Method, system, device and storage medium for determining test parameters of a memory

Abstract

The present invention discloses a method, system, device and storage medium for determining test parameters of a memory. The method includes: obtaining performance requirement information of the memory; determining a first parameter type according to the performance requirement information and a preset performance parameter table; performing a first performance test on the memory based on the first parameter type to obtain a first test result; when the first test result indicates that the performance value of the memory is greater than or equal to the performance expected value, configuring the first parameter type as the test parameter of the memory; when the first test result indicates that the performance value of the memory is less than the performance expected value, correcting the first parameter type to obtain a second parameter type; and configuring the second parameter type as the test parameter of the memory. It is possible to determine the corresponding parameter type according to the performance requirement of the memory, so that the test meets the performance requirement and improves the test efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of memory testing, and particularly to a method, system, device and storage medium for determining test parameters of a memory. Background Art

[0002] Currently, when testing the produced memories, usually specific parameter types are tested.

[0003] However, when the performance requirements of the memories are different, the tests of specific parameter types often cannot meet the performance requirements, or there are unnecessary tests of parameter types, resulting in low test efficiency. Summary of the Invention

[0004] In view of this, the purpose of the embodiments of the present invention is to provide a method, system, device and storage medium for determining test parameters of a memory, which can determine the corresponding parameter types according to the performance requirements of the memory, so that the test meets the performance requirements and improves the test efficiency.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining test parameters of a memory, including:

[0006] Obtain the performance requirement information of the memory, and the performance requirement information represents the performance expectation value of the memory;

[0007] Determine a first parameter type according to the performance requirement information and a preset performance parameter table, and the first parameter type includes one or more parameters;

[0008] Perform a first performance test on the memory based on the first parameter type to obtain a first test result. The first performance test represents testing the performance value of the memory after adjusting the parameter values of the first parameter type, and comparing the performance value of the memory with the performance expectation value to obtain a first comparison result;

[0009] In the case where the first test result represents that the performance value of the memory is greater than or equal to the performance expectation value, configure the first parameter type as the test parameter of the memory;

[0010] In the case where the first test result represents that the performance value of the memory is less than the performance expectation value, correct the first parameter type to obtain a second parameter type. Among them, the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type represents that the performance value of the memory is greater than or equal to the performance expectation value;

[0011] Configure the second parameter type as the test parameter of the memory.

[0012] In some optional embodiments, determining the first parameter type according to the performance requirement information and the preset performance parameter table includes:

[0013] Obtaining a first parameter range in the preset performance parameter table according to the performance requirement information, where the first parameter range represents a parameter range associated with the performance requirement information;

[0014] Determining a first threshold according to the performance requirement information and the preset relevance table, where the first threshold represents a relevance threshold between a parameter and the performance requirement information;

[0015] Determining the first parameter type according to the first threshold and the first parameter range.

[0016] In some optional embodiments, obtaining the first parameter range in the preset performance parameter table according to the performance requirement information includes:

[0017] Obtaining the performance impact information of each parameter in the preset performance parameter table, where the performance impact information represents the change information of each performance index when the parameter changes;

[0018] Determining multiple parameters associated with the performance requirement information according to the performance requirement information and the performance impact information;

[0019] Configuring the multiple parameters associated with the performance requirement information as the first parameter range.

[0020] In some optional embodiments, determining the first parameter type according to the first threshold and the first parameter range includes:

[0021] Determining the relevance value between each parameter in the first parameter range and the preset performance index in the performance requirement information according to the performance impact information;

[0022] Respectively comparing the relevance value of each parameter in the first parameter range with the first threshold to obtain a second comparison result;

[0023] Configuring the parameter whose relevance value indicated by the second comparison result is greater than or equal to the first threshold as the first parameter type.

[0024] In some optional embodiments, determining the relevance value between each parameter in the first parameter range and the preset performance index in the performance requirement information according to the performance impact information includes:

[0025] Determining the change amplitude of the preset performance index when the M parameter unit in the first parameter range changes according to the performance impact information, where the M parameter represents any parameter in the first parameter range;

[0026] Determine the correlation value between the M parameter and the preset performance index according to the change range.

[0027] In some optional embodiments, the modification of the first parameter type to obtain the second parameter type includes:

[0028] Establish a Gaussian distribution of the correlation value, where the first coordinate value of the Gaussian distribution represents the correlation value, and the second coordinate value of the Gaussian distribution represents each parameter in the first parameter range;

[0029] Obtain the first distribution range of the first parameter type in the Gaussian distribution, and the first distribution range represents the first value interval of the second coordinate value;

[0030] Add a preset unit interval to the first value interval to obtain a second value interval;

[0031] Configure all parameters in the second value interval as the second parameter type.

[0032] In some optional embodiments, when the performance requirement information represents the performance expectation values of multiple preset performance indexes of the memory, the method further includes:

[0033] Obtain the performance difference between the performance expectation value of each preset performance index and the current performance value, where the current performance value represents the performance value of the preset performance index before the first performance test;

[0034] Obtain the performance impact information of the N parameter in the second parameter type, where the performance impact information represents the change information of each performance index when the N parameter changes, and the N parameter represents any parameter in the second parameter type;

[0035] Determine the correlation map between the N parameter and multiple preset performance indexes based on the performance impact information, and the correlation map is used to indicate the performance value change of multiple preset performance indexes when the N parameter changes by one unit;

[0036] Determine the parameter value of the N parameter according to the correlation map and the performance difference.

[0037] In a second aspect, an embodiment of the present invention provides a memory test system, including:

[0038] A first module, configured to obtain performance requirement information of a memory, where the performance requirement information represents the performance expectation value of the memory; [[ID=z9]]

[0039] A second module, configured to determine a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters;

[0040] A third module, configured to perform a first performance test on the memory based on the first parameter type to obtain a first test result, where the first performance test is to test the performance value of the memory after adjusting the parameter value of the first parameter type, and compare the performance value of the memory with a performance expected value to obtain a first comparison result;

[0041] A fourth module, configured to, when the first test result indicates that the performance value of the memory is greater than or equal to the performance expected value, configure the first parameter type as the test parameter of the memory;

[0042] A fifth module, configured to, when the first test result indicates that the performance value of the memory is less than the performance expected value, correct the first parameter type to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the performance expected value;

[0043] A sixth module, configured to configure the second parameter type as the test parameter of the memory.

[0044] In a third aspect, an embodiment of the present invention provides a memory testing device, which is applied to a smart card. The device includes:

[0045] At least one processor;

[0046] At least one memory, configured to store at least one program;

[0047] When the at least one program is executed by the at least one processor, the at least one processor implements the method as described above.

[0048] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a processor-executable program is stored, and the processor-executable program is used to execute the method as described above when executed by a processor.

[0049] The implementation of the embodiments of the present invention has the following beneficial effects: The embodiments of the present invention provide a method for determining test parameters of a memory, including: obtaining performance requirement information of the memory, where the performance requirement information represents the expected performance value of the memory; determining a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters; performing a first performance test on the memory based on the first parameter type to obtain a first test result, where the first performance test represents testing the performance value of the memory after adjusting the parameter values of the first parameter type, and comparing the performance value of the memory with the expected performance value to obtain a first comparison result; in the case where the first test result indicates that the performance value of the memory is greater than or equal to the expected performance value, configuring the first parameter type as the test parameter of the memory; in the case where the first test result indicates that the performance value of the memory is less than the expected performance value, correcting the first parameter type to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the expected performance value; configuring the second parameter type as the test parameter of the memory. This application can effectively compensate for the deficiencies of existing test methods, select different parameter types for performance testing through the performance requirement information of the memory, so as to determine the parameter type that meets the expected performance value. Therefore, it can determine the corresponding parameter type according to the performance requirement of the memory, make the test meet the performance requirement, and improve the test efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 FIG. is a schematic flowchart of the steps of a method for determining test parameters of a memory provided by an embodiment of the present invention;

[0051] Figure 2 FIG. is a structural block diagram of a memory test system provided by an embodiment of the present invention;

[0052] Figure 3 FIG. is a structural block diagram of a memory test device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0054] It should be noted that although functional modules are divided in the device schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the sequence in the flowchart. Terms such as "first" and "second" in the specification, claims, or the above-mentioned drawings are used to distinguish similar objects and do not necessarily describe a specific order or sequence.

[0055] An embodiment of the present invention provides a method for determining test parameters of a memory, including: obtaining performance requirement information of the memory, where the performance requirement information represents the expected performance value of the memory; determining a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters; performing a first performance test on the memory based on the first parameter type to obtain a first test result, where the first performance test represents testing the performance value of the memory after adjusting the parameter values of the first parameter type and comparing the performance value of the memory with the expected performance value to obtain a first comparison result; when the first test result indicates that the performance value of the memory is greater than or equal to the expected performance value, configuring the first parameter type as the test parameter of the memory; when the first test result indicates that the performance value of the memory is less than the expected performance value, correcting the first parameter type to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the expected performance value; configuring the second parameter type as the test parameter of the memory. This application can effectively compensate for the deficiencies of existing test methods, select different parameter types for performance testing through the performance requirement information of the memory, so as to determine the parameter type that meets the expected performance value. Therefore, it can determine the corresponding parameter type according to the performance requirement of the memory, making the test meet the performance requirement and improving the test efficiency.

[0056] The following further elaborates on the embodiments of the present invention with reference to the accompanying drawings.

[0057] As Figure 1 shown, an embodiment of the present invention provides a method for determining test parameters of a memory, and the steps included are as follows.

[0058] S100. Obtain performance requirement information of the memory, where the performance requirement information represents the expected performance value of the memory.

[0059] Specifically, the performance requirement information can be the performance requirements requested by the customer or the performance requirements determined according to the corresponding usage scenarios of the memory, and there is no specific limitation. For example, for the memory used in a smart phone, considering the user's frequent operations such as application startup, multitasking, and photo and video storage, there are specific requirements for aspects such as read and write speed, storage capacity, and power consumption. In terms of read and write speed, it is expected to respond quickly to ensure smooth system operation; the storage capacity should be large enough to accommodate various applications, photos, videos and other data; the power consumption should be as low as possible to extend the battery life of the mobile phone. In the data center scenario, the memory used by the server focuses more on performance expectations such as large-capacity storage, stable performance under high-concurrency read and write operations, and data reliability. These specific expectation indicators together constitute the performance requirement information of the memory, which clarifies the performance goals to be achieved when determining parameters subsequently.

[0060] S200. Determine the first parameter type according to the performance requirement information and the preset performance parameter table, where the first parameter type includes one or more parameters.

[0061] Specifically, the preset performance parameter table is an important reference material summarized after long-term memory research and development, production, and actual application feedback. It covers various parameters closely related to memory performance and the corresponding performance performance under different value combinations of these parameters. From the hardware level, it includes the process parameters of the storage chip (such as transistor size, gate oxide layer thickness, etc., parameters that affect the electrical and physical properties of the chip), the organizational structure parameters of the storage unit (the number of rows and columns of the storage array, the number of Banks (storage channels), etc., parameters that affect storage capacity and read and write methods); from the perspective of electrical performance, there are parameters such as working voltage, current driving ability, signal timing, etc.; there are also parameters such as cache policy and address mapping method at the software control level. Through a large number of experimental tests and data analysis in actual applications, the performance levels that the memory can generally achieve in terms of read and write speed, storage capacity, power consumption, data reliability, etc. under different parameter combinations are clarified, providing a basis for determining the appropriate parameter type according to performance requirements.

[0062] Based on the obtained performance requirement information, search for the parameter combinations in the preset performance parameter table that can meet the corresponding performance requirements. For example, if the performance requirement information emphasizes high read and write speed and large-capacity storage, then search in the parameter table for parameters related to improving read and write speed (such as optimizing signal timing, increasing the working voltage to speed up the transistor switching speed, etc.) and parameters for increasing storage capacity (such as increasing the scale of the storage array, expanding the number of Banks, etc.), and classify these found relevant parameters as the first parameter type. This type often includes multiple parameters, and they work together to affect whether the memory can achieve the expected performance.

[0063] In some optional embodiments, determining the first parameter type according to the performance requirement information and the preset performance parameter table includes: obtaining a first parameter range in the preset performance parameter table according to the performance requirement information, where the first parameter range represents a parameter range associated with the performance requirement information; determining a first threshold according to the performance requirement information and the preset relevance table, where the first threshold represents the relevance threshold between the parameter and the performance requirement information; and determining the first parameter type according to the first threshold and the first parameter range.

[0064] Specifically, based on the obtained performance requirement information, search in the preset performance parameter table for parameters that are relevant to the performance requirement information, that is, the performance expectation values represented by the performance requirement information can be adjusted through the associated parameters. For example, if the performance requirement information emphasizes achieving high read / write speeds and low power consumption, then search in the table for parameters related to improving read / write speeds and reducing power consumption. For read / write speeds, parameters related to optimizing signal timing may be found; these parameters associated with the performance requirement information thus constitute the first parameter range. Through such a search process, a parameter range that preliminarily meets the performance expectations can be screened out, preparing for further determining the specific parameter type.

[0065] The preset relevance table is mainly used to measure the degree of association between different parameters and the performance requirement information, and it is constructed through a large number of experimental analyses, statistical data, and expert knowledge in the professional field. The table records the weights or association degree values of the impacts of each parameter on different performance indicators (such as read / write speeds, storage capacities, power consumption, etc.), and these values are the manifestations of the relevance values. For example, for the parameter of working voltage, in terms of its impact on read / write speeds, the relevance value is set to 0.8 (the value here is only an example, and it is actually obtained based on specific quantitative analyses), indicating a strong association between the working voltage and read / write speeds; while for a secondary structural parameter inside a certain storage unit, its relevance value in terms of its impact on read / write speeds is 0.2, indicating a relatively weak degree of association. Combining the performance indicators that are focused on in the performance requirement information, search for the corresponding parameter relevance threshold in the preset relevance table. For example, if the performance requirement information focuses on read / write speeds, then search in the relevance table for the relevance thresholds corresponding to all parameters related to read / write speeds, such as setting the parameters in the first parameter range whose relevance values with respect to read / write speeds are greater than or equal to 0.4 (the first threshold) as the first parameter type. Among them, for those key parameters that have a greater impact on improving read / write speeds (such as the key time node parameters in signal timing, etc.), their corresponding relevance values are relatively high, meaning that small changes in these parameters may have a relatively significant impact on the desired performance indicator of read / write speeds; while for some auxiliary parameters that have a relatively small impact on read / write speeds, their relevance values are lower.

[0066] After obtaining the first parameter range and the first threshold, the parameters ultimately included in the first parameter type are determined based on the level of the correlation value and the first parameter range. Generally, parameters within the first parameter range and with a relatively high correlation value (greater than the first threshold) will be preferentially selected into the first parameter type. For example, for a signal timing parameter related to read / write speed, within the previously determined first parameter range, and the correlation value of this parameter with respect to read / write speed in the preset correlation table reaches 0.8 (greater than the first threshold of 0.4), indicating that this parameter plays a key role in meeting the read / write speed requirement in the performance requirement information, then it will be included in the first parameter type. On the contrary, for some parameters that are within the first parameter range but have a relatively low correlation value (less than the first threshold of 0.4), it is determined whether to include them according to the actual situation (such as overall parameter quantity control, the synergistic effect of other parameters, etc.). Through such a screening process, those parameters that have a greater impact on achieving the performance expectation and a strong correlation are selected from among the numerous parameter ranges that initially meet the performance requirements to form the first parameter type. The parameters in this type will be the key objects for subsequent performance testing and parameter adjustment, with the expectation of adjusting the values of these parameters to make the performance of the memory reach or approach the expected value set in the performance requirement information.

[0067] In some optional embodiments, the obtaining of the first parameter range from the preset performance parameter table according to the performance requirement information includes: obtaining the performance impact information of each parameter in the preset performance parameter table, where the performance impact information represents the change information of each performance index when the parameter changes; determining multiple parameters associated with the performance requirement information according to the performance requirement information and the performance impact information; and configuring the multiple parameters associated with the performance requirement information as the first parameter range.

[0068] Specifically, the performance impact information reflects the change situation of each performance index when the parameter changes. Taking the parameter of operating voltage as an example, its performance impact information will record how performance indexes such as the read / write speed, power consumption, and data stability of the memory change correspondingly when the operating voltage increases or decreases. For example, as the operating voltage increases, the read / write speed will increase because the switching speed of the transistor will increase with the increase in voltage, enabling data read / write operations to be completed more quickly; but at the same time, the power consumption will also increase, and too high voltage may also affect the data stability, resulting in an increase in the error probability. For storage cell structure parameters, such as when the number of rows and columns of the storage array changes, the storage capacity will change correspondingly (usually an increase in the number of rows and columns will increase the storage capacity), but the read / write speed may be affected to a certain extent due to the complexity of operations such as address decoding. These specific change information is the manifestation of the performance impact information.

[0069] Generally, performance requirement information involves multiple performance metrics. Therefore, it is necessary to comprehensively consider various influences to determine the associated parameters. For example, in the case of requiring both high read / write speed and low power consumption, not only the parameters that can improve the read / write speed (such as the signal timing parameters mentioned above) need to be found, but also the parameters that will not excessively increase power consumption while adjusting the read / write speed and even help reduce power consumption need to be considered (for example, adopting a suitable caching strategy can not only speed up the read / write speed but also reduce the power consumption increase caused by frequent access to storage units. Therefore, the caching strategy parameters will also be determined as the associated parameters). Through such a comprehensive and all-round analysis, multiple associated parameters that meet different performance expectation requirements are determined. The multiple parameters associated with the performance requirement information that are screened out are configured as the first parameter range. Here, the "range" emphasizes the range composed of multiple parameters, that is, the first parameter range is composed of multiple parameters.

[0070] In some optional embodiments, determining the first parameter type according to the first threshold and the first parameter range includes: determining the relevance value between each parameter in the first parameter range and the preset performance metrics in the performance requirement information according to the performance impact information; respectively comparing the relevance value of each parameter in the first parameter range with the first threshold to obtain a second comparison result; configuring the parameters whose relevance value indicated by the second comparison result is greater than or equal to the first threshold as the first parameter type.

[0071] Specifically, for each parameter within the first parameter range, its relevance value to the preset performance indicators in the performance requirement information is determined based on the performance impact information of each parameter. The relevance value is a quantitative representation of the degree to which a parameter affects the preset performance indicators. For example, if the preset performance indicators in the performance requirement information include three items: read / write speed, power consumption, and storage capacity, for the parameter of working voltage within the first parameter range, by checking the performance impact information, it can be seen that when the working voltage changes within a certain range, the read / write speed may show a relatively obvious linear change (for example, when the voltage increases by 0.1V, the read / write speed increases by 20%), the impact on power consumption is also relatively significant (the increase in voltage will cause the power consumption to increase by a certain proportion), and there is basically no direct impact on the storage capacity. At this time, a quantitative method can be used to determine the relevance value. For example, a score system with a full score of 1 is set, and scores are given according to the importance degree and change range of the impact of this parameter on each preset performance indicator. Suppose after comprehensive evaluation, the relevance value of the working voltage to the read / write speed is scored 0.4 (indicating a relatively high degree of influence), the relevance value to the power consumption is scored 0.3 (the influence is also relatively obvious), and the relevance value to the storage capacity is scored 0 (no direct impact). These relevance values for different preset performance indicators together constitute the relevance value situation of the parameter of working voltage to the entire performance requirement information. In the same way, the relevance values of other parameters within the first parameter range (such as signal timing, cache strategy, etc.) to the preset performance indicators are also determined respectively, which will not be elaborated here.

[0072] The first threshold is previously determined according to the performance requirement information and the preset relevance table. It represents the relevance threshold between the parameter and the performance requirement information, and is a boundary threshold for distinguishing the importance degree of the parameter to meeting the performance requirement information. By comparing the relevance values of each parameter within the first parameter range with the first threshold, those parameters that truly play a key role and have a relatively strong relevance to meeting the performance requirement information can be screened out, providing a basis for finally determining the first parameter type. For example, assume that the relevance value of a certain parameter is lower than the first threshold, which means that although this parameter is within the first parameter range (that is, initially judged to be possibly related to the performance requirement), its actual impact on the preset performance indicators in the performance requirement information is relatively small, and it may not need to be considered as a key point when determining the parameter type later; while for a parameter whose relevance value is greater than or equal to the first threshold, it means that it plays a more important role in achieving the performance expectation.

[0073] Continuing with the example of the working voltage parameters mentioned above, assume that the determined first threshold is 0.3 for the preset performance metric of read / write speed (which means that parameters with a correlation value of 0.3 or above have a relatively important impact on read / write speed). The previously determined correlation value of the working voltage with respect to read / write speed is 0.4. Thus, the correlation value of the working voltage parameter in terms of read / write speed is greater than the first threshold. Looking at the preset performance metric of power consumption, if the first threshold is set to 0.2 and the correlation value of the working voltage with respect to power consumption is 0.3, it is also greater than the first threshold. According to this comparison method, for all parameters within the first parameter range, their correlation values are compared with the corresponding first threshold for each preset performance metric, obtaining the results of the comparison of each parameter with the first threshold in different performance metrics. Combining these results forms the second comparison result.

[0074] Based on the second comparison result, parameters with a correlation value greater than or equal to the first threshold are configured as the first parameter type. Parameters with a correlation value greater than or equal to the first threshold have a sufficiently important impact on meeting the preset performance metrics in the performance requirement information. They are the key parameters most likely to enable the memory to reach the desired performance level by adjusting their values. For example, if the correlation value of the signal timing parameter in terms of read / write speed is greater than the first threshold, and the correlation values of the cache policy parameter in both reducing power consumption and improving read / write speed are greater than or equal to the first threshold, then eligible parameters such as the signal timing parameter and the cache policy parameter will be included in the first parameter type. Subsequently, during the memory performance test, the values of the parameters in these first parameter types will be adjusted and tested, observing the actual performance changes of the memory, so as to find the optimal combination of parameter values, making the performance of the memory as close as possible to the expected requirements set by the performance requirement information, and ensuring the efficient and stable operation of the memory in the actual application scenario.

[0075] In some optional embodiments, determining the correlation value between each parameter in the first parameter range and the preset performance metric in the performance requirement information according to the performance impact information includes: determining the change amplitude of the preset performance metric when the M parameter in the first parameter range changes by one unit according to the performance impact information, where the M parameter represents any parameter in the first parameter range; and determining the correlation value between the M parameter and the preset performance metric according to the change amplitude.

[0076] Specifically, the performance impact information records the impact of each parameter on different performance metrics of the memory under different value changes. For each parameter in the first parameter range (represented by the M parameter here), the specific change situation of the preset performance metric when this parameter changes by one unit is extracted.

[0077] Example of the method for determining the change range: Taking the common parameter of working voltage as an example, assume it is within the first parameter range. From the performance impact information, check the impact on the preset performance indicator of read / write speed when the working voltage increases by 1V (here, 1V is the unit change). Suppose through a large number of experimental tests or summary of past application experience, it is known that when the working voltage increases by 1V, the read / write speed correspondingly increases by 100MB / s (only an example value). Then this "100MB / s" is the change range of the preset performance indicator of read / write speed when the working voltage, as the M parameter, has a unit change. Another example is for the parameter of the number of rows and columns in a storage array (which can also be regarded as an M parameter). When the number of rows and columns increases by 1 row (unit change), the preset performance indicator of storage capacity may increase by 1GB (assumed value). Then "1GB" is the change range of storage capacity when the parameter of the number of rows and columns has a unit change. For different M parameters and their corresponding different preset performance indicators, the specific change range values under this unit change can be determined by referring to the experimental data, simulation analysis results, or actual application feedback recorded in the performance impact information. The specific determination method is not limited here.

[0078] The correlation value aims to quantify the influence degree of the M parameter on the preset performance indicator, and its numerical value reflects the importance of this parameter in influencing the corresponding performance indicator. And the change range is the key basis for determining the correlation value. Generally, the larger the change range, the more significant the influence of the parameter on the corresponding performance indicator, and the higher the correlation value.

[0079] A relative quantitative scoring system can be adopted to determine the correlation value. For example, a score ranging from 0 to 1 is set to represent the strength of the correlation. Taking the preset performance indicator of read / write speed as an example, if a reference standard is set: when the change range of read / write speed exceeds a certain established large value (such as 500MB / s, only for illustration) when a certain M parameter has a unit change, it is considered that the correlation between this parameter and read / write speed is very strong, and a correlation value of 0.8 can be assigned; if the change range is at a medium level (such as 100MB / s to 300MB / s), the correlation value is set to 0.5; if the change range is small (less than 100MB / s), the correlation value is set to 0.2, etc.

[0080] For example, as previously mentioned, for every 1V increase in the working voltage, the read / write speed increases by 100MB / s. According to the above-set scoring rules, the correlation value between the M parameter of the working voltage and the preset performance index of the read / write speed can be determined to be 0.5. For other M parameters and their corresponding different preset performance indicators, their correlation values are determined based on their respective change ranges and by referring to similar quantitative scoring rules, so as to clearly reflect the differences in the influence degrees of each parameter on different performance indicators and provide strong data support for subsequent steps such as screening key parameters and determining parameter types.

[0081] S300. Perform a first performance test on the memory based on the first parameter type to obtain a first test result. The first performance test characterizes the performance value of the memory after adjusting the parameter value of the first parameter type and compares the performance value of the memory with the performance expectation value to obtain a first comparison result.

[0082] Specifically, for each parameter in the first parameter type, its value needs to be adjusted within a reasonable range. This reasonable range is usually determined based on relevant technical specifications, past experience, and theoretical analysis. Taking the working voltage as an example, if it is in the first parameter type and the normal range of the working voltage of this memory is known to be between 1.0V and 1.8V, then different values within this range are selected for experimentation. For example, starting from a lower value, the voltage is gradually increased in a certain step (such as 0.1V) to observe the performance changes of the memory at different voltage values. The same operation method is applicable to other parameters. For example, for the cache policy parameter, different cache levels or cache sizes can be set, and different cache levels are switched or the values of the cache size are changed in turn, and comprehensive experimentation is carried out within its allowed effective setting range to comprehensively examine the performance of the memory under different parameter value combinations.

[0083] In the process of adjusting the parameter values of the first parameter type, it is necessary to test multiple key performance indicators of the memory. Common performance indicators include read / write speed, storage capacity, power consumption, and data stability (such as the error rate of read / write data, etc.). For the test of read / write speed, a large number of data blocks will be written into the memory through a dedicated test tool, and then these data will be read out according to certain rules, and the time spent to complete the read / write operation will be recorded, so as to calculate the specific value of the read / write speed, such as how many read / write operations can be completed per second or the size of the data read / written within a unit time, etc.; for the storage capacity, the data will be gradually filled until the memory reaches the full-load state or a prompt indicating that it cannot continue to store appears, so as to accurately measure the actual amount of data that can be stably stored; the power consumption test uses a professional power consumption measurement instrument to record the power consumption of the memory when it is in different working states (such as standby, read / write operation, etc.); the data stability test can be carried out by repeating the read / write operation multiple times, comparing whether the written and read data are consistent, and counting the proportion of incorrect data, so as to measure the data stability.

[0084] Compare each performance value of the memory obtained through the above tests with the performance expectation values set in the initially obtained performance requirement information one by one. For example, if the performance expectation value requires the read / write speed to reach 1000 MB / s, and the actually measured read / write speed is 800 MB / s, it means that under the current parameter value setting, the performance indicator of read / write speed does not meet the expectation value; another example is that the power consumption set in the performance expectation value cannot exceed 1 W under a specific working state, and the actually measured power consumption is 0.8 W, which indicates that the power consumption indicator meets the expectation value requirement. Through such a comprehensive and detailed comparison, a comprehensive first comparison result is formed, which clearly reflects whether the current parameter adjustment based on the first parameter type can make the overall performance of the memory reach the expected level. If each performance indicator is greater than or equal to the corresponding performance expectation value, then the first comparison result is that the performance of the memory meets the expectation; on the contrary, if there are some performance indicators lower than the expectation value, it means that the parameter type needs to be further adjusted and optimized, hoping that through subsequent improvement measures, the performance of the memory can better meet the goals set by the performance requirement information.

[0085] S400. When the first test result indicates that the performance value of the memory is greater than or equal to the performance expectation value, configure the first parameter type as the test parameter of the memory.

[0086] Specifically, when the first test result indicates that the performance value of the memory is greater than or equal to the expected performance value, it means that the parameter value combination set based on the first parameter type can enable the memory to reach the expected performance level. Configuring the first parameter type as the test parameter of the memory provides a reliable reference standard for subsequent production, quality inspection, and actual application. During the mass production of the memory, configuring and detecting according to these parameters can ensure that each produced memory product can stably reach the expected performance, guarantee the consistency of product performance, and contribute to improving the overall quality and market competitiveness of the product.

[0087] S500. In the case where the first test result indicates that the performance value of the memory is less than the expected performance value, the first parameter type is corrected to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not equal to the empty set, and the second test result of the second performance test of the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the expected performance value.

[0088] Specifically, when the first test result shows that the performance value of the memory is less than the expected performance value, it means that the parameter value combination corresponding to the first parameter type cannot enable the memory to reach the expected performance level and there is a performance bottleneck, so it needs to be corrected. This is to further determine a more appropriate parameter setting to meet the performance requirements of the memory in terms of read and write speed, storage capacity, power consumption, etc. in the actual application scenario, and ensure that the memory can operate stably and efficiently in the corresponding system.

[0089] First, it is necessary to analyze which performance indicators in the first test result do not reach the expected value and the magnitude of the gap from the expected value, and find the reasons by combining the working principle of the memory and the mutual influence relationship between parameters. For example, if the read and write speed is much lower than the expected value, it may be that the signal timing parameters involved in the first parameter type are not set optimally, resulting in a large delay in data transmission; or the organizational structure of the storage unit limits the efficiency of parallel read and write operations; it is also possible that the cache strategy fails to play its full role, causing frequent access to the storage unit and reducing the overall read and write speed. For power consumption, if the actual power consumption is higher than the expected value, it may be that the working voltage is set too high, or some circuit modules are in an unnecessary high-energy consumption state under the current parameter configuration. Through such a comprehensive and in-depth analysis, the key factors affecting performance are identified, providing a targeted direction for correcting the parameter type.

[0090] The first parameter type is corrected to obtain a second parameter type. The main correction methods include:

[0091] Adjust the value range or adjustment method of existing parameters: For example, for signal timing parameters, if the previous value range was narrow or the adjustment step size was inappropriate, resulting in the failure to find the optimal timing configuration, then in the second parameter type, the value range can be expanded and the adjustment step size can be refined to more precisely explore the appropriate signal timing; for the working voltage parameter, if it is found that the current voltage setting is not conducive to the balance of power consumption and read / write speed, then the value can be reselected and tried within a more reasonable voltage range. For example, it can be further reduced from the original relatively wide voltage range to a sub-range that is more likely to meet the performance requirements based on experience or theoretical analysis for value adjustment.

[0092] Introduce new parameters: If analysis shows that certain factors not included in the first parameter type limit the memory performance, relevant parameters need to be introduced into the second parameter type. For example, if it is suspected that the internal error correction mechanism of the memory cell (such as ECC, error checking and correction) affects the read / write speed and data stability, but this parameter was not considered before, then at this time, the parameters related to ECC (such as the encoding method of the error correction code, the length of the parity bit, etc.) can be included in the second parameter type. By adjusting the values of these newly introduced parameters and their combined effect with the original parameters, the performance of the memory can be improved.

[0093] The process of conducting the second performance test on the memory based on the second parameter type is similar to the first performance test. Similarly, for each parameter included in the second parameter type, it needs to be adjusted within a reasonable value range, and the key performance indicators of the memory (such as read / write speed, storage capacity, power consumption, data stability, etc.) need to be tested. For example, for the newly introduced ECC-related parameters, different combinations of the encoding method of the error correction code and the length of the parity bit will be set. At the same time, in combination with the adjusted signal timing, working voltage and other parameters, data will be written to and read from the memory according to a certain test process, and the actual performance values of the memory under these different parameter value combinations will be measured and recorded. When the performance value does not meet the performance expectation value, repeat the above method of introducing new parameters to correct the first parameter type and continue the second performance test until the corrected first parameter type can meet the performance expectation value, then configure the corrected first parameter type as the second parameter type.

[0094] When the second test result indicates that the performance value of the memory is greater than or equal to the performance expectation value, it means that the memory can reach the expected performance level after correcting the parameter type. After configuring the second parameter type as the test parameters of the memory, it can ensure the consistency of product quality during the production process, and ensure that the memory can operate stably according to the expected performance in system integration applications, enabling it to better meet the requirements of actual applications for memory performance and improving the performance and reliability of the entire system.

[0095] In some optional embodiments, the modification of the first parameter type to obtain the second parameter type includes: establishing a Gaussian distribution of the correlation value, where the first coordinate value of the Gaussian distribution represents the correlation value, and the second coordinate value of the Gaussian distribution represents each parameter in the first parameter range; obtaining a first distribution range of the first parameter type in the Gaussian distribution, where the first distribution range represents a first value interval of the second coordinate value; adding a preset unit interval to the first value interval to obtain a second value interval; and configuring all parameters in the second value interval as the second parameter type.

[0096] Specifically, the Gaussian distribution, also known as the normal distribution, is a common probability distribution model, characterized by being high in the middle, low on both sides, and symmetric left and right. Introducing the Gaussian distribution of the correlation value here is to analyze the distribution of the correlation values between each parameter in the first parameter range and the performance requirement information from the perspective of probability statistics, so as to more scientifically and reasonably screen and adjust the parameters, and find parameter combinations that may be more helpful for improving the memory performance.

[0097] In this Gaussian distribution, the first coordinate value represents the correlation value, which reflects the quantitative value of the influence degree of each parameter on the preset performance index in the performance requirement information, and its range is usually between 0 and 1 (specifically determined according to the set correlation value quantization rule, which is not limited here). The second coordinate value represents each parameter in the first parameter range, that is, each parameter participating in the analysis is taken as a variable on the horizontal axis of this distribution. By collecting and organizing the correlation values corresponding to each parameter, this distribution is constructed according to the statistical law of the Gaussian distribution, so that the parameters with higher correlation values are in relatively closer positions to the center and have greater probability density in the distribution, while the parameters with lower correlation values are distributed on both sides with relatively smaller probability density.

[0098] For example, assume that there are multiple parameters such as working voltage, signal timing, and cache policy in the first parameter range. According to the correlation values determined by their respective relationships with preset performance indicators such as read / write speed and power consumption, these parameters and their correlation values are correspondingly plotted in the Gaussian distribution. If the working voltage has a relatively high comprehensive correlation value with the read / write speed and power consumption, then the corresponding point in this distribution will be closer to the center position, reflecting its relatively more important and greater impact on meeting the performance requirements.

[0099] The first parameter type is a set of parameters that have been determined through a series of screenings before and have an important impact on meeting performance requirements. In the constructed Gaussian distribution of correlation values, determine the distribution range where these parameters are located, that is, the first distribution range, which characterizes the first value range of the second coordinate values (i.e., each parameter). This value range is determined by observing the positions of the parameters in the first parameter type in the Gaussian distribution, and usually covers the ranges where the correlation values are relatively high, in the central region of the Gaussian distribution, or meet the previously set conditions of relatively strong correlation. For example, through analysis, it is found that the working voltage, signal timing, etc. parameters in the first parameter type correspond to an abscissa range in the Gaussian distribution of correlation values that is roughly from parameter A to parameter B (where parameter A and B are specific parameters arranged in order, just examples), and this range is the first distribution range. It represents the distribution interval corresponding to the set of parameters initially determined to have a greater impact on performance, reflecting the concentration degree and range of these parameters in terms of correlation values, and providing a basic reference for further adjusting the parameter range later.

[0100] Since the first test result shows that the performance value of the memory is less than the performance expectation value, it indicates that the parameter range covered by the current first parameter type is not comprehensive or accurate enough, and the best parameter combination has not been found to meet the performance requirements. Increase the first value range by a preset unit range. The purpose is to expand the parameter search range, try to include some parameters with slightly lower correlation values originally but may be helpful for improving performance in a more extensive exploration, and provide more possibilities for finding a suitable parameter combination.

[0101] The setting of the preset unit range needs to consider various factors comprehensively, such as the total number of parameters, the complexity of the correlation between parameters, and past experience. If the preset unit range is set too large, it may introduce too many parameters with weak correlations and not very critical for performance improvement, increasing the complexity and workload of the test; while if it is set too small, it may not effectively expand to a more suitable parameter range, and it is still difficult to find a parameter combination that meets the performance requirements. For example, according to the overall distribution of parameters, the preset unit range can be set as a range containing several adjacent parameters, and the specific number is determined according to actual analysis and experience.

[0102] An example of obtaining the second value range: Suppose the first value range is from parameter A to parameter B, and the preset unit range is set as a range containing 3 adjacent parameters. Then, after expanding the first value range by a range containing 3 adjacent parameters on both sides, the new range obtained is the second value range. For example, from parameter C to parameter D (where C is before A and D is after B, and the corresponding expanded parameter ranges are included), and this second value range covers a more extensive parameter range than the first value range.

[0103] Configuring all parameters within the second value range to the second parameter type is based on the idea of expanding the parameter search scope, considering as comprehensively as possible those parameters that may play a role in improving the memory performance. Although these newly included parameters may have relatively low correlation values in the initial evaluation, through further performance tests after expanding the scope, it is found that under different combinations of parameter values or in cooperation with other parameters, they can make the memory performance reach or exceed the performance expectation value. After being configured as the second parameter type, subsequent second performance tests will be carried out based on these parameters. By adjusting their values and observing the actual performance changes of the memory, it is verified whether this expansion of the parameter scope is effective and whether a suitable parameter combination that meets the performance requirements can be found, thus promoting the process of optimizing the entire memory performance and enabling it to finally reach the expected performance level in actual applications.

[0104] In some optional embodiments, when the performance requirement information represents the performance expectation values of multiple preset performance indicators of the memory, the method further includes: obtaining the performance difference between the performance expectation value of each preset performance indicator and the current performance value, where the current performance value represents the performance value of the preset performance indicator before the first performance test; obtaining the performance impact information of the N parameters in the second parameter type, where the performance impact information represents the change information of each performance indicator when the N parameters change, and the N parameters represent any one parameter in the second parameter type; determining the association map between the N parameters and multiple preset performance indicators based on the performance impact information, where the association map is used to indicate the performance value changes of multiple preset performance indicators when the N parameters change by one unit; and determining the parameter values of the N parameters according to the association map and the performance difference.

[0105] Specifically, the performance difference is a key indicator for measuring how much each preset performance indicator is short of the expected performance level. By comparing the performance expectation value with the current performance value before the first performance test, the improvement or reduction space in each performance aspect can be clarified. For example, for the preset performance indicator of read / write speed, if the performance expectation value is 1000 MB / s and the current performance value is 600 MB / s, then the performance difference is 400 MB / s, which indicates that in terms of read / write speed, a certain increase needs to be achieved by adjusting the parameters to meet the expectation.

[0106] These performance differences are obtained by calculating the performance expectation values recorded in the performance requirement information and the known current performance values (which can be obtained through preliminary tests or based on the existing specifications of the memory, etc.). Such calculations are carried out for each preset performance indicator to obtain a complete set of performance differences, which can comprehensively reflect the improvement requirements of the memory in various performance aspects and provide a basis for subsequent targeted parameter adjustment.

[0107] The performance impact information has been mentioned above, which details the changes in each performance metric when the parameters change. For any N parameter in the second parameter type, obtaining its performance impact information is to understand the specific impact manner and degree of this parameter on all preset performance metrics, so as to construct an association graph and determine appropriate parameter values subsequently.

[0108] Example of acquisition method: Assume that the N parameter is the working voltage, and obtain its performance impact information from the previously established performance parameter table or through experimental test data. For example, when the working voltage increases, the read / write speed will increase accordingly, but the power consumption will also increase, and the storage capacity is not affected or only slightly changed (these change situations may be reflected by specific numerical relationships, such as when the working voltage increases by 0.1V, the read / write speed increases by 100MB / s, and the power consumption increases by 0.1W). The acquisition of this information helps to comprehensively understand the actual impact of the N parameter on each preset performance metric under different value changes.

[0109] The association graph visually represents the mutual relationship between the N parameter and multiple preset performance metrics. Through the association graph, it is possible to show the changes in the performance values of each preset performance metric when the N parameter changes by one unit, obtain the comprehensive impact of the N parameter on different performance aspects, and then determine appropriate parameter values considering the balanced improvement of multiple performance metrics.

[0110] Example of construction process: Taking the N parameter as the working voltage as an example, assume a simple association graph is constructed. In the graph, the abscissa represents the unit change of the working voltage (such as taking 0.1V as one unit), and the ordinate corresponds to different preset performance metrics (such as read / write speed, power consumption, storage capacity). According to the obtained performance impact information, when the working voltage starts to change from an initial value, draw a curve in the graph showing that the read / write speed rises as the voltage increases (such as increasing by 100MB / s per 0.1V as mentioned before), and at the same time draw a curve showing the increase in power consumption (increasing by 0.1W per 0.1V) and a straight line showing that the storage capacity remains basically unchanged (assuming the storage capacity is not affected by the working voltage). In this way, an association graph between the working voltage and multiple preset performance metrics is constructed, and through this graph, the comprehensive impact of the working voltage change on each performance metric can be visually seen.

[0111] Using the association graph and performance difference to determine the parameter value of the N parameter is a process that comprehensively considers the balanced improvement of multiple performance metrics. First, determine the amplitude direction of improvement or adjustment required for each preset performance metric according to the performance difference (such as the read / write speed needs to be improved, the power consumption needs to be reduced, etc.). Then, find the value range or specific value of the N parameter in the association graph that can simultaneously meet these performance adjustment requirements.

[0112] Determination process example: Assume that according to the associated graph, it is known that when the unit of parameter A increases, performance index a increases by 20%, performance index b increases by 10%, and performance index c decreases by 10%; when the unit of parameter B increases, performance index a increases by 10%, performance index b increases by 20%, and performance index c increases by 10%; when the unit of parameter C increases, performance index a increases by 10%, performance index b increases by 10%, and performance index c increases by 20%. The current performance value of performance index a is 50% and the expected performance value is 80%; the current performance value of performance index b is 60% and the expected performance value is 80%; the current performance value of performance index c is 60% and the expected performance value is 90%. Then the performance difference of performance index a is 30%, the performance difference of performance index b is 20%, and the performance difference of performance index c is 30%. Then according to the calculation formula AX + BY + CZ = (performance index a ≥ 80%, performance index b ≥ 80%, performance index c ≥ 90%), the corresponding values of X, Y, and Z are determined, that is, AX is used as the parameter value of parameter A, BX is used as the parameter value of parameter B, and CX is used as the parameter value of parameter C.

[0113] S600. Configure the second parameter type as the test parameter of the memory.

[0114] Specifically, the second parameter type was obtained by correcting the first parameter type before, and the second performance test results based on the second parameter type indicate that the performance value of the memory can be greater than or equal to the expected performance value. This means that the parameters and their value combinations included in the second parameter type have been verified to be effective configurations that can enable the memory to reach the expected performance level. Therefore, it is configured as a test parameter to provide test parameters for the subsequent testing of other memories of the same type.

[0115] In multiple links such as the production, quality inspection, and actual application integration of the memory, it is crucial to have clear and effective test parameters. Using the second parameter type as the test parameter can ensure that when mass-producing memories, each product is set with parameters according to this standard configuration, thereby guaranteeing the consistency of product performance and reducing quality problems caused by performance fluctuations.

[0116] Record each parameter involved in the second parameter type and its corresponding values in detail and accurately to form a standardized documentation. For example, if the second parameter type includes working voltage, signal timing, cache policy, and parameters related to ECC (error checking and correction), etc., then the specific value of the working voltage (such as 1.5V), the exact duration of each key time node of the signal timing (such as the rising edge time of the read / write signal is 1ns, etc.), the specific mode of the cache policy (such as adopting a certain specific cache algorithm and cache size setting), and the specific configuration of the ECC parameters (such as the encoding method of the error correction code, the length of the parity bit, etc.) should be recorded respectively. These recorded documents will serve as the official reference for the memory test parameters, facilitating subsequent query, use, and sharing.

[0117] Implementing the embodiments of the present invention includes the following beneficial effects: The embodiments of the present invention provide a method for determining test parameters of a memory, including: obtaining performance requirement information of the memory, where the performance requirement information represents the expected performance value of the memory; determining a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters; performing a first performance test on the memory based on the first parameter type to obtain a first test result, where the first performance test represents testing the performance value of the memory after adjusting the parameter values of the first parameter type and comparing the performance value of the memory with the expected performance value to obtain a first comparison result; in the case where the first test result represents that the performance value of the memory is greater than or equal to the expected performance value, configuring the first parameter type as the test parameter of the memory; in the case where the first test result represents that the performance value of the memory is less than the expected performance value, correcting the first parameter type to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type represents that the performance value of the memory is greater than or equal to the expected performance value; configuring the second parameter type as the test parameter of the memory. This application can effectively compensate for the deficiencies of existing test methods, select different parameter types for performance testing through the performance requirement information of the memory, so as to determine the parameter type that meets the expected performance value. Therefore, parameter types under different performance requirements can be obtained, and the corresponding parameter type can be determined according to the performance requirement of the memory, making the test meet the performance requirement and improving the test efficiency.

[0118] In a second aspect, referring to Figure 2 , the embodiments of the present invention provide a memory test system, including:

[0119] A first module, configured to obtain performance requirement information of the memory, where the performance requirement information represents the expected performance value of the memory;

[0120] A second module, configured to determine a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters;

[0121] A third module, configured to perform a first performance test on the memory based on the first parameter type to obtain a first test result, where the first performance test is to test the performance value of the memory after adjusting the parameter value of the first parameter type, and compare the performance value of the memory with a performance expected value to obtain a first comparison result;

[0122] A fourth module, configured to, when the first test result indicates that the performance value of the memory is greater than or equal to the performance expected value, configure the first parameter type as the test parameter of the memory;

[0123] A fifth module, configured to, when the first test result indicates that the performance value of the memory is less than the performance expected value, correct the first parameter type to obtain a second parameter type, where the difference set between the second parameter type and the first parameter type is not an empty set, and the second test result of performing a second performance test on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the performance expected value;

[0124] A sixth module, configured to configure the second parameter type as the test parameter of the memory.

[0125] It can be seen that the content in the above method embodiments is applicable to the system embodiments. The functions specifically implemented by the system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0126] In a third aspect, referring to Figure 3 , an embodiment of the present invention provides a memory test device, including:

[0127] At least one processor;

[0128] At least one memory, configured to store at least one program;

[0129] When at least one program is executed by at least one processor, at least one processor implements the method as described above.

[0130] It can be seen that the content in the above method embodiments is applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0131] Fourthly, in addition, the embodiments of the present application also disclose a computer program product or a computer program. The computer program product or the computer program is stored in a computer-readable storage medium. The processor of the computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the computer device executes the above-mentioned method or the above-mentioned system. Similarly, the content in the above-mentioned method embodiments is applicable to the present storage medium embodiment. The functions specifically implemented by the present storage medium embodiment are the same as those of the above-mentioned method embodiments, and the beneficial effects achieved are also the same as those of the above-mentioned method embodiments.

[0132] It can be understood that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and their appropriate combinations. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital information processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cassette, tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0133] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those of ordinary skill in the art, various changes can be made without departing from the gist of the present invention.

Claims

1. A method for determining test parameters of a memory, characterized in that: include: Acquire performance requirement information of a memory, where the performance requirement information represents an expected performance value of the memory; Determining a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters; performing a first performance test on the memory based on the first parameter type to obtain a first test result, wherein the first performance test represents testing a performance value of the memory after adjusting a parameter value of the first parameter type, and comparing the performance value of the memory with an expected performance value to obtain a first comparison result; If the first test result indicates that the performance value of the memory is greater than or equal to the expected performance value, configuring the first parameter type as a test parameter of the memory; If the first test result indicates that the performance value of the memory is less than the expected performance value, modify the first parameter type to obtain a second parameter type, wherein a difference set between the second parameter type and the first parameter type is not equal to an empty set, and a second test result of performing a second performance test on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the expected performance value; configuring the second parameter type as a test parameter of the memory; In a case where the performance requirement information represents the performance expectation values of multiple preset performance indicators of the memory: obtain the performance difference between the performance expectation values of the multiple preset performance indicators and the current performance value, where the current performance value represents the performance value of the preset performance indicator before the first performance test is performed; obtain an association map between the N parameter in the second parameter type and the multiple preset performance indicators; determine the parameter value of the N parameter based on the association map and the performance difference, where the N parameter represents any parameter in the second parameter type, and the association map is used to indicate the change in the performance values of the multiple preset performance indicators when the unit of the N parameter changes.

2. The method according to claim 1, characterized in that The determining the first parameter type according to the performance requirement information and the preset performance parameter table includes: Obtaining a first parameter range in the preset performance parameter table according to the performance requirement information, where the first parameter range represents a parameter range associated with the performance requirement information; Determining a first threshold value according to the performance requirement information and a preset correlation table, wherein the first threshold value represents a correlation threshold value between a parameter and the performance requirement information; The first parameter type is determined according to the first threshold and the first parameter range.

3. The method according to claim 2, characterized in that Obtaining a first parameter range in the preset performance parameter table according to the performance requirement information includes: Obtaining performance impact information of each parameter in the preset performance parameter table, wherein the performance impact information represents change information of each performance indicator when the parameter changes; determining a plurality of parameters associated with the performance requirement information according to the performance requirement information and the performance impact information; A plurality of parameters associated with the performance requirement information are configured as the first parameter range.

4. The method according to claim 3, characterized in that The determining the first parameter type according to the first threshold and the first parameter range includes: Determining, based on the performance impact information, a correlation value between each parameter in the first parameter range and a preset performance indicator in the performance requirement information; Comparing the correlation value of each parameter in the first parameter range with the first threshold value to obtain a second comparison result; The parameter whose correlation value indicated by the second comparison result is greater than or equal to the first threshold is configured as the first parameter type.

5. The method according to claim 4, characterized in that The determining, based on the performance impact information, a correlation value between each parameter in the first parameter range and a preset performance indicator in the performance requirement information includes: When determining, based on the performance impact information, a change in the M parameter unit of the first parameter range, a change range of the preset performance indicator, the M parameter representing any parameter in the first parameter range; A correlation value between the M parameter and the preset performance indicator is determined according to the variation range.

6. The method according to claim 4, characterized in that The modifying the first parameter type to obtain the second parameter type includes: Establishing a Gaussian distribution of the correlation value, wherein a first coordinate value of the Gaussian distribution represents the correlation value, and a second coordinate value of the Gaussian distribution represents each parameter within the first parameter range; Obtaining a first distribution range of the first parameter type in the Gaussian distribution, where the first distribution range represents a first value interval of the second coordinate value; The first value interval is increased by a preset unit interval to obtain a second value interval; All parameters in the second value range are configured as the second parameter type.

7. The method according to claim 1, characterized in that In a case where the performance requirement information represents expected performance values of a plurality of preset performance indicators of the memory, the method further includes: Obtaining performance impact information of the N parameter in the second parameter type, where the performance impact information represents change information of each performance indicator when the N parameter changes; A correlation map between the N parameter and the plurality of preset performance indicators is determined based on the performance impact information.

8. A memory testing system, characterized in that: include: A first module is configured to obtain performance requirement information of a memory, wherein the performance requirement information represents an expected performance value of the memory; A second module is configured to determine a first parameter type according to the performance requirement information and a preset performance parameter table, where the first parameter type includes one or more parameters; a third module, configured to perform a first performance test on the memory based on the first parameter type to obtain a first test result, wherein the first performance test represents testing a performance value of the memory after adjusting a parameter value of the first parameter type, and comparing the performance value of the memory with an expected performance value to obtain a first comparison result; A fourth module is configured to configure the first parameter type as a test parameter of the memory when the first test result indicates that the performance value of the memory is greater than or equal to the expected performance value; A fifth module is configured to, if the first test result indicates that the performance value of the memory is less than the expected performance value, modify the first parameter type to obtain a second parameter type, wherein a difference set between the second parameter type and the first parameter type is not equal to an empty set, and a second test result of a second performance test performed on the memory based on the second parameter type indicates that the performance value of the memory is greater than or equal to the expected performance value; A sixth module, configured to configure the second parameter type as a test parameter of the memory; It also includes: in a case where the performance requirement information represents the performance expected values of multiple preset performance indicators of the memory: obtaining the performance difference between the performance expected values of the multiple preset performance indicators and the current performance value, the current performance value representing the performance value of the preset performance indicator before the first performance test is performed; obtaining an association map between the N parameter in the second parameter type and the multiple preset performance indicators; determining the parameter value of the N parameter based on the association map and the performance difference, the N parameter representing any one parameter in the second parameter type, and the association map being used to indicate the change in the performance values of the multiple preset performance indicators when the unit of the N parameter changes.

9. A memory testing device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to perform the method according to any one of claims 1 to 7 when executed by the processor.

Images