Method and device for testing performance of storage unit and electronic equipment

Abstract

The invention relates to the technical field of storages, and discloses a test method and device for performance of a storage unit and electronic equipment, the test method for the performance of the storage unit comprises the following steps: step S101, obtaining hardware parameters of hardware equipment where a solid state disk is located; s102, generating a hard disk performance test condition; step S103, generating a hardware score; step S104, executing the hard disk performance test condition to generate a first performance score in the sequential read mode and the random read mode respectively; step S105, executing the hard disk performance test condition to generate a second performance score in the sequential write mode and the random write mode respectively; s106, generating a comprehensive score of the solid state disk; according to the method, the hard disk performance test condition and the hardware score are generated according to the hardware parameters of the hardware equipment where the solid state disk is located, the performance scores are obtained in different read-write modes, and deviation generated due to different hardware configurations is eliminated by subtracting the hardware score from the performance score.

Description

Technical Field

[0001] The present invention relates to the field of memory technology, and more specifically, to a method and device for testing the performance of a storage unit and electronic equipment. Background Art

[0002] With the rapid development of information technology, the demand for data storage is growing. Solid-state drives (SSDs) are becoming the mainstream storage devices on the market because they can directly access storage units through electronic signals and do not require a magnetic head seek process like a mechanical hard disk (HDD). They can read and write data at high speeds. As the storage density and capacity of SSDs continue to increase, the testing and evaluation of their performance is also crucial.

[0003] Existing SSD performance testing methods mainly test the read and write speeds of files of different sizes, such as using standard testing tools for sequential read and write or random read and write tests. Standard testing tools can be CrystalDiskMark or ATTO Disk Benchmark, etc. However, existing SSD performance testing methods often ignore the impact of hardware configuration on SSD performance. SSDs will also have read and write differences under different hardware configurations, such as CPU, memory, and motherboard, which may cause deviations in SSD performance test results due to different hardware configurations. In addition, in order to eliminate the impact of hardware configuration, the SSD is disassembled and installed on a unified hardware device for testing. This method is not only cumbersome to operate, but may also cause physical damage to the SSD.

[0004] Therefore, a performance testing method for a solid state drive is urgently needed to solve the above problems. Summary of the invention

[0005] The present invention provides a method, device and electronic device for testing the performance of a storage unit, which solves the technical problem that the existing solid-state hard disk performance testing method in the related art ignores the influence of the hardware configuration of the hardware device where the solid-state hard disk is located on its read and write performance, resulting in deviations in its performance test results due to different hardware configurations.

[0006] The present invention provides a method for testing the performance of a storage unit, comprising the following steps: Step S101, obtaining hardware parameters of the hardware device where the solid state drive is located; Hardware parameters include: number of CPU cores, CPU main frequency, CPU cache size, memory size, memory frequency, power supply power and cooling type; Step S102, generating hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state hard disk is located; Hard disk performance test conditions include: test file size, queue depth and test duration; Step S103, generating a hardware score through a second neural network model according to the hardware parameters of the hardware device where the solid state drive is located; The hardware score ranges from 1 to 10. A larger hardware score indicates better performance of the hardware device where the SSD is located. Step S104, executing the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generating a first performance score according to the first test result; The first test results include: sequential read speed, random read speed, SSD temperature, and CPU temperature; A larger value of the first performance score indicates a better data reading performance of the solid state drive; Step S105, executing the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generating a second performance score according to the second test result; The second test results include: sequential write speed, sequential write amplification factor, random write speed, random write amplification factor, SSD temperature, and CPU temperature; A larger value of the second performance score indicates a better data writing performance of the solid state drive; Step S106: Generate a comprehensive score of the solid state drive according to the hardware score, the first performance score, and the second performance score.

[0007] Furthermore, the hardware parameters are represented by eight-bit binary codes, and the heat dissipation types include: air cooling and liquid cooling.

[0008] Further, the first neural network model includes: a feature conversion layer, a feature extraction layer, a maximum pooling layer, a feature expansion layer and a first classifier; The feature conversion layer inputs hardware parameters and outputs a feature matrix; The calculation formula of the feature conversion layer is as follows: F = W × X, where F represents the feature matrix, X represents the hardware parameters, which are represented by a 1 × 7 vector, and W represents the weight parameters, which are represented by a 32 × 1 vector. The size of F is 32 × 7. The feature extraction layer includes a first convolution kernel and a second convolution kernel, which respectively extract features from the feature matrix to obtain a first feature map and a second feature map, wherein the sizes of the first convolution kernel and the second convolution kernel are 3×3 and 5×5 respectively, and the step sizes are both 1, so the sizes of the first feature map and the second feature map are 30×5 and 28×3 respectively; The first feature map and the second feature map are input into the maximum pooling layer, and the third feature map and the fourth feature map are output. The size of the pooling window of the maximum pooling layer is 3×3, and the step size is 1. Then, the sizes of the third feature map and the fourth feature map are 28×3 and 26×1 respectively. The feature expansion layer is used to expand the third feature map and the fourth feature map into vector representations and concatenate them to obtain a combined vector. The size of the combined vector is 110×1. The combined vector is input to the first classifier, and the classification space of the first classifier represents the hardware score.

[0009] Further, the second neural network model includes a first hidden layer, a second hidden layer, a second classifier, a third classifier and a fourth classifier; The first hidden layer inputs hardware parameters and outputs a first update vector; The calculation formula for the first hidden layer is as follows: ; in represents the first update vector of the first hidden layer output, X represents the hardware parameters of the first hidden layer input, represented by a 1×7 size vector, represents the first weight parameter, represents the second weight parameter, Swish represents the Swish activation function; The second hidden layer inputs the first update vector and outputs the second update vector; The calculation formula for the second hidden layer is as follows: ; in represents the second update vector for the output of the second hidden layer, represents the second weight parameter, represents the second bias parameter; The second update vector is input into the second classifier, the third classifier and the fourth classifier respectively, and the classification spaces of the second classifier, the third classifier and the fourth classifier respectively represent the test file size, the queue depth and the test duration.

[0010] Furthermore, sample labels of training samples used to train the first neural network model and the second neural network model are obtained by manual labeling.

[0011] Furthermore, the first performance score The calculation formula is as follows: ; in Indicates the sequential read speed, Indicates random read speed, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

[0012] Furthermore, the second performance score The calculation formula is as follows: ; in Indicates the sequential write speed, Indicates random write speed, represents the sequential write amplification factor, represents the random write amplification factor, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

[0013] Furthermore, the calculation formula of the comprehensive score Score is as follows: ; in and They represent the first performance score and the second performance score respectively, handware represents the hardware score, , and represent the first weight coefficient, the second weight coefficient and the third weight coefficient respectively, , and are custom parameters, and + + =1.

[0014] The present invention provides a storage unit performance testing device, comprising: A hardware parameter acquisition module is used to obtain the hardware parameters of the hardware device where the solid state drive is located; A hard disk performance test condition generation module, which is used to generate hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state hard disk is located; A hardware score generating module, which is used to generate a hardware score through a second neural network model according to hardware parameters of the hardware device where the solid-state hard disk is located; A first performance score generating module, which is used to execute the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generate a first performance score according to the first test result; A second performance score generating module, which is used to execute the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generate a second performance score according to the second test result; The comprehensive score generating module is used to generate a comprehensive score of the solid state drive according to the hardware score, the first performance score and the second performance score.

[0015] The present invention provides an electronic device for testing the performance of a storage unit, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the above-mentioned method for testing the performance of a storage unit is executed.

[0016] The beneficial effect of the present invention is that the present invention generates hard disk performance test conditions and hardware scores according to the hardware parameters of the hardware device where the solid-state hard disk is located, and obtains performance scores under different read and write modes respectively, and eliminates the deviation caused by different hardware configurations by subtracting the hardware score from the performance score, thereby improving the accuracy of the solid-state hard disk performance test. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of a method for testing performance of a storage unit of the present invention; Figure 2 It is a schematic diagram of a storage unit performance testing device of the present invention.

[0018] In the figure: a hardware parameter acquisition module 201, a hard disk performance test condition generation module 202, a hardware score generation module 203, a first performance score generation module 204, a second performance score generation module 205, and a comprehensive score generation module 206. DETAILED DESCRIPTION

[0019] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0020] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] like Figure 1-2 As shown, a method for testing the performance of a storage unit comprises the following steps: Step S101, obtaining hardware parameters of the hardware device where the solid state drive is located; Hardware parameters include: number of CPU cores, CPU main frequency, CPU cache size, memory size, memory frequency, power supply power and cooling type; Step S102, generating hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state hard disk is located; Hard disk performance test conditions include: test file size, queue depth and test duration; Step S103, generating a hardware score through a second neural network model according to the hardware parameters of the hardware device where the solid state drive is located; The hardware score ranges from 1 to 10. A larger hardware score indicates better performance of the hardware device where the SSD is located. Step S104, executing the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generating a first performance score according to the first test result; The first test results include: sequential read speed, random read speed, SSD temperature, and CPU temperature; A larger value of the first performance score indicates a better data reading performance of the solid state drive; Step S105, executing the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generating a second performance score according to the second test result; The second test results include: sequential write speed, sequential write amplification factor, random write speed, random write amplification factor, SSD temperature, and CPU temperature; A larger value of the second performance score indicates a better data writing performance of the solid state drive; Step S106: Generate a comprehensive score of the solid state drive according to the hardware score, the first performance score, and the second performance score.

[0022] It should be noted that the hard disk performance test conditions can also be executed in sequential read mode and sequential write mode to obtain a third test result, and a third performance score can be generated based on the third test result. At the same time, the hard disk performance test conditions can be executed in random read mode and random write mode to obtain a fourth test result, and a fourth performance score can be generated based on the fourth test result, thereby mixing different read and write modes to approximate the actual operating state of the solid-state drive. In addition, the concept of the present invention can also be used to complete the performance test of a mechanical hard disk. Although the working principles of the two are different, the factors affecting their performance are the same, such as CPU, memory, etc. Therefore, the present invention has a certain universality.

[0023] In one embodiment of the present invention, the hardware parameters are represented by eight-bit binary codes, wherein the heat dissipation types include: air cooling and liquid cooling. For example, the eight-bit binary code of the heat dissipation type of air cooling is 00000001.

[0024] It should be noted that each hardware parameter corresponds to a custom set, and eight-bit binary encoding is performed according to the element position of the corresponding custom set where the hardware parameter is located. For example, the number of CPU cores includes: 2 cores, 4 cores, 6 cores, 8 cores, 12 cores and 16 cores. The element position of the custom set where the CPU core number is 4 cores is 2, and the corresponding eight-bit binary encoding is 00000010. For example, the CPU main frequencies include 2.4 GHz, 3.5 GHz and 4.0 GHz. The element of the custom set where the CPU main frequency is 4.0 GHz is 3, and the corresponding eight-bit binary encoding is 00000011. The hardware parameters can also be represented by real number encoding. For example, the element position of the custom set where the CPU core number is 4 cores is 2, and the corresponding real number encoding is 1.2, which will not be elaborated here.

[0025] In one embodiment of the present invention, the first neural network model includes: a feature conversion layer, a feature extraction layer, a maximum pooling layer, a feature expansion layer and a first classifier; The feature conversion layer inputs hardware parameters and outputs a feature matrix; The calculation formula of the feature conversion layer is as follows: F = W × X, where F represents the feature matrix, X represents the hardware parameters, which are represented by a 1 × 7 vector, and W represents the weight parameters, which are represented by a 32 × 1 vector. The size of F is 32 × 7. The feature extraction layer includes a first convolution kernel and a second convolution kernel, which respectively extract features from the feature matrix to obtain a first feature map and a second feature map, wherein the sizes of the first convolution kernel and the second convolution kernel are 3×3 and 5×5 respectively, and the step sizes are both 1, so the sizes of the first feature map and the second feature map are 30×5 and 28×3 respectively; The first feature map and the second feature map are input into the maximum pooling layer, and the third feature map and the fourth feature map are output. The size of the pooling window of the maximum pooling layer is 3×3, and the step size is 1. Then, the sizes of the third feature map and the fourth feature map are 28×3 and 26×1 respectively. The feature expansion layer is used to expand the third feature map and the fourth feature map into vector representations and concatenate them to obtain a combined vector. The size of the combined vector is 110×1. The combined vector is input to the first classifier, and the classification space of the first classifier represents the hardware score.

[0026] It should be noted that the hardware parameters are represented by a 1×7 vector. Each dimension of the vector represents the number of CPU cores, CPU main frequency, CPU cache size, memory size, memory frequency, power supply power and cooling type. The first classifier is a softmax classifier, which is used to convert the combined vector into a probability distribution, that is, to convert each output element into a real number between 0 and 1, and to ensure that the sum of all these numbers is 1. For example, the hardware score is divided into 10 levels from 1 to 10, and the probability value of the hardware score is 9 is 90%. The sum of the probability values ​​of the remaining hardware scores is 20%. In this case, the hardware score with the largest probability value (9 points) is selected for output.

[0027] In one embodiment of the present invention, the second neural network model includes a first hidden layer, a second hidden layer, a second classifier, a third classifier, and a fourth classifier; The first hidden layer inputs hardware parameters and outputs a first update vector; The second hidden layer inputs the first update vector and outputs the second update vector; The second update vector is input into the second classifier, the third classifier and the fourth classifier respectively, and the classification spaces of the second classifier, the third classifier and the fourth classifier respectively represent the test file size, the queue depth and the test duration.

[0028] In one embodiment of the present invention, the calculation formula of the first hidden layer is as follows: ; in represents the first update vector of the first hidden layer output, X represents the hardware parameters of the first hidden layer input, represented by a 1×7 size vector, represents the first weight parameter, represents the second weight parameter, and Swish represents the Swish activation function.

[0029] For example, the first weight parameter may be designed as a matrix of size 7×16, and the size of the first update vector obtained by multiplying the vector of size 1×7 with the matrix is ​​1×16.

[0030] In one embodiment of the present invention, the calculation formula of the second hidden layer is as follows: ; in represents the second update vector for the output of the second hidden layer, represents the second weight parameter, Represents the second bias parameter.

[0031] Similarly, the second weight parameter may also be designed as a matrix of size 7×16, and the size of the second update vector obtained by multiplying the vector of size 1×7 with the matrix is ​​also 1×16.

[0032] It should be noted that queue depth refers to the ability of an SSD to process multiple input and output (I / O) requests simultaneously. Queue depth is used to assist in identifying performance bottlenecks of SSDs in high-concurrency environments.

[0033] In one embodiment of the present invention, sample labels of training samples used to train the first neural network model and the second neural network model are obtained by manual labeling.

[0034] It should be noted that during the training process of the first neural network model and the second neural network model, the gradient information of the weight parameters and bias parameters of the first neural network model and the second neural network model are calculated by the chain rule, and the weight parameters and bias parameters are updated by the gradient descent algorithm according to the gradient information.

[0035] In one embodiment of the present invention, the first performance score The calculation formula is as follows: ; in Indicates the sequential read speed, Indicates random read speed, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

[0036] It should be noted that the weight coefficients corresponding to the sequential read speed, random read speed, SSD temperature and CPU temperature are the reciprocals of their maximum values, which are used to control the size of the first performance score. For example, the maximum value of the random read speed is around 5000MB / s, and the corresponding weight coefficient is 1 / 5000. The maximum value of the random read speed is around 1400K IOPS, and the corresponding weight coefficient is 1 / 1400.

[0037] In one embodiment of the present invention, the second performance score The calculation formula is as follows: ; in Indicates the sequential write speed, Indicates random write speed, represents the sequential write amplification factor, represents the random write amplification factor, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

[0038] It should be noted that the write amplification factor represents the ratio of the amount of data actually written by the SSD to the amount of data requested to be written by the host system. The value range of the write amplification factor is between 0 and 1. The closer it is to 1, the better the write performance of the SSD. Similarly, the corresponding weight coefficient is the inverse of its maximum value, which will not be elaborated here.

[0039] In one embodiment of the present invention, the calculation formula of the comprehensive score Score is as follows: ; in and They represent the first performance score and the second performance score respectively, handware represents the hardware score, , and represent the first weight coefficient, the second weight coefficient and the third weight coefficient respectively, , and are custom parameters, and + + =1.

[0040] In one embodiment of the present invention, Figure 2 As shown, the present invention provides a storage unit performance testing device, comprising: A hardware parameter acquisition module 201 is used to acquire hardware parameters of the hardware device where the solid state drive is located; A hard disk performance test condition generating module 202, which is used to generate hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state disk is located; A hardware score generating module 203, which is used to generate a hardware score through a second neural network model according to hardware parameters of the hardware device where the solid state drive is located; A first performance score generating module 204, which is used to execute the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generate a first performance score according to the first test result; A second performance score generating module 205, which is used to execute the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generate a second performance score according to the second test result; The comprehensive score generating module 206 is used to generate a comprehensive score of the solid state drive according to the hardware score, the first performance score and the second performance score.

[0041] In one embodiment of the present invention, the present invention provides an electronic device for testing storage unit performance, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the above-mentioned method for testing storage unit performance is executed. In order to avoid repetition, it is not described here.

[0042] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A method for testing the performance of a storage unit, characterized in that: The following steps are involved: Step S101, obtaining hardware parameters of the hardware device where the solid state drive is located; Hardware parameters include: number of CPU cores, CPU main frequency, CPU cache size, memory size, memory frequency, power supply power and cooling type; Step S102, generating hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state hard disk is located; Hard disk performance test conditions include: test file size, queue depth and test duration; Step S103, generating a hardware score through a second neural network model according to the hardware parameters of the hardware device where the solid state drive is located; The hardware score ranges from 1 to 10. A larger hardware score indicates better performance of the hardware device where the SSD is located. Step S104, executing the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generating a first performance score according to the first test result; The first test results include: sequential read speed, random read speed, SSD temperature, and CPU temperature; A larger value of the first performance score indicates a better data reading performance of the solid state drive; Step S105, executing the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generating a second performance score according to the second test result; The second test results include: sequential write speed, sequential write amplification factor, random write speed, random write amplification factor, SSD temperature, and CPU temperature; A larger value of the second performance score indicates a better data writing performance of the solid state drive; Step S106: Generate a comprehensive score of the solid state drive according to the hardware score, the first performance score, and the second performance score.

2. A method for testing the performance of a storage unit according to claim 1, characterized in that: Hardware parameters are represented by eight-bit binary codes. Cooling types include air cooling and liquid cooling.

3. A method for testing the performance of a storage unit according to claim 1, characterized in that: The first neural network model includes: a feature conversion layer, a feature extraction layer, a maximum pooling layer, a feature expansion layer and a first classifier; The feature conversion layer inputs hardware parameters and outputs a feature matrix; The calculation formula of the feature conversion layer is as follows: F = W × X, where F represents the feature matrix, X represents the hardware parameters, which are represented by a 1 × 7 vector, and W represents the weight parameters, which are represented by a 32 × 1 vector. The size of F is 32 × 7. The feature extraction layer includes a first convolution kernel and a second convolution kernel, which respectively extract features from the feature matrix to obtain a first feature map and a second feature map, wherein the sizes of the first convolution kernel and the second convolution kernel are 3×3 and 5×5 respectively, and the step sizes are both 1, so the sizes of the first feature map and the second feature map are 30×5 and 28×3 respectively; The first feature map and the second feature map are input into the maximum pooling layer, and the third feature map and the fourth feature map are output. The size of the pooling window of the maximum pooling layer is 3×3, and the step size is 1. Then, the sizes of the third feature map and the fourth feature map are 28×3 and 26×1 respectively. The feature expansion layer is used to expand the third feature map and the fourth feature map into vector representations and concatenate them to obtain a combined vector. The size of the combined vector is 110×1. The combined vector is input to the first classifier, and the classification space of the first classifier represents the hardware score.

4. A method for testing the performance of a storage unit according to claim 1, characterized in that: The second neural network model includes a first hidden layer, a second hidden layer, a second classifier, a third classifier, and a fourth classifier; The first hidden layer inputs hardware parameters and outputs a first update vector; The calculation formula for the first hidden layer is as follows: ； in represents the first update vector of the first hidden layer output, X represents the hardware parameters of the first hidden layer input, represented by a 1×7 size vector, represents the first weight parameter, represents the second weight parameter, Swish represents the Swish activation function; The second hidden layer inputs the first update vector and outputs the second update vector; The calculation formula for the second hidden layer is as follows: ； in represents the second update vector for the output of the second hidden layer, represents the second weight parameter, represents the second bias parameter; The second update vector is input into the second classifier, the third classifier and the fourth classifier respectively, and the classification spaces of the second classifier, the third classifier and the fourth classifier respectively represent the test file size, the queue depth and the test duration.

5. A method for testing the performance of a storage unit according to claim 1, characterized in that: The sample labels of the training samples used to train the first neural network model and the second neural network model are obtained by manual labeling.

6. A method for testing the performance of a storage unit according to claim 1, characterized in that: First Performance Rating The calculation formula is as follows: ； in Indicates the sequential read speed, Indicates random read speed, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

7. A method for testing the performance of a storage unit according to claim 1, characterized in that: Second performance rating The calculation formula is as follows: ； in Indicates the sequential write speed, Indicates random write speed, represents the sequential write amplification factor, represents the random write amplification factor, Indicates the temperature of the solid state drive. Indicates the CPU temperature.

8. A method for testing the performance of a storage unit according to claim 1, characterized in that: The calculation formula of the comprehensive score is as follows: ； in and They represent the first performance score and the second performance score respectively, handware represents the hardware score, , and represent the first weight coefficient, the second weight coefficient and the third weight coefficient respectively, , and are custom parameters, and + + =1.

9. A device for testing the performance of a storage unit, characterized in that: The method for testing the performance of a storage unit according to any one of claims 1 to 8 comprises: A hardware parameter acquisition module is used to obtain the hardware parameters of the hardware device where the solid state drive is located; A hard disk performance test condition generation module, which is used to generate hard disk performance test conditions through a first neural network model according to hardware parameters of the hardware device where the solid state hard disk is located; A hardware score generating module, which is used to generate a hardware score through a second neural network model according to hardware parameters of the hardware device where the solid-state hard disk is located; A first performance score generating module, which is used to execute the hard disk performance test condition in the sequential read mode and the random read mode to obtain a first test result, and generate a first performance score according to the first test result; A second performance score generating module, which is used to execute the hard disk performance test condition in the sequential write mode and the random write mode to obtain a second test result, and generate a second performance score according to the second test result; The comprehensive score generating module is used to generate a comprehensive score of the solid state drive according to the hardware score, the first performance score and the second performance score.

10. An electronic device for testing the performance of a storage unit, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements a method for testing the performance of a storage unit as claimed in any one of claims 1 to 8.

Images