A fan heat dissipation algorithm evaluation method, device, equipment and medium

Abstract

The application relates to a fan heat dissipation algorithm evaluation method, device, equipment and medium. The method comprises the following steps: constructing test files with random content, random size and equal to a preset number in quantity for a compression test program; constructing running parameters for the compression test program; selecting one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time; executing the compression test program on a target server adopting the target fan heat dissipation algorithm to compress the test files with the preset number based on the running parameters; recording the execution time length, temperature value, whole machine power consumption and noise value of the target server executing the compression test program; returning to the step of selecting one from the plurality of fan heat dissipation algorithms to be evaluated as the target fan heat dissipation algorithm each time until all the fan heat dissipation algorithms to be evaluated are traversed; and calculating an evaluation value according to the recorded execution time length, temperature value, whole machine power consumption and noise value. The scheme of the application realizes quantitative evaluation of the quality of the fan heat dissipation algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of server testing, and in particular to a fan heat dissipation algorithm evaluation method, device, equipment and medium. BACKGROUND

[0002] Air-cooled servers mainly rely on fans for heat dissipation to ensure stable operation. However, there are many types of fan heat dissipation algorithms at present, and how to quickly land a suitable fan heat dissipation algorithm on the existing model has always been a research hotspot. Moreover, even for the same algorithm, how to quickly determine the parameters will also consume a certain amount of energy.

[0003] In the prior art, Chinese patent CN 113110962 A discloses a fan heat dissipation performance test method and test equipment, which discloses testing the physical performance of the fan. However, this test belongs to the category of material detection and does not involve fan heat dissipation algorithms, so there is an urgent need for a method for evaluating the quality of server fan heat dissipation algorithms. SUMMARY

[0004] Therefore, it is necessary to provide a fan heat dissipation algorithm evaluation method, device, equipment and medium to solve the above technical problems.

[0005] According to a first aspect of the present application, a fan heat dissipation algorithm evaluation method is provided, the method comprising:

[0006] constructing test files with random content, random size and equal to a preset number for a compression test program;

[0007] constructing running parameters for the compression test program, wherein the running parameters include execution times, interval times and the number of threads started each time;

[0008] selecting one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time;

[0009] executing the compression test program on a target server using the target fan heat dissipation algorithm to compress the preset number of test files based on the running parameters;

[0010] recording the execution duration, temperature value, whole machine power consumption and noise value of the target server executing the compression test program;

[0011] if the test of the current target fan heat dissipation algorithm is completed, returning to selecting one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time until all fan heat dissipation algorithms to be evaluated are traversed;

[0012] calculating the evaluation value of each fan heat dissipation algorithm to be evaluated according to the recorded execution duration, temperature value, whole machine power consumption and noise value.

[0013] In some embodiments, the method for constructing test files for the compression test program includes:

[0014] creating a file and generating a random number;

[0015] randomly reading a preset size of content from a preset file and appending it to the created file;

[0016] determining whether the size of the file exceeds the random number;

[0017] if the size of the file does not exceed the random number, returning to the step of randomly reading a preset size of content from a preset file and appending it to the created file;

[0018] if the size of the file exceeds the random number, confirming the current file generation;

[0019] determining whether the number of generated files reaches a preset number;

[0020] if the number of generated files does not reach the preset number, returning to the step of creating a file and generating a random number;

[0021] if the number of generated files reaches the preset number, taking all the generated files as test files.

[0022] In some embodiments, the method for constructing running parameters for the compression test program includes:

[0023] customizing the number of executions of the compression test program, wherein the number of executions is greater than or equal to two;

[0024] randomly generating a set of time intervals, wherein the set of time intervals is composed of a number of random times that is one less than the number of executions, and each random time is between a preset minimum interval time and a preset maximum interval time;

[0025] defining the number of start threads for each execution of the compression test program, wherein the number of start threads corresponding to each execution of the compression test program is between 1 and ten times the number of CPU cores of the target server.

[0026] In some embodiments, the method further includes:

[0027] if each fan cooling algorithm to be evaluated is traversed, returning to the step of constructing running parameters for the compression test program until the number of returns equals a preset number of rounds.

[0028] In some embodiments, the evaluation value of each fan cooling algorithm to be evaluated is calculated according to the recorded execution duration, temperature value, overall power consumption, and noise value, including:

[0029] According to the formula one, the average execution time of the compressed test procedure corresponding to the target fan heat dissipation algorithm is calculated:

[0030]

[0031] Wherein, Dura g represents the average execution time of the compressed test procedure corresponding to each target fan heat dissipation algorithm, TotalDura g represents the total execution time of all rounds of the target fan heat dissipation algorithm, and TotalN represents the preset number of rounds.

[0032] According to the formula two, the average temperature detection value corresponding to the target fan heat dissipation algorithm is calculated.

[0033]

[0034] Wherein, Temp g represents the average temperature detection value, and ∑ represents the cumulative operator symbol. represents the detection value of each temperature sensor at the qth second when testing the target fan heat dissipation algorithm, and is a vector; SensNum represents the total number of temperature sensors, represents the temperature sensor weighting coefficient vector, and 0≤WeightSensor s ≤1, WeightSensor s represents the weighting coefficient of the st temperature sensor, which is an element of the vector WeightSensor, wherein 0≤s≤SensNum.

[0035] When the target fan heat dissipation algorithm runs for the preset number of rounds, the total power consumption value of the target server is obtained.

[0036] According to the formula three, the average noise value corresponding to the target fan heat dissipation algorithm is calculated.

[0037]

[0038] Wherein, Noise g represents the average noise value, FrameN g represents the total number of frames after the recording with a time length of TotalDura g is divided into frames; ValFreq g,r,f represents the fth frequency point of the rth frame after the recording generated during the running of the target fan heat dissipation algorithm is divided into frames.

[0039] According to the formula four, the evaluation value of the target fan algorithm is calculated.

[0040] Measurement g =Weight T ×Valueg Formula Four;

[0041] wherein, Measurement g represents the evaluation value of the target fan heat dissipation algorithm, Weight = [W1 W2 W3 W4] represents the weighting coefficient of each measurement dimension, and 0≤W1,W2,W3,W4≤1.

[0042] In some embodiments, after each selection of a new target fan algorithm in the same round, the compression test procedure based on the running parameters is executed after waiting for a first preset time.

[0043] In some embodiments, the method further comprises:

[0044] sorting the evaluation values corresponding to all fan heat dissipation algorithms to be evaluated according to the size;

[0045] taking the lowest evaluation value corresponding to each fan heat dissipation algorithm to be evaluated as the optimal fan heat dissipation algorithm;

[0046] wherein, the plurality of fan heat dissipation algorithms to be evaluated refers to fan heat dissipation algorithms with the same algorithm and different parameters, and / or fan heat dissipation algorithms with different algorithms.

[0047] According to a second aspect of the present application, a fan heat dissipation algorithm evaluation device is provided, which comprises:

[0048] a file construction module configured to construct test files with random content, random size and equal to a preset number of test files for the compression test procedure;

[0049] a parameter construction module configured to construct running parameters for the compression test procedure, wherein the running parameters include the number of executions, the interval time and the number of threads started at each execution;

[0050] a selection module configured to select one from the plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time;

[0051] a test module configured to execute the compression test procedure based on the running parameters on the target server using the target fan heat dissipation algorithm to compress the preset number of test files;

[0052] a recording module configured to record the execution time, temperature value, overall power consumption and noise value of the target server executing the compression test procedure;

[0053] a traversal module configured to return to the selection of one from the plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time if the test of the current target fan heat dissipation algorithm is completed, until all fan heat dissipation algorithms to be evaluated are traversed.

[0054] The calculation module is configured to calculate the evaluation value of each fan cooling algorithm to be evaluated based on the recorded execution time, temperature value, total power consumption, and noise value.

[0055] According to a third aspect of the present invention, a computer device is also provided, the computer device comprising:

[0056] At least one processor; and

[0057] The memory stores computer programs that can run on the processor, and the processor executes the aforementioned fan cooling algorithm evaluation method when executing the program.

[0058] According to a fourth aspect of the present invention, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, performs the aforementioned fan cooling algorithm evaluation method.

[0059] The aforementioned fan cooling algorithm evaluation method achieves quantitative evaluation of fan cooling algorithm quality, solving the pain points of lacking objective evaluation methods and indicators for fan cooling algorithms and slow implementation of fan cooling algorithms. It can quickly evaluate and implement the quality of fan cooling algorithms, and forms an objective standard that can be used to compare the quality of various fan cooling algorithms, select the optimal algorithm, improve the overall performance of the server, and is low in cost and easy to promote.

[0060] In addition, the present invention also provides a fan cooling algorithm evaluation device, a computer device, and a computer-readable storage medium, which can achieve the above-mentioned technical effects, and will not be described in detail here. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0062] Figure 1 A flowchart illustrating a fan cooling algorithm evaluation method provided in one embodiment of the present invention;

[0063] Figure 2 A flowchart illustrating the overall implementation of another fan cooling algorithm evaluation method provided in another embodiment of the present invention;

[0064] Figure 3 A flowchart illustrating the construction of a test file is provided as an embodiment of the present invention;

[0065] Figure 4 a flow chart of running a compression test program provided by an embodiment of the present application;

[0066] Figure 5 a structural schematic diagram of a fan heat dissipation algorithm evaluation device provided by another embodiment of the present application;

[0067] Figure 6 an internal structure diagram of a computer device in another embodiment of the present application. DETAILED DESCRIPTION

[0068] In order to make the objectives, technical solutions and advantages of the present application clearer and more comprehensible, the embodiments of the present application are further described in detail below with reference to the accompanying drawings.

[0069] It should be noted that all the expressions of "first" and "second" in the embodiments of the present application are used to distinguish two same-named different entities or different parameters, and it can be seen that "first" and "second" are only used for the convenience of description, and should not be understood as a limitation on the embodiments of the present application, and the subsequent embodiments will not be described one by one.

[0070] In an embodiment, referring to FIG. 1, the present application provides a fan heat dissipation algorithm evaluation method 100, specifically, the method comprises the following steps: Figure 1

[0071] Step 101, constructing test files with random content, random size and equal to a preset number for a compression test program;

[0072] Step 102, constructing running parameters for the compression test program, wherein the running parameters include execution times, interval time and the number of threads started each time;

[0073] Step 103, selecting one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time;

[0074] Step 104, executing the compression test program based on the running parameters on a target server using the target fan heat dissipation algorithm to compress the preset number of test files;

[0075] Step 105, recording the execution duration, temperature value, whole machine power consumption and noise value of the target server executing the compression test program;

[0076] Step 106, if the test of the current target fan heat dissipation algorithm is completed, returning to step 103 to execute the selection of one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time, until all the fan heat dissipation algorithms to be evaluated are traversed;

[0077] ​Step 107, calculating the evaluation value of each fan heat dissipation algorithm to be evaluated according to the recorded execution duration, temperature value, whole machine power consumption and noise value.

[0078] The fan heat dissipation algorithm evaluation method realizes quantitative evaluation of the quality of the fan heat dissipation algorithm, solves the pain points of no objective evaluation method and index for the fan heat dissipation algorithm and slow landing of the fan heat dissipation algorithm, can quickly evaluate and land the quality of the fan heat dissipation algorithm, and forms an objective standard that can be used for horizontal comparison of the quality of various fan heat dissipation algorithms, selection of the optimal algorithm and improvement of the performance of the whole server, and has low cost and is easy to popularize.

[0079] In some embodiments, step 101, a plurality of test files with random content, random size and equal to a preset number of quantity are constructed for the compression test program, including:

[0080] Creating a file and generating a random number;

[0081] Randomly reading a preset size of content from a preset file and appending it to the created file;

[0082] Judging whether the size of the file exceeds the random number;

[0083] If the size of the file does not exceed the random number, return to execute the step of randomly reading a preset size of content from a preset file and appending it to the created file;

[0084] If the size of the file exceeds the random number, confirm the current file generation;

[0085] Judging whether the number of generated files reaches the preset number;

[0086] If the number of generated files does not reach the preset number, return to execute the step of creating a file and generating a random number;

[0087] If the number of generated files reaches the preset number, all generated files are used as test files.

[0088] In some embodiments, step 102, the running parameters are constructed for the compression test program, including:

[0089] Customizing the execution times of the compression test program, wherein the execution times are greater than or equal to two times;

[0090] Randomly generating a set of time intervals, wherein the set of time intervals is composed of one less random time than the execution times, and each random time is between a preset minimum interval time and a preset maximum interval time;

[0091] Defining the number of starting threads for each execution of the compression test program, wherein the number of starting threads corresponding to each execution of the compression test program is between 1 and ten times the number of CPU cores of the target server.

[0092] In some embodiments, the method further comprises:

[0093] If each fan cooling algorithm to be evaluated is traversed, returning to the step of constructing the running parameters for the compression test program until the number of returns is equal to the preset number of rounds.

[0094] In some embodiments, step 107, calculating the evaluation value of each fan cooling algorithm to be evaluated according to the recorded execution duration, temperature value, total machine power consumption and noise value, comprises:

[0095] According to formula one, calculating the average execution duration of the target fan cooling algorithm corresponding to the compression test program:

[0096]

[0097] Wherein, Dura g represents the average execution duration of the compression test program corresponding to each target fan cooling algorithm, TotalDura g represents the total execution time of all rounds of the target fan cooling algorithm, and TotalN represents the preset number of rounds.

[0098] According to formula two, calculating the average temperature detection value corresponding to the target fan cooling algorithm:

[0099]

[0100] Wherein, Temp g represents the average temperature detection value, and ∑ represents the cumulative operator symbol. represents the detection value of each temperature sensor at the qth second when testing the target fan cooling algorithm, and is a vector; SensNum is the total number of temperature sensors, represents the temperature sensor weighting coefficient vector, and 0≤WeightSensor s ≤1, WeightSensor s represents the weighting coefficient of the st temperature sensor, which is an element of the vector WeightSensor, wherein 0≤s≤SensNum.

[0101] Obtaining the total machine power consumption value of the target server when the target fan cooling algorithm runs for the preset number of rounds;

[0102] According to formula three, calculating the average noise value corresponding to the target fan cooling algorithm:

[0103]

[0104] wherein, Noise g represents the noise mean value, FrameN g represents the total number of frames after the recording with a time length of TotalDura g is framed; ValFreq g,r,f represents the fth frequency point of the rth frame after the recording generated during the running of the target fan heat dissipation algorithm is framed.

[0105] The evaluation value of the target fan algorithm is calculated according to Formula Four.

[0106] Measurement g = Weight T * Value g Formula Four.

[0107] wherein, Measurement g represents the evaluation value of the target fan heat dissipation algorithm, Weight = [W1 W2 W3 W4] represents the weighting coefficient of each measurement dimension, and 0≤W1,W2,W3,W4≤1.

[0108] In some embodiments, after selecting a new target fan algorithm in each round, waiting for a first preset time before executing the compression test procedure based on the running parameters.

[0109] In some embodiments, the method further comprises:

[0110] sorting the evaluation values corresponding to all fan heat dissipation algorithms to be evaluated according to the size;

[0111] taking the lowest evaluation value corresponding to each fan heat dissipation algorithm to be evaluated as the optimal fan heat dissipation algorithm;

[0112] wherein, the plurality of fan heat dissipation algorithms to be evaluated refer to fan heat dissipation algorithms with the same algorithm and different parameters, and / or fan heat dissipation algorithms with different algorithms.

[0113] In yet another embodiment, in order to solve the pain points in the prior art that there is no objective evaluation method and index for fan heat dissipation algorithms, and fan heat dissipation algorithms are slow to land, and in order to achieve the purpose of quickly measuring alternative fan heat dissipation algorithms, accelerating the landing efficiency, and improving the overall performance of the server, another embodiment of the present application provides another fan heat dissipation algorithm evaluation method, which is described in detail in combination with Figure 2As shown, the specific implementation includes four parts: the first part is global parameters and compression test program settings; the second part is to generate a test file set and running parameters; the third part is to run the compression test program while recording data; and the fourth part is to evaluate the algorithm quality. For ease of understanding, the specific implementation will be described in detail below in combination with each part.

[0114] Regarding the first part, global parameters and compression test program settings:

[0115] For TotalAlgo fan cooling algorithms, a total of TotalN tests are performed, each of which traverses the cooling algorithm, runs the custom compression test program, and finally takes the test average to evaluate the quality of each algorithm. Among them, TotalAlgo refers to the number of fan cooling algorithms to be evaluated; the larger the value of TotalN, the more accurate the evaluation, but at the same time, the test will also be more time-consuming, and the value of TotalN is recommended to be 100. In each round of testing, the server is kept offline and single-machine running, and the running software and hardware platforms are kept fixed and unchanged; and the surrounding natural environment factors such as temperature, humidity, wind speed, altitude, and other key physical factors are kept stable and unchanged. This will ensure that the final test results are objective and reliable, serving as a measurement standard.

[0116] Regarding the second part, generating a test file set and running parameters:

[0117] (1) Generate a test file set FileSet, which contains FileNum files, each with a size of 1Kb~SizeMax Kb. The recommended value of FileNum is 100000, and SizeMax is 16. The specific value can be based on the hard disk capacity of the test platform, and should not exceed the upper limit of its available capacity. The specific generation process can refer to the flow shown in Figure 3 The flow shown in the figure constructs the test file, so that the file set FileSet in each round of testing contains random file content and random size, ensuring the objectivity of the test.

[0118] (2) Generate compression test program running parameters,

[0119] The custom test program is ProgUnit, which is run Times times in each round, and the number of ProgUnit running each time is Times is recommended to be 100, the larger the evaluation is more accurate, but the evaluation process will be more time-consuming. Among them, ProgUnit is a custom program: that is, compressing the file set FileSet generated in the last step. Taking the Linux platform as an example, the application calls the zip standard command, and at the same time outputs the compressed result file to the / dev / null device (i.e. the recycle bin), without the need to test and then delete the compressed file, avoiding insufficient hard disk storage capacity. Among them, ProcNum represents a one-dimensional vector with a dimension of Timesx1, and its elements are pn m , ProcNumMin≤pn m ≤ProcNumMax, where pn m is an element belonging to ProcNum, 1≤m≤Times, ProcNumMin is the minimum number of started compression programs, and ProcNumMax is the maximum number of started compression programs. The application recommends the following values: ProcNumMin=1, ProcNumMax=CPU core number of the running platform x 10.

[0120] At the same time, a set of random time intervals is generated: whose elements are ti n , TimeInteMin≤ti n ≤TimeInteMax, where ti n is an element belonging to TimeInte, 1≤n≤Times-1, TimeInteMin is the minimum time interval, and TimeInteMax is the maximum time interval. The application recommends the following values: ProcNumMin=1, ProcNumMax=300, unit: seconds.

[0121] Regarding the third part of running the compression test program and recording data:

[0122] Please refer to Figure 4 , based on the set parameters, generated file set, running parameters, start the compression test program, and record the duration of each server heat dissipation algorithm running the compression test program, all temperature sensor detection values, server whole machine power consumption and server generated noise, and end the test program based on the determination condition shown in Figure 4 , the end conditions are as follows: whether the test is completed for the set number of rounds. And whether the test is completed for all fan heat dissipation algorithms.

[0123] Regarding the fourth part of evaluating algorithm quality:

[0124] If the server component is overheated and down during the test, the fan cooling algorithm is directly excluded. The embodiment proposes four measurement dimensions: server performance, component temperature, overall power consumption and noise generation, which are integrated into a scalar for evaluation by the following Formula Five:

[0125] Measurement g = Weight T × Value g Formula Five;

[0126] wherein Measurement g represents the final evaluation value of the gth fan cooling algorithm, 1≤g≤TotalAlgo; Weight=[W1 W2 W3 W4] T represents the weighting coefficient of each measurement dimension, and 0≤W1,W2,W3,W4≤1, which can be adjusted based on the specific project preferences. The greater the weight coefficient, the more important the dimension is; X T represents the transposition of matrix X;

[0127] Value g represents the measurement value of the gth fan cooling algorithm when it is running, specifically Value g =[Dura g Temp g Power g Noise g ] T The calculation methods of Dura g , Temp g , Power g , and Noise g will be described in detail below:

[0128] Dura g represents the average running time in seconds when the gth fan cooling algorithm finishes running the test program, and the specific calculation method is shown in the following Formula Six:

[0129]

[0130] wherein TotalDura g represents the total running time of all rounds of the gth cooling algorithm, in seconds.

[0131] Temp g represents the average detection value of the temperature sensor during the running of the gth fan cooling algorithm, and its calculation method is shown in the following Formula Seven:

[0132]

[0133] wherein, ∑ represents the cumulative operator symbol; represents the detection value of each temperature sensor at the qth second when testing the gth heat dissipation algorithm, which is a vector; SensNum represents the total number of temperature sensors; represents the temperature sensor weighting coefficient vector, and 0≤WeightSensor s ≤1, WeightSensor s represents the weighting coefficient of the st temperature sensor, which is an element of the vector WeightSensor, wherein, 0≤s≤SensNum. The weighted cumulative operation on each temperature sensor detection value can be adjusted based on the importance of the server components in the specific project. The more important a certain component is, the greater the corresponding weight coefficient is.

[0134] Power g represents the total power consumption of the whole machine after running the gth fan heat dissipation algorithm for all rounds of test procedures;

[0135] Noise g represents the average noise generated during the test of the gth fan heat dissipation algorithm. The specific calculation method is as follows: record the noise using an 8K sampling rate, then perform 512-point FFT based on the overlap-save method, then perform A-weighting on the calculated noise spectrum, and finally output 256 valid value points, the calculation method is shown in the following formula eight:

[0136]

[0137] wherein, FrameN g represents the total number of frames after the recording with a duration of TotalDura g is divided into frames; ValFreq g,r,f represents the fth frequency point of the rth frame after the recording generated during the running of the gth fan heat dissipation algorithm is divided into frames.

[0138] Finally, sort the evaluation value Measurement g of each fan heat dissipation algorithm, and the smaller the value is, the better the quality of the fan heat dissipation algorithm is.

[0139] The fan heat dissipation algorithm evaluation method of the embodiment has at least the following beneficial technical effects: first, the quality of the fan heat dissipation algorithm can be quickly evaluated and landed, and an objective test process is provided to quickly evaluate the selected fan heat dissipation algorithm and accelerate the landing; second, an objective and unified evaluation standard is proposed, which can be used for horizontal comparison of the quality of each fan heat dissipation algorithm to select the optimal algorithm and improve the overall performance of the server; finally, it does not need to design specific hardware for the running platform, but is realized based on algorithm logic and software coding, which has low implementation cost and is easy to promote.

[0140] In some embodiments, referring to Figure 5 The application also provides a fan heat dissipation algorithm evaluation device 200, which comprises:

[0141] a file construction module 201 configured to construct test files with random content, random size and equal to a preset number of times for a compression test program;

[0142] a parameter construction module 202 configured to construct running parameters for the compression test program, wherein the running parameters comprise execution times, interval times and the number of threads started each time;

[0143] a selection module 203 configured to select one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time;

[0144] a test module 204 configured to execute the compression test program based on the running parameters on a target server using the target fan heat dissipation algorithm to compress the test files with a preset number of times;

[0145] a recording module 205 configured to record the execution time, temperature value, whole machine power consumption and noise value of the target server executing the compression test program;

[0146] a traversal module 206 configured to return to selecting one from a plurality of fan heat dissipation algorithms to be evaluated as a target fan heat dissipation algorithm each time if the test of the current target fan heat dissipation algorithm is completed, until all the fan heat dissipation algorithms to be evaluated are traversed;

[0147] a calculation module 207 configured to calculate the evaluation value of each fan heat dissipation algorithm to be evaluated according to the recorded execution time, temperature value, whole machine power consumption and noise value.

[0148] The above fan heat dissipation algorithm evaluation device realizes quantitative evaluation of the quality of fan heat dissipation algorithms, solves the pain points of no objective evaluation method and index for fan heat dissipation algorithms and slow landing of fan heat dissipation algorithms, can quickly evaluate and land the quality of fan heat dissipation algorithms, and forms an objective standard which can be used for horizontal comparison of the quality of various fan heat dissipation algorithms, selection of the optimal algorithm and improvement of the performance of the whole server, and has low cost and is easy to popularize.

[0149] In some embodiments, the file construction module 201 is further configured to:

[0150] create a file and generate a random number;

[0151] randomly read a content with a preset size from a preset file and append it to the created file;

[0152] determining whether the size of the file exceeds the random number;

[0153] if the size of the file does not exceed the random number, returning to perform the step of randomly reading a preset size of content from a preset file and appending to the created file;

[0154] if the size of the file exceeds the random number, confirming the current file generation;

[0155] determining whether the number of generated files reaches a preset number;

[0156] if the number of generated files does not reach the preset number, returning to perform the step of creating a file and generating a random number;

[0157] if the number of generated files reaches the preset number, taking all generated files as test files.

[0158] In some embodiments, the parameter construction module 202 is further configured to:

[0159] customizing the number of execution times of the compression test program, wherein the number of execution times is greater than or equal to two;

[0160] randomly generating a set of time intervals, wherein the set of time intervals is composed of one less random time than the number of execution times, and each random time is between a preset minimum interval time and a preset maximum interval time;

[0161] defining the number of start threads for each execution of the compression test program, wherein the number of start threads corresponding to each execution of the compression test program is between 1 and ten times the number of CPU cores of the target server.

[0162] In some embodiments, the apparatus further comprises a module configured to perform the following steps:

[0163] if each fan cooling algorithm to be evaluated is traversed, returning to perform the step of constructing running parameters for the compression test program until the number of returns is equal to a preset number of rounds.

[0164] In some embodiments, the calculation module 207 is further configured to:

[0165] calculating the average execution time of the compression test program corresponding to the target fan cooling algorithm according to Formula One:

[0166]

[0167] wherein Dura g represents the average execution time of the compression test program corresponding to each target fan cooling algorithm, TotalDura gTotalN represents the preset number of rounds;

[0168] The temperature average detection value corresponding to the target fan heat dissipation algorithm is calculated according to Formula Two;

[0169]

[0170] wherein, Temp g represents the temperature average detection value, and ∑ represents the cumulative operator symbol; represents the detection value of each temperature sensor at the qth second when testing the target fan heat dissipation algorithm, which is a vector; SensNum is the total number of temperature sensors, represents the temperature sensor weighting coefficient vector, and 0≤WeightSensor s ≤1, WeightSensor s represents the weighting coefficient of the st temperature sensor, which is an element of the vector WeightSensor, wherein 0≤s≤SensNum;

[0171] The total power consumption value of the target server when the target fan heat dissipation algorithm runs for the preset number of rounds is obtained;

[0172] The noise average value corresponding to the target fan heat dissipation algorithm is calculated according to Formula Three;

[0173]

[0174] wherein, Noise g represents the noise average value, FrameN g represents the total number of frames after the recording with a time length of TotalDura g is divided into frames; ValFreq g,r,f represents the fth frequency point of the rth frame after the recording generated during the running of the target fan heat dissipation algorithm is divided into frames;

[0175] The evaluation value of the target fan algorithm is calculated according to Formula Four;

[0176] Measurement g = Weight T × Value g Formula Four;

[0177] wherein, Measurement g represents the evaluation value of the target fan heat dissipation algorithm, Weight = [W1 W2 W3 W4] represents the weighting coefficient of each measurement dimension, and 0≤W1,W2,W3,W4≤1.

[0178] In some embodiments, after selecting a new target fan algorithm in each round, the method waits for a first preset time before executing the compression test procedure based on the running parameters.

[0179] In some embodiments, the device further comprises a module configured to perform the following steps:

[0180] Ranking the evaluation values corresponding to all fan heat dissipation algorithms to be evaluated according to their sizes;

[0181] Taking the lowest evaluation value corresponding to each fan heat dissipation algorithm to be evaluated as the optimal fan heat dissipation algorithm;

[0182] The plurality of fan heat dissipation algorithms to be evaluated refer to fan heat dissipation algorithms with the same algorithm but different parameters, and / or fan heat dissipation algorithms with different algorithms.

[0183] It should be noted that the specific limitations of the fan heat dissipation algorithm evaluation device can be referred to the limitations of the fan heat dissipation algorithm evaluation method described above, which will not be repeated here. The various modules in the above fan heat dissipation algorithm evaluation device can be realized by software, hardware and their combinations, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above modules by the processor.

[0184] According to another aspect of the present application, a computer device is provided, which can be a server, and its internal structure is shown in Figure 6 The computer device comprises a processor, a memory, a network interface and a database connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement the above-mentioned fan heat dissipation algorithm evaluation method. Specifically, the method comprises the following steps:

[0185] Constructing test files with random content, random size and equal to a preset number of quantity for the compression test procedure;

[0186] Constructing running parameters for the compression test procedure, wherein the running parameters include the number of executions, the interval time and the number of threads started at each execution;

[0187] selecting one of the plurality of fan heat dissipation algorithms as a target fan heat dissipation algorithm each time;

[0188] executing the compression test program on the target server based on the running parameters to compress the preset number of test files;

[0189] recording the execution duration, temperature value, overall power consumption and noise value of the target server executing the compression test program;

[0190] if the test on the current target fan heat dissipation algorithm is completed, returning to selecting one of the plurality of fan heat dissipation algorithms as a target fan heat dissipation algorithm each time until all the fan heat dissipation algorithms are traversed;

[0191] calculating the evaluation value of each of the plurality of fan heat dissipation algorithms according to the recorded execution duration, temperature value, overall power consumption and noise value.

[0192] According to still another aspect of the present application, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the fan heat dissipation algorithm evaluation method as described above, specifically including the following steps:

[0193] constructing test files with random content, random size and equal to the preset number of test files for the compression test program;

[0194] constructing running parameters for the compression test program, wherein the running parameters include the number of executions, interval time and the number of threads started each time;

[0195] selecting one of the plurality of fan heat dissipation algorithms as a target fan heat dissipation algorithm each time;

[0196] executing the compression test program on the target server based on the running parameters to compress the preset number of test files;

[0197] recording the execution duration, temperature value, overall power consumption and noise value of the target server executing the compression test program;

[0198] if the test on the current target fan heat dissipation algorithm is completed, returning to selecting one of the plurality of fan heat dissipation algorithms as a target fan heat dissipation algorithm each time until all the fan heat dissipation algorithms are traversed;

[0199] calculating the evaluation value of each of the plurality of fan heat dissipation algorithms according to the recorded execution duration, temperature value, overall power consumption and noise value.

[0200] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, databases, or other media in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0201] The technical features of the above embodiments can be combined in any way. In order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0202] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for evaluating fan cooling algorithms, characterized in that, The method includes: To construct test files with random content, random size, and a preset number of files for the compression test program; The runtime parameters for the compression test program are constructed, including the number of executions, the interval time, and the number of threads started each time the program is executed. One fan cooling algorithm is selected as the target fan cooling algorithm from multiple fan cooling algorithms to be evaluated each time. On the target server employing the target fan cooling algorithm, the compression test program is executed based on the operating parameters to compress a preset number of test files; Record the execution time, temperature, power consumption, and noise level of the target server when executing the compression test program; If the test of the current target fan cooling algorithm is completed, return to the previous step and select one of the multiple fan cooling algorithms to be evaluated as the target fan cooling algorithm each time, until all the fan cooling algorithms to be evaluated have been traversed. The evaluation value of each fan cooling algorithm to be evaluated is calculated based on the recorded execution time, temperature value, total power consumption, and noise value.

2. The fan cooling algorithm evaluation method according to claim 1, characterized in that, To construct test files for the compression test program, the content, size, and number of files must be randomized and equal to a preset number. This includes: Create a file and generate random numbers; Randomly read content of a preset size from a preset file and append it to the created file; Determine if the file size exceeds the random number; If the file size does not exceed the random number, then return to the step of randomly reading content of a preset size from a preset file and appending it to the created file; If the file size exceeds the random number, then confirm the generation of the current file; Determine if the number of generated files has reached the preset number; If the number of generated files has not reached the preset number, return to the step of creating files and generating random numbers; If the number of generated files reaches the preset number, all generated files will be used as test files.

3. The fan cooling algorithm evaluation method according to claim 1, characterized in that, To construct runtime parameters for the compression test program, including: The number of times the custom compression test program is executed, wherein the number of executions is greater than or equal to two; A random time interval set is generated, wherein the time interval set consists of a number of random times less than the number of executions, and each random time interval is between a preset minimum time interval and a preset maximum time interval; Define the number of threads to start for each execution of the compression test program. The number of threads to start for each execution of the compression test program is between 1 and ten times the number of CPU cores of the target server.

4. The fan cooling algorithm evaluation method according to claim 1, characterized in that, The method further includes: If each fan cooling algorithm to be evaluated has been traversed, the process returns to the step of constructing running parameters for the compression test program, until the number of returns equals the preset number of rounds.

5. The fan cooling algorithm evaluation method according to claim 4, characterized in that, The evaluation value for each fan cooling algorithm to be evaluated is calculated based on the recorded execution time, temperature value, total power consumption, and noise value, including: Calculate the average execution time of the compression test program corresponding to the target fan cooling algorithm using Formula 1: Among them, Dura g TotalDura represents the average execution time of the compression test program corresponding to each target fan cooling algorithm. g TotalN represents the total execution time of all rounds of the target fan cooling algorithm, where TotalN represents the preset number of rounds. Calculate the average temperature detection value corresponding to the target fan cooling algorithm according to Formula 2; Among them, Temp g This represents the average temperature reading, and ∑ represents the accumulation operator. This represents the detection value of each temperature sensor at second q during the testing of the target fan cooling algorithm, expressed as a vector; SensNum represents the total number of temperature sensors. Let represent the weighted coefficient vector of the temperature sensor, and 0 ≤ WeightSensor s ≤1, WeightSensor s Let represent the weighting coefficient of the s-th temperature sensor, which is an element of the vector WeightSensor, where 0≤s≤SensNum; Obtain the total power consumption of the target server after the target fan cooling algorithm has completed the preset number of runs; Calculate the average noise value corresponding to the target fan cooling algorithm according to Formula 3; Among them, Noise g FrameN represents the noise mean. g This indicates a duration of TotalDura g The total number of frames after the recording is framed; ValFreq g,r,f This represents the f-th frequency point of the r-th frame after the recording frames generated during the operation of the target fan cooling algorithm; The evaluation value of the target fan algorithm is calculated according to Formula 4; Measurement g =Weight T ×Value g Official 4; Among them, Measurement g This represents the evaluation value of the target fan cooling algorithm. Weight = [W1 W2 W3 W4] represents the weighting coefficient for each measurement dimension, and 0 ≤ W1, W2, W3, W4 ≤ 1.

6. The fan cooling algorithm evaluation method according to claim 4, characterized in that, Within the same round, after selecting a new target fan algorithm each time, wait for a first preset time before executing the compression test program based on the running parameters. The interval between different rounds is a second preset time.

7. The fan cooling algorithm evaluation method according to claim 1, characterized in that, The method further includes: Sort the evaluation values ​​of all the fan cooling algorithms to be evaluated by size; The lowest evaluation value among all the fan cooling algorithms to be evaluated is taken as the optimal fan cooling algorithm. Among them, multiple fan cooling algorithms to be evaluated refer to fan cooling algorithms with the same algorithm but different parameters, and / or fan cooling algorithms with different algorithms.

8. A fan cooling algorithm evaluation device, characterized in that, The device includes: The file construction module is configured to construct test files for the compression test program with random content, random size, and a number equal to a preset number. The parameter construction module is configured to construct runtime parameters for the compression test program, wherein the runtime parameters include the number of executions, the interval time, and the number of threads started each time the program is executed. The selection module is configured to select one fan cooling algorithm from multiple fan cooling algorithms to be evaluated as the target fan cooling algorithm each time. The testing module is configured to execute the compression test program on a target server employing the target fan cooling algorithm, based on the operating parameters, to compress a preset number of test files. The recording module is configured to record the execution time, temperature value, total power consumption, and noise value of the target server when executing the compression test program. The traversal module is configured to return to the execution if the test of the current target fan cooling algorithm is completed, and select one of the multiple fan cooling algorithms to be evaluated as the target fan cooling algorithm each time, until all fan cooling algorithms to be evaluated are traversed. The calculation module is configured to calculate the evaluation value of each fan cooling algorithm to be evaluated based on the recorded execution time, temperature value, total power consumption, and noise value.

9. A computer device, characterized in that, include: At least one processor; as well as A memory storing a computer program executable in the processor, wherein the processor executes the program to perform the method according to any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it performs the method described in any one of claims 1-7.

Images