Server operation management method, system and device, server and storage medium

Abstract

The application discloses a kind of server operation management method, system, device, server and storage medium, it is related to the field of server, which comprises: determining a plurality of preset power consumption values;Each preset power consumption value is set as the power consumption ceiling value of server, and the stress test is carried out on the server under each preset power consumption value, to obtain the maximum power consumption value and performance value of server;Based on all maximum power consumption values and all performance values, the best energy efficiency ratio is calculated;The optimal power consumption value obtained according to the best energy efficiency ratio is set as the power consumption ceiling value of server.The application can ensure that the energy efficiency conversion ratio of server is best when high load service, avoid causing resource waste, so as to promote green data center construction.

Description

Technical Field

[0001] This application relates to the field of servers, and in particular to server operation and management methods, systems, devices, servers, and storage media. Background Technology

[0002] With the promotion of green data center construction, the requirements for energy efficiency conversion and energy conservation and emission reduction of equipment in data centers are becoming increasingly stringent. To meet server energy consumption control needs, improve server performance, and enhance data center energy utilization, server management software offers a power consumption capping function to limit system power consumption, ensuring that server operating power consumption is below the set safe power consumption value. This technology can avoid sudden power consumption spikes and improve server operational security. Currently, the maximum theoretical power consumption of the server is set as the power consumption cap. However, in actual use, the probability of server operating power consumption approaching the rated maximum theoretical power consumption is extremely low, and the idle rate of server resources reaches over 50%, meaning that the energy efficiency conversion ratio corresponding to the maximum theoretical power consumption is not high, resulting in wasted server resources.

[0003] Therefore, how to provide a solution to the above-mentioned technical problems is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a server operation management method, system, device, server, and storage medium that can ensure the optimal energy efficiency conversion ratio of the server under high load business, avoid resource waste, and thus promote the construction of green data centers.

[0005] To address the aforementioned technical problems, this application provides a server operation management method, comprising:

[0006] Determine multiple preset power consumption values;

[0007] Each of the preset power consumption values ​​is set as the power consumption cap of the server, and a stress test is performed on the server under each preset power consumption value to obtain the maximum power consumption value and performance value of the server.

[0008] Calculate the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values;

[0009] The optimal power consumption value obtained based on the optimal energy efficiency ratio is set as the power consumption cap of the server.

[0010] Optionally, the process of determining multiple preset power consumption values ​​includes:

[0011] Determine the maximum theoretical power consumption and reference power consumption of the server;

[0012] The target power consumption interval [p, q] is determined based on the reference power consumption value and the maximum theoretical power consumption value. The target power consumption interval is divided into n equal parts to obtain n sub-intervals. The boundary values ​​of the n sub-intervals are used as the preset power consumption value. p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

[0013] Optionally, the server includes a BMC and a CPU, and the process of setting each of the preset power consumption values ​​as the server's power consumption cap includes:

[0014] The CPU's operating frequency is adjusted by the BMC so that the preset power consumption value is set as the server's power consumption cap.

[0015] Optionally, the process of calculating the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values ​​includes:

[0016] Curve fitting is performed on all the maximum power consumption values ​​and all the performance values ​​to obtain a curve function;

[0017] Calculate the maximum value of the derivative of the curve function, and determine the maximum value as the optimal energy efficiency ratio.

[0018] Optionally, the process of performing curve fitting on all the maximum power consumption values ​​and all the performance values ​​to obtain a curve function includes:

[0019] Curve functions are obtained by curve fitting of all the maximum power consumption values ​​and all the performance values ​​using the least squares method.

[0020] Optionally, the process of stress testing the server at each of the preset power consumption values ​​includes:

[0021] The server was subjected to stress testing using a stress testing tool at each of the preset power consumption values.

[0022] The server operation and management method also includes:

[0023] Obtain the square root count of the stress testing tool;

[0024] The number of square roots is used as the performance value of the server.

[0025] Optionally, after calculating the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values, and before setting the optimal power consumption value obtained based on the optimal energy efficiency ratio as the power consumption cap of the server, the server operation management method further includes:

[0026] Determine the power consumption value to be measured;

[0027] The power consumption value to be measured is set as the power consumption cap of the server, and the stress test is performed on the server under each power consumption value to be measured to obtain the maximum power consumption value and performance value of the server.

[0028] A new optimal energy efficiency ratio is calculated based on the server's maximum power consumption and performance values ​​under all the preset power consumption values ​​and the server's maximum power consumption and performance values ​​under all the power consumption values ​​to be tested;

[0029] Determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the set conditions;

[0030] If so, the power consumption value corresponding to the new optimal energy efficiency ratio shall be taken as the optimal power consumption value;

[0031] If not, repeat the operation of determining the power consumption value to be measured.

[0032] Optionally, the process of determining the power consumption value to be measured includes:

[0033] Determine the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio. If the current iteration number is 0, the candidate power consumption value is the preset power consumption value. If the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number.

[0034] Calculate the average value of the power consumption value corresponding to the candidate power consumption value and the optimal energy efficiency ratio;

[0035] The average value is used as the power consumption value to be measured corresponding to the current iteration number.

[0036] Optionally, the setting conditions include the difference between the power consumption value corresponding to the optimal energy efficiency ratio in the current iteration number and the power consumption value corresponding to the optimal energy efficiency ratio in the previous iteration number being less than a preset value.

[0037] Optionally, the setting condition is that the current iteration number is a preset number.

[0038] To address the aforementioned technical problems, this application also provides a server operation and management system, comprising:

[0039] Determine the module and determine multiple preset power consumption values;

[0040] The testing module is used to set each of the preset power consumption values ​​as the power consumption cap of the server, and to perform stress tests on the server under each preset power consumption value to obtain the maximum power consumption value and performance value of the server.

[0041] The calculation module is used to calculate the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values;

[0042] The setting module is used to set the optimal power consumption value obtained based on the optimal energy efficiency ratio as the power consumption cap value of the server.

[0043] Optionally, the determining module is further configured to determine the power consumption value to be measured;

[0044] The testing module is also used to set the power consumption value to be tested as the power consumption cap value of the server, and to perform the stress test on the server under each power consumption value to be tested, so as to obtain the maximum power consumption value and performance value of the server.

[0045] The calculation module is also used to calculate a new optimal energy efficiency ratio based on the maximum power consumption and performance value of the server under all the preset power consumption values ​​and the maximum power consumption and performance value of the server under all the power consumption values ​​to be tested;

[0046] The server operation and management system also includes:

[0047] The judgment module is used to determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions. If yes, the power consumption value corresponding to the new optimal energy efficiency ratio is taken as the optimal power consumption value. If not, the test module is triggered so that the test module can re-determine the power consumption value to be tested.

[0048] To address the aforementioned technical problems, this application also provides a server operation management device, comprising:

[0049] Memory, used to store computer programs;

[0050] A processor, used to implement the steps of the server operation management method as described in any of the above when executing the computer program.

[0051] To address the aforementioned technical problems, this application also provides a server, including the server operation management device as described above.

[0052] To address the aforementioned technical problems, this application also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the server operation management method described in any of the above descriptions.

[0053] This application provides a server operation management method. First, multiple preset power consumption values ​​are determined, and each preset power consumption value is used as a power consumption cap for the server. Under each preset power consumption value as the server's power consumption cap, a stress test is performed on the server to obtain the maximum power consumption and performance values ​​of the server at each preset power consumption value. This allows for the calculation of the server's optimal energy efficiency ratio (EER). The optimal power consumption value corresponding to the optimal EER is then set as the server's power consumption cap, ensuring the server achieves optimal energy conversion efficiency under high-load business conditions, avoiding resource waste, and thus promoting the construction of green data centers. This application also provides a server operation management system, device, server, and storage medium, which have the same beneficial effects as the aforementioned server operation management method. Attached Figure Description

[0054] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the 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.

[0055] Figure 1 A flowchart of the steps of a server operation management method provided in this application;

[0056] Figure 2 A schematic diagram of a fitting curve provided in this application;

[0057] Figure 3 A flowchart illustrating the steps of another server operation management method provided in this application;

[0058] Figure 4 Another schematic diagram of the fitting curve provided in this application;

[0059] Figure 5 This is a schematic diagram of the structure of a server operation and management system provided in this application. Detailed Implementation

[0060] The core of this application is to provide a server operation management method, system, device, server, and storage medium that can ensure the optimal energy efficiency conversion ratio of the server under high load business, avoid resource waste, and thus promote the construction of green data centers.

[0061] 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.

[0062] Please refer to Figure 1 , Figure 1 The present application provides a flowchart of a server operation management method, which includes:

[0063] S101: Determine multiple preset power consumption values;

[0064] Each preset power consumption value is different and does not exceed the server's theoretical maximum power consumption. The number of preset power consumption values ​​can be determined according to actual needs. Understandably, the more preset power consumption values ​​there are, the more accurate the optimal energy efficiency ratio will be in subsequent calculations.

[0065] S102: Set each preset power consumption value as the server's power consumption cap value, and perform a stress test on the server under each preset power consumption value to obtain the server's maximum power consumption value and performance value;

[0066] It is understandable that the power consumption cap is used to limit the overall power consumption of the server. In S101, multiple preset power consumption values ​​are determined, and each preset power consumption value is set as the power consumption cap of the server. This application does not make specific restrictions on the setting order of each preset power consumption value. Assuming four preset power consumption values ​​are determined, namely a, b, c, and d, we can first set 'a' as the maximum power consumption value and perform a stress test on the server to obtain the maximum power consumption value Wa and performance value Ka when 'a' is the maximum power consumption value. Next, we set 'b' as the maximum power consumption value and perform a stress test on the server to obtain the maximum power consumption value Wb and performance value Kb when 'b' is the maximum power consumption value. Then, we set 'c' as the maximum power consumption value and perform a stress test on the server to obtain the maximum power consumption value Wc and performance value Kc when 'c' is the maximum power consumption value. Finally, we set 'd' as the maximum power consumption value and perform a stress test on the server to obtain the maximum power consumption value Wd and performance value Kd when 'd' is the maximum power consumption value. Here, the performance value refers to the server's total performance value in this stress test. The stress test duration can be determined according to actual needs, such as setting it to 10 minutes.

[0067] As an optional embodiment, the server includes a BMC (Baseboard Management Controller) and a CPU (Central Processing Unit). During each stress test, the maximum power consumption value of the server can be obtained by polling the BMC. Specifically, at the beginning of the current stress test, the BMC obtains the power consumption value of the server once in each polling cycle. At the end of the current stress test, the BMC determines a maximum power consumption value from the power consumption values ​​obtained in all polling cycles, which is used as the maximum power consumption value of the server corresponding to the current stress test.

[0068] As an optional embodiment, the process of setting each preset power consumption value as the server's power consumption cap includes:

[0069] The CPU's operating frequency is adjusted via the BMC so that the preset power consumption value is set as the server's power consumption cap.

[0070] Specifically, considering that the CPU's operating frequency has a significant impact on the server's overall power consumption, a higher CPU operating frequency results in higher overall power consumption, while a lower CPU operating frequency results in lower overall power consumption, the CPU's operating frequency can be adjusted through the BMC to set a power consumption cap for the server.

[0071] As an optional embodiment, the process of stress testing the server at each preset power consumption value includes:

[0072] The server was stress-tested using a stress testing tool at each preset power consumption value;

[0073] The server operation and management method also includes:

[0074] Obtain the square root count of the load testing tool;

[0075] The number of square roots is used as a performance value for the server.

[0076] Specifically, a full-core stress test can be performed on the server using a stress testing tool. When performing a full-core stress test on the server using a stress testing tool, the stress testing tool will generate n worker processes to perform square root calculations to generate load. Based on this, at the end of each full-core stress test, the number of square roots taken by the stress testing tool is obtained, and this number of square roots is used as the performance value of the server.

[0077] As another optional implementation, the floating-point operation value of the CPU during this full-core stress test can also be counted, and the performance value of the server can be determined based on the floating-point operation value of the CPU.

[0078] Of course, in addition to the two methods mentioned above, other methods can also be used to obtain server performance values, and this application does not make any specific limitations here.

[0079] S103: Calculates the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values;

[0080] It is understandable that the energy efficiency ratio is the ratio of the performance value to the maximum power consumption value, that is, the energy efficiency ratio Fi = performance value Ki / maximum power consumption value Wi. Taking four preset power consumption values ​​a, b, c, and d as an example, we can obtain the energy efficiency ratios Fa, Fb, Fc, and Fd of the server when each preset power consumption value is the power consumption cap. Among them, Fa = Ka / Wa, Fb = Kb / Wb, Fc = Kc / Wc, and Fd = Kd / Wd. The optimal energy efficiency ratio is the maximum value among all energy efficiency ratios.

[0081] Furthermore, in order to further improve the accuracy and reliability of the obtained optimal energy efficiency ratio, as an optional embodiment, the process of calculating the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values ​​includes: performing curve fitting on all maximum power consumption values ​​and all performance values ​​to obtain a curve function; calculating the maximum value of the derivative of the curve function, and determining the maximum value as the optimal energy efficiency ratio.

[0082] Specifically, using maximum power consumption as the x-axis and performance values ​​as the y-axis, curve fitting is performed on all maximum power consumption values ​​and all performance values ​​to obtain a curve function. Then, the maximum value of the energy efficiency ratio on the curve is obtained. Specifically, the derivative of the curve function can be taken, and the maximum value of the derivative is the optimal energy efficiency ratio. The least squares method can be used for curve fitting, as it can easily obtain unknown data and minimize the sum of squares of the errors between the obtained data and the actual data. Of course, other methods can also be used for curve fitting, and this application does not impose specific limitations on them.

[0083] S104: Set the optimal power consumption value obtained based on the best energy efficiency ratio as the power consumption cap for the server.

[0084] Understandably, the optimal energy efficiency ratio corresponds to a maximum power consumption value and a performance value. The maximum power consumption value corresponding to the optimal energy efficiency ratio is taken as the optimal power consumption value, and this optimal power consumption value is set as the power consumption cap of the server. This ensures that the server operates at this power consumption cap, guaranteeing the optimal energy efficiency conversion ratio of the server under high load business and avoiding resource waste.

[0085] As can be seen, in this embodiment, multiple preset power consumption values ​​are first determined and used as the power consumption cap of the server. When each preset power consumption value is used as the power consumption cap of the server, a stress test is performed on the server to obtain the maximum power consumption and performance value of the server under each preset power consumption value. The optimal energy efficiency ratio of the server is then calculated, and the optimal power consumption value corresponding to the optimal energy efficiency ratio is set as the power consumption cap of the server. This ensures that the server has the best energy efficiency conversion ratio under high load business, avoids resource waste, and promotes the construction of green data centers.

[0086] Based on the above embodiments:

[0087] As an optional embodiment, the process of determining multiple preset power consumption values ​​includes:

[0088] Determine the server's maximum theoretical power consumption and reference power consumption;

[0089] The target power consumption range [p, q] is determined based on the reference power consumption value and the maximum theoretical power consumption value. The target power consumption range is divided into n equal parts to obtain n sub-ranges. The boundary values ​​of the n sub-ranges are used as preset power consumption values. p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

[0090] Specifically, first, the maximum theoretical power consumption and reference power consumption of the server are determined. The reference power consumption can be determined by referring to the minimum operating power consumption of the server. Then, the reference power consumption and the maximum theoretical power consumption are used as two boundary values ​​of the target power consumption range. Assuming that the reference power consumption p is 200W and the maximum theoretical power consumption q of the server is 800W, the target power consumption range is [200, 800]. The target power consumption range is divided into n equal parts to obtain n sub-ranges. In this embodiment, the boundary values ​​of the n sub-ranges are used as preset power consumption values. Therefore, the larger n is, the more preset power consumption values ​​are determined. At the same time, taking the equally divided boundary values ​​can ensure that there is a corresponding preset power consumption value in each power consumption range within the target power consumption range, which is more comprehensive. Thus, the accuracy and reliability of obtaining the optimal energy efficiency ratio are higher.

[0091] For example, in this embodiment, n=6, that is, the target power consumption range [200, 800] is divided into 6 equal parts, which can be obtained into 6 sub-ranges, namely [200, 300], [300, 400], [400, 500], [500, 600], [600, 700], and [700, 800]. Then, the boundary values ​​of each of the above sub-ranges are used as preset power consumption values, that is, 200, 300, 400, 500, 600, 700, and 800 are used as preset power consumption values.

[0092] Of course, after determining the target power consumption range, the preset power consumption value can be obtained without dividing it into equal parts, or it can be randomly determined within the target power consumption range.

[0093] Furthermore, the pre-defined power consumption values ​​obtained from the above equal division are set as the server's power consumption cap. A stress testing tool is then used to perform stress tests on the server, obtaining the maximum power consumption and performance values ​​during each stress test. Curve fitting is then performed based on all maximum power consumption and performance values. The fitted curve can be referenced... Figure 2 As shown.

[0094] Please refer to Figure 3 , Figure 3 This is a flowchart of another server operation management method provided in this application. In this embodiment, the determination of the optimal power consumption value is divided into two parts: preliminary determination and iterative optimization. The preliminary determination part includes:

[0095] S201: Determine multiple preset power consumption values;

[0096] S202: Set each preset power consumption value as the server's power consumption cap value, and perform a stress test on the server under each preset power consumption value to obtain the server's maximum power consumption value and performance value;

[0097] S203: Calculate the optimal energy efficiency ratio by curve fitting based on all maximum power consumption values ​​and all performance values;

[0098] Understandably, the initial determination of the power consumption value corresponding to the optimal energy efficiency ratio can be used to directly set the power consumption cap of the server as the optimal power consumption value. Alternatively, subsequent iterative optimizations can be carried out to determine a more accurate optimal power consumption value.

[0099] The iterative optimization part includes:

[0100] S204: Determine the power consumption value to be measured;

[0101] S205: Set the power consumption value to be measured as the server's power consumption cap value, and perform stress tests on the server under each power consumption value to obtain the server's maximum power consumption value and performance value;

[0102] S206: Based on the maximum power consumption and performance values ​​of the server under all preset power consumption values ​​and the maximum power consumption and performance values ​​of the server under all test power consumption values, curve fitting is performed to calculate the new optimal energy efficiency ratio;

[0103] S207: Determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions. If yes, execute S208; otherwise, execute S204.

[0104] S208: Take the power consumption value corresponding to the new optimal energy efficiency ratio as the optimal power consumption value and set it as the power consumption cap value of the server.

[0105] Specifically, in addition to the multiple preset power consumption values ​​determined in S101, a power consumption value to be measured is determined. The number of power consumption values ​​to be measured can be one or more. The determined power consumption values ​​to be measured are set as the power consumption cap values ​​of the server. The server is subjected to stress testing to obtain the server's maximum power consumption value and performance value, so as to increase the amount of data for subsequent curve fitting and further improve the accuracy and reliability of the obtained optimal energy efficiency ratio.

[0106] Understandably, when determining the power consumption value to be measured, the current iteration number can be considered. If the current iteration number is 0, the power consumption value to be measured under the current iteration number needs to be determined based on the power consumption value corresponding to the optimal energy efficiency ratio obtained in the preliminary determination part. If the current iteration number is not 0, the power consumption value to be measured under the current iteration number needs to be determined based on the power consumption value corresponding to the optimal energy efficiency ratio obtained in the previous iteration number.

[0107] As an optional embodiment, the process of determining the power consumption value to be measured includes: determining the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio; if the current iteration number is 0, the candidate power consumption value is the preset power consumption value; if the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number; calculating the average value of the candidate power consumption value and the power consumption value corresponding to the optimal energy efficiency ratio; and using the average value as the power consumption value to be measured corresponding to the current iteration number.

[0108] Specifically, with Figure 4 For example, assuming point A is the point where the optimal energy efficiency ratio is initially determined from the partial calculation, then a power consumption value W0 can be determined based on point A. Assuming W0 is 460W, in the first iteration, the two preset power consumption values ​​adjacent to the power consumption value W0 can be determined as candidate power consumption values, that is, 400W and 500W are determined as candidate power consumption values. Then, the average value between the candidate power consumption value and the power consumption value W0 corresponding to point A is calculated, that is, the average value of 400W and 460W, and the average value of 500W and 460W are calculated. The two average values ​​of 430W and 480W are used as the power consumption value to be measured in the current iteration.

[0109] Next, 460W is set as the server's power consumption cap, and a stress test is performed on the server to obtain the maximum power consumption and performance values ​​during the test. Then, 480W is set as the server's power consumption cap, and a stress test is performed on the server to obtain the maximum power consumption and performance values ​​during the test. The maximum power consumption and performance values ​​corresponding to 460W and 480W are then curve-fitted with the maximum power consumption and performance values ​​corresponding to the 200W-800W range mentioned above, resulting in a new curve. Based on the new curve, subsequent calculations can be performed to obtain a new optimal energy efficiency ratio. It is then determined whether the power consumption value W1 corresponding to the new optimal energy efficiency ratio meets the setting conditions. If it does, then this power consumption value W1 is set as the server's power consumption cap. If the conditions are not met, subsequent iterations are performed to re-obtain the power consumption value to be measured. At this point, one or two candidate power consumption values ​​adjacent to the power consumption value W1 are determined. Assuming W1 is 470W, the two candidate power consumption values ​​adjacent to it, i.e., the power consumption values ​​to be measured determined in the previous iteration, are 430W and 480W. Then, the average value of 430W and 470W, and the average value of 480W and 470W are calculated. 450W and 475W are taken as the two power consumption values ​​to be measured corresponding to the current iteration number, and the above steps are repeated until the power consumption value corresponding to the optimal energy efficiency ratio calculated under the current iteration number meets the setting conditions. The iteration is then stopped, and the optimal power consumption determined after multiple iterations is set as the power consumption cap value of the server, which enables the server to have the best energy efficiency conversion ratio under high load business.

[0110] The setting conditions can include the difference between the power consumption value corresponding to the best energy efficiency ratio in the current iteration and the power consumption value corresponding to the best energy efficiency ratio in the previous iteration being less than a preset value, or the current iteration reaching a preset number.

[0111] Specifically, the absolute value of the difference between the power consumption value corresponding to the best energy efficiency ratio in the current iteration and the power consumption value corresponding to the best energy efficiency ratio in the previous iteration is less than a preset value to determine that the setting condition is met. Assuming the preset value is set to 5W, in the above steps, if |W1-W0|<5W, then W1 is determined to meet the setting condition, and W1 is determined as the optimal power consumption value and set as the power consumption cap of the server.

[0112] In summary, this application first presets several different server power consumption values ​​(i.e., preset power consumption values). Using a performance testing tool, different preset power consumption points are set as power consumption caps for testing, yielding the corresponding maximum power consumption and performance values. By curve fitting of different maximum power consumption and performance values, the optimal energy efficiency conversion ratio (EER) point is determined. The average power consumption point (i.e., the power consumption point to be tested) between this point and its two adjacent points is calculated. Power consumption caps and performance stress tests are then performed on these two power consumption points to obtain the corresponding maximum power consumption and performance values. The data from these two points are then re-curved with the previously tested data to determine the optimal EER point. The power consumption value corresponding to this optimal EER point is compared with the power consumption value corresponding to the previously calculated optimal EER point. If the difference between the two power consumption values ​​is less than 5W, this point is considered the optimal EER point, and the power consumption value corresponding to this optimal EER point is the optimal power consumption value. If the difference is greater than 5W, the previous step is iterated until the difference between the two power consumption values ​​is less than 5W or the maximum number of iterations is reached. Finally, the power consumption value at the optimal energy efficiency conversion ratio point is set as the power consumption cap for the server, thereby ensuring that the server has the best energy efficiency conversion ratio under high load business, thus promoting the construction of green data centers.

[0113] Please refer to Figure 5 , Figure 5 This application provides a schematic diagram of the structure of a server operation management system, which includes:

[0114] Module 1 is defined to determine multiple preset power consumption values;

[0115] Test module 2 is used to set each preset power consumption value as the server's power consumption cap value, and to perform stress tests on the server under each preset power consumption value to obtain the server's maximum power consumption value and performance value;

[0116] Calculation module 3 is used to calculate the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values;

[0117] Setting module 4 is used to set the optimal power consumption value obtained based on the best energy efficiency ratio as the power consumption cap value of the server.

[0118] As can be seen, in this embodiment, multiple preset power consumption values ​​are first determined and used as the power consumption cap of the server. When each preset power consumption value is used as the power consumption cap of the server, a stress test is performed on the server to obtain the maximum power consumption and performance value of the server under each preset power consumption value. The optimal energy efficiency ratio of the server is then calculated, and the optimal power consumption value corresponding to the optimal energy efficiency ratio is set as the power consumption cap of the server. This ensures that the server has the best energy efficiency conversion ratio under high load business, avoids resource waste, and promotes the construction of green data centers.

[0119] As an optional embodiment, the determining module 1 is also used to determine the power consumption value to be measured;

[0120] Test module 2 is also used to set the power consumption value to be tested as the power consumption cap of the server, and to perform stress tests on the server under each power consumption value to obtain the server's maximum power consumption value and performance value;

[0121] Calculation module 3 is also used to calculate a new optimal energy efficiency ratio based on the maximum power consumption and performance of the server under all preset power consumption values ​​and the maximum power consumption and performance of the server under all test power consumption values;

[0122] The server operation and management system also includes:

[0123] The judgment module is used to determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions. If yes, the power consumption value corresponding to the new optimal energy efficiency ratio is taken as the optimal power consumption value. If not, the test module 2 is triggered so that the test module 2 can re-determine the power consumption value to be tested.

[0124] As an optional embodiment, the process of determining multiple preset power consumption values ​​includes:

[0125] Determine the server's maximum theoretical power consumption and reference power consumption;

[0126] The target power consumption range [p, q] is determined based on the reference power consumption value and the maximum theoretical power consumption value. The target power consumption range is divided into n equal parts to obtain n sub-ranges. The boundary values ​​of the n sub-ranges are used as preset power consumption values. p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

[0127] As an optional embodiment, the server includes a BMC and a CPU, and the process of setting each preset power consumption value as the server's power consumption cap includes:

[0128] The CPU's operating frequency is adjusted via the BMC so that the preset power consumption value is set as the server's power consumption cap.

[0129] As an optional implementation, the process of calculating the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values ​​includes:

[0130] Curve fitting is performed on all maximum power consumption values ​​and all performance values ​​to obtain curve functions;

[0131] Calculate the maximum value of the derivative of the curve function, and determine the maximum value as the optimal energy efficiency ratio.

[0132] As an optional embodiment, the process of curve fitting to all maximum power consumption values ​​and all performance values ​​to obtain the curve function includes:

[0133] Curve functions are obtained by curve fitting of all maximum power consumption values ​​and all performance values ​​using the least squares method.

[0134] As an optional embodiment, the process of stress testing the server at each preset power consumption value includes:

[0135] The server was stress-tested using a stress testing tool at each preset power consumption value;

[0136] The server operation and management system also includes:

[0137] The statistics module is used to obtain the number of square roots taken by the load testing tool, and the number of square roots is used as the performance value of the server.

[0138] As an optional embodiment, the process of determining the power consumption value to be measured includes:

[0139] Determine the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio. If the current iteration number is 0, the candidate power consumption value is the preset power consumption value. If the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number.

[0140] Calculate the average of the power consumption values ​​corresponding to the candidate power consumption values ​​and the optimal energy efficiency ratio;

[0141] The average value is used as the power consumption value to be measured for the current iteration number.

[0142] As an optional embodiment, the setting condition includes that the difference between the power consumption value corresponding to the best energy efficiency ratio in the current iteration number and the power consumption value corresponding to the best energy efficiency ratio in the previous iteration number is less than a preset value.

[0143] As an optional implementation, the condition is set to the current iteration number being a preset number.

[0144] On the other hand, this application also provides a server operation management device, including:

[0145] Memory, used to store computer programs;

[0146] A processor is used to execute computer programs to implement the steps of the server operation management method as described in any of the embodiments above.

[0147] Specifically, the memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer-readable instructions, and the internal memory provides an environment for the operation of the operating system and computer-readable instructions stored in the non-volatile storage medium. When the processor executes the computer program stored in the memory, it can perform the following steps: determine multiple preset power consumption values; set each preset power consumption value as the server's power consumption cap, and perform a stress test on the server at each preset power consumption value to obtain the server's maximum power consumption value and performance value; calculate the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values; and set the optimal power consumption value obtained based on the optimal energy efficiency ratio as the server's power consumption cap.

[0148] As can be seen, in this embodiment, multiple preset power consumption values ​​are first determined and used as the power consumption cap of the server. When each preset power consumption value is used as the power consumption cap of the server, a stress test is performed on the server to obtain the maximum power consumption and performance value of the server under each preset power consumption value. The optimal energy efficiency ratio of the server is then calculated, and the optimal power consumption value corresponding to the optimal energy efficiency ratio is set as the power consumption cap of the server. This ensures that the server has the best energy efficiency conversion ratio under high load business, avoids resource waste, and promotes the construction of green data centers.

[0149] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: determine the maximum theoretical power consumption value and the reference power consumption value of the server; determine the target power consumption interval [p, q] based on the reference power consumption value and the maximum theoretical power consumption value; divide the target power consumption interval into n equal parts to obtain n sub-intervals; and use the boundary values ​​of the n sub-intervals as preset power consumption values; where p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

[0150] As an optional embodiment, when the processor executes a computer subroutine stored in memory, it can perform the following steps: adjust the CPU's operating frequency through the BMC so that the preset power consumption value is set as the server's power consumption cap.

[0151] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: curve fitting on all maximum power consumption values ​​and all performance values ​​to obtain a curve function; calculate the maximum value of the derivative of the curve function, and determine the maximum value as the optimal energy efficiency ratio.

[0152] As an optional embodiment, when the processor executes a computer subroutine stored in memory, it can perform the following steps: curve fitting of all maximum power consumption values ​​and all performance values ​​using the least squares method to obtain a curve function.

[0153] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: stress test the server using a stress testing tool at each preset power consumption value; obtain the number of square roots taken by the stress testing tool; and use the number of square roots taken as the performance value of the server.

[0154] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: determine the power consumption value to be measured; set the power consumption value to be measured as the power consumption cap of the server, and perform a stress test on the server under each power consumption value to be measured to obtain the maximum power consumption value and performance value of the server; calculate a new optimal energy efficiency ratio based on the maximum power consumption value and performance value of the server under all preset power consumption values ​​and the maximum power consumption value and performance value of the server under all power consumption values ​​to be measured; determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions; if yes, take the power consumption value corresponding to the new optimal energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0155] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: determine the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio; if the current iteration number is 0, the candidate power consumption value is the preset power consumption value; if the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number; calculate the average value of the candidate power consumption value and the power consumption value corresponding to the optimal energy efficiency ratio; and use the average value as the power consumption value to be measured corresponding to the current iteration number.

[0156] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: determine whether the difference between the power consumption value corresponding to the best energy efficiency ratio in the current iteration number and the power consumption value corresponding to the best energy efficiency ratio in the previous iteration number is less than a preset value; if so, take the power consumption value corresponding to the new best energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0157] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: determine whether the current iteration number is a preset number; if so, take the power consumption value corresponding to the new optimal energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0158] Based on the above embodiments, the server operation management device further includes:

[0159] An input interface, connected to the processor, is used to acquire externally imported computer programs, parameters, and instructions, which are then stored in memory under the processor's control. This input interface can be connected to an input device to receive parameters or instructions manually entered by the user. This input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the terminal casing.

[0160] The display unit, connected to the processor, is used to display the data sent by the processor. This display unit can be an LCD screen or an e-ink screen, etc.

[0161] The network port, connected to the processor, is used for communication with external terminal devices. The communication technology used for this connection can be wired or wireless, such as Mobile High Definition Link (MHL), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), Wireless Fidelity (WiFi), Bluetooth, Bluetooth Low Energy, or IEEE 802.11s-based communication technologies.

[0162] On the other hand, this application also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the server operation management method as described in any of the embodiments above.

[0163] Specifically, this application also provides a computer-readable storage medium, which may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. The storage medium stores a computer program, which, when executed by a processor, performs the following steps: determining multiple preset power consumption values; setting each preset power consumption value as a power consumption cap for the server, and performing a stress test on the server at each preset power consumption value to obtain the server's maximum power consumption value and performance value; calculating the optimal energy efficiency ratio based on all maximum power consumption values ​​and all performance values; and setting the optimal power consumption value obtained based on the optimal energy efficiency ratio as the server's power consumption cap.

[0164] As can be seen, in this embodiment, multiple preset power consumption values ​​are first determined and used as the power consumption cap of the server. When each preset power consumption value is used as the power consumption cap of the server, a stress test is performed on the server to obtain the maximum power consumption and performance value of the server under each preset power consumption value. The optimal energy efficiency ratio of the server is then calculated, and the optimal power consumption value corresponding to the optimal energy efficiency ratio is set as the power consumption cap of the server. This ensures that the server has the best energy efficiency conversion ratio under high load business, avoids resource waste, and promotes the construction of green data centers.

[0165] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically implement the following steps: determining the maximum theoretical power consumption value and the reference power consumption value of the server; determining the target power consumption interval [p, q] based on the reference power consumption value and the maximum theoretical power consumption value; dividing the target power consumption interval into n equal parts to obtain n sub-intervals; and using the boundary values ​​of the n sub-intervals as preset power consumption values; where p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

[0166] As an optional embodiment, when a computer subroutine stored in a computer-readable storage medium is executed by a processor, it can specifically perform the following steps: adjusting the CPU's operating frequency through the BMC so as to set a preset power consumption value as the server's power consumption cap.

[0167] As an optional embodiment, when the processor executes the computer subroutine stored in the memory, it can perform the following steps: curve fitting on all maximum power consumption values ​​and all performance values ​​to obtain a curve function; calculate the maximum value of the derivative of the curve function, and determine the maximum value as the optimal energy efficiency ratio.

[0168] As an optional embodiment, when a computer subroutine stored in a computer-readable storage medium is executed by a processor, it may specifically perform the following steps: curve fitting of all maximum power consumption values ​​and all performance values ​​using the least squares method to obtain a curve function.

[0169] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically perform the following steps: stress test the server using a stress testing tool at each preset power consumption value; obtain the number of square roots obtained by the stress testing tool; and use the number of square roots as the performance value of the server.

[0170] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically perform the following steps: determine the power consumption value to be measured; set the power consumption value to be measured as the power consumption cap of the server, and perform stress tests on the server under each power consumption value to be measured to obtain the maximum power consumption value and performance value of the server; calculate a new optimal energy efficiency ratio based on the maximum power consumption value and performance value of the server under all preset power consumption values ​​and the maximum power consumption value and performance value of the server under all power consumption values ​​to be measured; determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions; if yes, take the power consumption value corresponding to the new optimal energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0171] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically implement the following steps: determining the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio; if the current iteration number is 0, the candidate power consumption value is the preset power consumption value; if the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number; calculating the average value of the candidate power consumption value and the power consumption value corresponding to the optimal energy efficiency ratio; and using the average value as the power consumption value to be measured corresponding to the current iteration number.

[0172] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically implement the following steps: determine whether the difference between the power consumption value corresponding to the best energy efficiency ratio in the current iteration number and the power consumption value corresponding to the best energy efficiency ratio in the previous iteration number is less than a preset value; if so, take the power consumption value corresponding to the new best energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0173] As an optional embodiment, when the computer subroutine stored in the computer-readable storage medium is executed by the processor, it can specifically implement the following steps: determine whether the current iteration number is a preset number; if so, take the power consumption value corresponding to the new optimal energy efficiency ratio as the optimal power consumption value; if not, re-execute the operation of determining the power consumption value to be measured.

[0174] On the other hand, this application also provides a server, including the server operation management device as described in the embodiments above.

[0175] For a description of the server provided in this application, please refer to the above embodiments; further details will not be repeated here.

[0176] The server provided in this application has the same beneficial effects as the server operation and management method described above.

[0177] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or 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 limitations, 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.

[0178] 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 server operation management method, characterized in that, include: Determine multiple preset power consumption values; Each of the preset power consumption values ​​is set as the power consumption cap of the server, and a stress test is performed on the server under each preset power consumption value to obtain the maximum power consumption value and performance value of the server. Calculate the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values; The optimal power consumption value obtained based on the optimal energy efficiency ratio is set as the power consumption cap of the server; After calculating the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values, and before setting the optimal power consumption value obtained based on the optimal energy efficiency ratio as the power consumption cap of the server, the server operation management method further includes: Determine the power consumption value to be measured; The power consumption value to be measured is set as the power consumption cap of the server, and the stress test is performed on the server under each power consumption value to be measured to obtain the maximum power consumption value and performance value of the server. A new optimal energy efficiency ratio is calculated based on the server's maximum power consumption and performance values ​​under all the preset power consumption values ​​and the server's maximum power consumption and performance values ​​under all the power consumption values ​​to be tested; Determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the set conditions; If so, the power consumption value corresponding to the new optimal energy efficiency ratio shall be taken as the optimal power consumption value; If not, repeat the operation of determining the power consumption value to be measured.

2. The server operation management method according to claim 1, characterized in that, The process of determining multiple preset power consumption values ​​includes: Determine the maximum theoretical power consumption and reference power consumption of the server; Determine the target power consumption range based on the reference power consumption value and the maximum theoretical power consumption value. The target power consumption range is divided into n equal parts to obtain n sub-ranges, and the boundary values ​​of the n sub-ranges are used as the preset power consumption values; p is the reference power consumption value, q is the maximum theoretical power consumption value, and n is an integer greater than 1.

3. The server operation management method according to claim 1, characterized in that, The server includes a BMC and a CPU. The process of setting each of the preset power consumption values ​​as the server's power consumption cap value includes: The CPU's operating frequency is adjusted by the BMC so that the preset power consumption value is set as the server's power consumption cap.

4. The server operation management method according to claim 1, characterized in that, The process of calculating the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values ​​includes: Curve fitting is performed on all the maximum power consumption values ​​and all the performance values ​​to obtain a curve function; Calculate the maximum value of the derivative of the curve function, and determine the maximum value as the optimal energy efficiency ratio.

5. The server operation management method according to claim 4, characterized in that, The process of performing curve fitting on all the maximum power consumption values ​​and all the performance values ​​to obtain the curve function includes: Curve functions are obtained by curve fitting of all the maximum power consumption values ​​and all the performance values ​​using the least squares method.

6. The server operation management method according to claim 1, characterized in that, The process of stress testing the server at each of the preset power consumption values ​​includes: The server was subjected to stress testing using a stress testing tool at each of the preset power consumption values. The server operation and management method also includes: Obtain the square root count of the stress testing tool; The number of square roots is used as the performance value of the server.

7. The server operation management method according to claim 1, characterized in that, The process of determining the power consumption value to be measured includes: Determine the candidate power consumption value adjacent to the power consumption value corresponding to the optimal energy efficiency ratio. If the current iteration number is 0, the candidate power consumption value is the preset power consumption value. If the current iteration number is not 0, the candidate power consumption value is the preset power consumption value and / or the power consumption value to be measured corresponding to any iteration number before the current iteration number. Calculate the average value of the power consumption value corresponding to the candidate power consumption value and the optimal energy efficiency ratio; The average value is used as the power consumption value to be measured corresponding to the current iteration number.

8. The server operation management method according to claim 1, characterized in that, The setting conditions include the difference between the power consumption value corresponding to the optimal energy efficiency ratio in the current iteration number and the power consumption value corresponding to the optimal energy efficiency ratio in the previous iteration number being less than a preset value.

9. The server operation management method according to claim 1, characterized in that, The setting condition is that the current iteration number is a preset number.

10. A server operation and management system, characterized in that, include: Determine the module and determine multiple preset power consumption values; The testing module is used to set each of the preset power consumption values ​​as the power consumption cap of the server, and to perform stress tests on the server under each preset power consumption value to obtain the maximum power consumption value and performance value of the server. The calculation module is used to calculate the optimal energy efficiency ratio based on all the maximum power consumption values ​​and all the performance values; The setting module is used to set the optimal power consumption value obtained according to the optimal energy efficiency ratio as the power consumption cap value of the server; the determining module is also used to determine the power consumption value to be measured. The testing module is also used to set the power consumption value to be tested as the power consumption cap value of the server, and to perform the stress test on the server under each power consumption value to be tested, so as to obtain the maximum power consumption value and performance value of the server. The calculation module is also used to calculate a new optimal energy efficiency ratio based on the maximum power consumption and performance value of the server under all the preset power consumption values ​​and the maximum power consumption and performance value of the server under all the power consumption values ​​to be tested; The server operation and management system also includes: The judgment module is used to determine whether the power consumption value corresponding to the new optimal energy efficiency ratio meets the setting conditions. If yes, the power consumption value corresponding to the new optimal energy efficiency ratio is taken as the optimal power consumption value. If not, the test module is triggered so that the test module can re-determine the power consumption value to be tested.

11. A server operation management device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the server operation management method as described in any one of claims 1-9 when executing the computer program.

12. A server, characterized in that, Includes the server operation management device as described in claim 11.

13. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the server operation management method as described in any one of claims 1-9.

Images