be adjusted appropriately according to the actual application. For example, during peak periods, when the frequency of tasks issued by external devices is high, the acquisition frequency can be increased (i.e., the current cycle size can be decreased), thereby improving the real-time performance of the power consumption adjustment method in this application. During off-peak periods, when the frequency of tasks issued by external devices is low, the acquisition frequency can be decreased (i.e., the current cycle size can be increased), reducing the additional resource consumption caused by processing the operating data of the DUT.

[0101] As a preferred embodiment, adjusting the resource utilization rate of the device under test to lower its score according to the descending order of scores includes:

[0102] Based on the scores from highest to lowest, the following steps are performed on the device under test:

[0103] When the resource utilization rate of the device under test is greater than the average resource utilization rate of all devices under test in the temperature range, limit the power consumption of the device under test and prohibit external devices from sending tasks to the device under test until its score drops below the preset threshold.

[0104] When the resource utilization rate of the device under test is less than the average resource utilization rate of all devices under test in the temperature range, the external device is given priority to send tasks to the device under test until its score drops below the preset threshold.

[0105] In this embodiment, a high score is divided into two cases, and the following measures are taken for each case:

[0106] First, if the resource utilization rate of the device under test is too high, it is necessary to limit the power consumption and migrate the load of the device under test. This is to prevent its power consumption from continuing to increase, and to set the priority of the device under test to the lowest level. The tasks should be sent to other devices under test for processing, so as to avoid sending more tasks to the device under test and causing its resource utilization rate to continue to increase. After completing the existing tasks, the resource utilization rate should be gradually reduced, thereby reducing the score until it is lower than the preset threshold, so that the resource utilization rate of the devices under test in the temperature range is basically the same.

[0107] Second, if the resource utilization rate of the device under test is too low, and if the device under test has a power consumption limit, it needs to be removed and the priority of the device under test should be increased. When a new task is issued, the device under test with a higher priority should be selected to execute the task, thereby improving the resource utilization rate and reducing the score until it is lower than the preset threshold, so that the resource utilization rate of the devices under test in the temperature range is basically the same.

[0108] In a preferred embodiment, resource utilization includes CPU utilization, memory utilization, and network utilization;

[0109] The score for each device under test is calculated based on its resource utilization rate. The absolute value of the difference between the resource utilization rate of the device under test and the average resource utilization rate of all devices under test within the same temperature range is positively correlated with the score, including:

[0110] According to the formula Calculate the score for each device under test, where m(T) is the average ambient temperature of all devices under test in the computer room, G is the temperature zone to which the device under test belongs, and m(T) GLet m(D) be the average ambient temperature of all devices under test within temperature region G, and let D be the CPU utilization of the device under test. G ) represents the average CPU utilization of all devices under test within temperature range G, and E represents the memory utilization of the devices under test. G ) represents the average memory utilization of all devices under test within temperature region G, and F represents the network utilization of the devices under test. G ) represents the average network utilization of all devices under test within temperature region G.

[0111] In this embodiment, after dividing the temperature zone, data statistics are performed separately for each temperature zone, namely, the average ambient temperature, average CPU utilization, average memory utilization, and average network utilization of all devices under test within the temperature zone, as well as the average ambient temperature of the computer room.

[0112] The score obtained by formula S reflects the degree of deviation between the operating status of the device under test (DUT) and the average operating status of the temperature range. A higher score indicates a greater deviation, requiring optimization of the load on the DUT, i.e., optimizing the number of tasks sent to and from the DUT. When the score approaches 0, it indicates that the DUT meets the operating benchmark of the current temperature range and no change in operating load is needed.

[0113] It should also be noted that the scoring primarily uses the absolute value of the difference between the resource utilization rate of the device under test (DUT) and the average resource utilization rate of all DUTs within its temperature range, supplemented by the average ambient temperature of all DUTs within the temperature range. When the resource utilization rates of multiple DUTs differ from the average resource utilization rate of all DUTs within their respective temperature ranges, priority is given to processing the DUT within the temperature range with the higher average ambient temperature. This allows for timely adjustment of its resource utilization rate to prevent further temperature increases, which could lead to greater heat generation and increased heat dissipation and power consumption.

[0114] Under the above scoring model, since the indicators used in the scoring of the device under test are all related to the real-time operating data of the device under test, the obtained evaluation is dynamically changing. Under this evaluation formula, the task distribution will be more evenly distributed to the device under test, improving the overall utilization efficiency of the data center. At the same time, since the benchmark for obtaining the score is the average temperature of the temperature range, the impact of immutable factors such as the location of air conditioners and columns in the data center's own design can be reduced.

[0115] Furthermore, when the device under test (DUT) is a server, the basic location unit is the server rack housing multiple servers. A commonly used physical facility monitoring platform, such as ISPIM, is used to manage and monitor the DUT within the data center. Real-time operational data of the DUT is acquired at a certain monitoring frequency. This operational data includes ambient temperature (T), CPU utilization (D), memory utilization (E), and network utilization (F). After acquiring the operational data, an operational data model can be built based on the data center map index (L). in This refers to the operational data of the cabinet located in the Mth column of the Nth row. This is the operating data for the x-th server in the rack located at position M in row N. The temperature sensor readings reflect the ambient temperature of the server, used to divide the data center into temperature zones; CPU utilization, memory utilization, and network utilization reflect the server's operating status and are used to score the server as described above.

[0116] The raw data in the running data model d can be processed. Since the coordinates in L are divided with server racks as the basic unit, the average value of the ambient temperature around the servers needs to be extracted as the ambient temperature around the server racks. Then, temperature zones are divided based on the ambient temperature around the server racks. After the division, a score is calculated according to the above formula, and the overall power consumption of the data center is reduced based on the score.

[0117] Please refer to Figure 2 , Figure 2 A schematic diagram of a power consumption regulation system provided in this application includes:

[0118] Data acquisition unit 31 is used to acquire the ambient temperature and resource utilization rate of each device under test in the computer room during the current period;

[0119] Temperature zone division unit 32 is used to divide temperature zones according to the ambient temperature of each device under test.

[0120] The scoring calculation unit 33 is used to calculate the score of each device under test based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of the device under test and the average resource utilization rate of all devices under test in the temperature range is positively correlated with the score.

[0121] The adjustment unit 34 is used to adjust the resource utilization rate of the device under test according to the order of the scores from large to small, so that the scores are reduced.

[0122] For a description of the power consumption regulation system provided in this application, please refer to the above embodiments; further details will not be repeated here.

[0123] In a preferred embodiment, the data acquisition unit 31 is specifically used for:

[0124] The ambient temperature of each device under test in the computer room during the current period is obtained by temperature sensors installed on each device under test, and the resource utilization rate of each device under test in the computer room during the current period is also obtained.

[0125] In a preferred embodiment, the temperature zone division unit 32 includes:

[0126] Location set partitioning unit, used to divide the computer room into N rows * M columns of location sets. Indicates the position of the Nth row and the Mth column;

[0127] The set creation and location determination unit is used to create a set Gi corresponding to the i-th temperature region, and to determine the location of the device under test with the highest ambient temperature within the location set L.

[0128] The temperature level determination unit is used to determine the temperature level of the highest ambient temperature. The temperature level corresponds one-to-one with the temperature range.

[0129] The set and pre-set are used to divide the units into positions. The devices under test are grouped into set Gi and their spatially associated locations are determined. The devices under test are assigned to the preliminary set RG.

[0130] The first judgment unit is used to determine whether there is a device under test with a temperature level in the preparatory set RG. If yes, it enters the set and preparatory set re-division unit; if no, it enters the second judgment unit.

[0131] The set and pre-set re-division unit is used to re-divide the devices under test with temperature levels into set Gi, and to divide the devices under test at the spatially associated positions of the devices under test re-divided into set Gi into pre-set RG, and then enter the first judgment unit.

[0132] The second judgment unit is used to determine whether there is a device under test in the preparatory set RG. If yes, it enters the update unit; otherwise, it stops.

[0133] The update unit is used to update the position set L to the preliminary set RG when i = i + 1, and then enters the set creation and position determination unit.

[0134] In a preferred embodiment, the temperature level determination unit includes:

[0135] The historical temperature determination unit is used to determine the maximum, minimum, and average historical ambient temperatures of all devices under test in the computer room.

[0136] The "Excellent" determination unit is used to determine the temperature level as "Excellent" when the highest ambient temperature is less than (a+b) / 2, where a is the average historical ambient temperature of all devices under test, and b is the minimum historical ambient temperature of all devices under test.

[0137] The "Good" determination unit is used to determine the temperature level as "Good" when the highest ambient temperature is greater than (a+b) / 2 and less than a.

[0138] The "Medium" determination unit is used to determine the temperature level as "Medium" when the highest ambient temperature is greater than a and less than (a+c) / 2, where c is the maximum historical ambient temperature of all devices under test.

[0139] The "Difference" determination unit is used to determine the temperature level as "Difference" when the highest ambient temperature is greater than (a+c) / 2.

[0140] As a preferred embodiment, it also includes:

[0141] The removal unit is used to remove positions in the location set L where the device under test cannot be placed after the location set is divided into units.

[0142] As a preferred embodiment, it also includes:

[0143] The period adjustment unit is used to adjust the size of the current period according to the frequency of tasks sent by external devices before the data acquisition unit 31. The task needs to be executed by the device under test, and the frequency of task sending is negatively correlated with the current period.

[0144] In a preferred embodiment, the adjustment unit 34 is specifically used to perform the following steps on the device under test in descending order of scores:

[0145] When the resource utilization rate of the device under test is greater than the average resource utilization rate of all devices under test in the temperature range, limit the power consumption of the device under test and prohibit external devices from sending tasks to the device under test until its score drops below the preset threshold.

[0146] When the resource utilization rate of the device under test is less than the average resource utilization rate of all devices under test in the temperature range, the external device is given priority to send tasks to the device under test until its score drops below the preset threshold.

[0147] In a preferred embodiment, resource utilization includes CPU utilization, memory utilization, and network utilization;

[0148] The scoring calculation unit 33 is specifically used to calculate the score based on the formula. Calculate the score for each device under test, where m(T) is the average ambient temperature of all devices under test in the computer room, G is the temperature zone to which the device under test belongs, and m(T) G Let m(D) be the average ambient temperature of all devices under test within temperature region G, and let D be the CPU utilization of the device under test. G ) represents the average CPU utilization of all devices under test within temperature range G, and E represents the memory utilization of the devices under test. G ) represents the average memory utilization of all devices under test within temperature region G, and F represents the network utilization of the devices under test. G ) represents the average network utilization of all devices under test within temperature region G.

[0149] Please refer to Figure 3 , Figure 3 A schematic diagram of a power consumption regulation device provided in this application includes:

[0150] Memory 41 is used to store computer programs;

[0151] Processor 42 is used to execute computer programs to implement the steps of the power consumption regulation method.

[0152] For a description of the power consumption regulation device provided in this application, please refer to the above embodiments; further details will not be repeated here.

[0153] This power consumption regulation device can be used for ISPIM, which provides users with functions such as asset management, intelligent monitoring, energy consumption management, automatic inspection and maintenance, stateless management, and report statistical analysis. It enables unified management of servers, storage, network devices, and firewalls, effectively helping enterprise users improve operation and maintenance efficiency, reduce operation and maintenance costs, and ensure the stable operation of data centers.

[0154] Based on ISPIM and the above embodiments, through three implementation processes: first, data center map creation; second, data center temperature zone division; and third, load migration based on temperature zone scoring, the cost of data center use and the effective utilization rate can be reduced.

[0155] It should be noted that, in this specification, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0156] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A power consumption regulation method, characterized in that, include: Obtain the ambient temperature and resource utilization rate of each device under test in the computer room during the current period; Temperature zones are divided based on the ambient temperature of each of the devices under test. The score of each device under test is calculated based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of each device under test and the average resource utilization rate of all devices under test in the temperature range is positively correlated with the score. The resource utilization rate of the device under test is adjusted to reduce its score according to the descending order of the scores. The temperature zone division based on the ambient temperature of each of the devices under test includes: S21: Divide the computer room into a set of N rows * M columns. , Indicates the position of the Nth row and the Mth column; S22: Create a new set Gi corresponding to the i-th temperature region, and identify the location of the device under test with the highest ambient temperature within the location set L. , i≥1, 1≤n≤N, 1≤m≤M; S23: Determine the highest ambient temperature level, which corresponds one-to-one with the temperature range; S24: Assign the device under test at the specified location to the set Gi, and spatially associate the locations. The devices under test are assigned to the preparatory set RG. ={ , , , , , , , }; S25: Determine whether there is a device under test with the temperature level in the pre-set set RG. If yes, proceed to S26; otherwise, proceed to S27. S26: The device under test with the temperature level is reclassified into the set Gi, and the device under test at the spatially associated position of the device under test reclassified into the set Gi is reclassified into the preliminary set RG, and then proceeds to S25. S27: Determine whether the device under test exists in the preparatory set RG. If yes, proceed to S28; otherwise, stop. S28: i = i + 1 and update the position set L to the preparatory set RG, then proceed to S22.

2. The power consumption adjustment method as described in claim 1, characterized in that, Obtain the ambient temperature and resource utilization of each device under test in the computer room during the current period, including: The ambient temperature of each device under test in the computer room during the current period is obtained by temperature sensors installed on each device under test, and the resource utilization rate of each device under test in the computer room during the current period is also obtained.

3. The power consumption adjustment method as described in claim 1, characterized in that, Determine the highest ambient temperature level, where each temperature level corresponds one-to-one with a temperature range, including: Determine the maximum, minimum, and average historical ambient temperatures of all the devices under test within the computer room; When the highest ambient temperature is less than (a+b) / 2, the temperature level is determined to be "excellent", where a is the average of the historical ambient temperatures of all the devices under test, and b is the minimum of the historical ambient temperatures of all the devices under test. When the highest ambient temperature is greater than (a+b) / 2 and less than a, the temperature level is determined to be "good". When the highest ambient temperature is greater than a and less than (a+c) / 2, the temperature level is determined to be "medium", where c is the maximum historical ambient temperature of all the devices under test; When the highest ambient temperature is greater than (a+c) / 2, the temperature level is determined to be "poor".

4. The power consumption adjustment method as described in claim 1, characterized in that, The computer room is divided into a set of N rows * M columns. Following that, it also includes: Remove locations from the location set L where the device under test cannot be placed.

5. The power consumption adjustment method as described in claim 1, characterized in that, Before obtaining the ambient temperature and resource utilization of each device under test in the computer room during the current period, the following steps are also included: The current period is adjusted according to the frequency of tasks issued by external devices. The tasks need to be executed by the device under test, and the frequency of the tasks is negatively correlated with the current period.

6. The power consumption adjustment method as described in claim 1, characterized in that, Adjusting the resource utilization rate of the device under test to lower its score according to the descending order of the scores includes: Based on the scores in descending order, the following steps are performed on the device under test: When the resource utilization rate of the device under test is greater than the average resource utilization rate of all devices under test in the temperature range, the power consumption of the device under test is limited and external devices are prohibited from sending tasks to the device under test until its score drops to less than a preset threshold. When the resource utilization rate of the device under test is less than the average resource utilization rate of all devices under test in the temperature range, the external device is given priority to send a task to the device under test until its score drops below the preset threshold.

7. The power consumption regulation method according to any one of claims 1 to 6, characterized in that, The resource utilization rate includes CPU utilization, memory utilization, and network utilization. A score is calculated for each device under test based on its resource utilization rate. The absolute value of the difference between the resource utilization rate of each device under test and the average resource utilization rate of all devices under test within the same temperature range is positively correlated with the score, including: According to the formula Calculate the score for each of the devices under test, where m(T) is the average ambient temperature of all the devices under test in the computer room, G is the temperature zone to which the device under test belongs, and m( ) is the average ambient temperature of all the devices under test within the temperature region G, D is the CPU utilization of the device under test, and m ( ) represents the average CPU utilization of all the devices under test within the temperature range G, E represents the memory utilization of the devices under test, and m( ) is the average memory utilization of all the devices under test within the temperature range G, F is the network utilization of the devices under test, and m ( ) represents the average network utilization of all the devices under test within the temperature range G.

8. A power consumption regulation system, characterized in that, include: The data acquisition unit is used to acquire the ambient temperature and resource utilization rate of each device under test in the computer room during the current period. A temperature zone division unit is used to divide the temperature zone according to the ambient temperature of each of the devices under test. The scoring calculation unit is used to calculate the score of each of the devices under test based on the resource utilization rate. The absolute value of the difference between the resource utilization rate of the device under test and the average resource utilization rate of all the devices under test in the temperature range is positively correlated with the score. An adjustment unit is used to adjust the resource utilization rate of the device under test according to the order of the scores from largest to smallest, so that the scores are reduced. The temperature zone division unit is specifically used to perform the following steps: S21: Divide the computer room into a set of N rows * M columns. , Indicates the position of the Nth row and the Mth column; S22: Create a new set Gi corresponding to the i-th temperature region, and identify the location of the device under test with the highest ambient temperature within the location set L. , i≥1, 1≤n≤N, 1≤m≤M; S23: Determine the highest ambient temperature level, which corresponds one-to-one with the temperature range; S24: Position The devices under test are assigned to the set Gi, and the spatially associated locations are determined. The devices under test are assigned to the preparatory set RG. ={ , , , , , , , }; S25: Determine whether there is a device under test with the temperature level in the pre-set set RG. If yes, proceed to S26; otherwise, proceed to S27. S26: The device under test with the temperature level is reclassified into the set Gi, and the device under test at the spatially associated position of the device under test reclassified into the set Gi is reclassified into the preliminary set RG, and then proceeds to S25. S27: Determine whether the device under test exists in the preparatory set RG. If yes, proceed to S28; otherwise, stop. S28: i = i + 1 and update the position set L to the preparatory set RG, then proceed to S22.

9. A power consumption regulation device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the power consumption regulation method as described in any one of claims 1 to 7.

Images