A method, apparatus, and storage medium for processing data

Abstract

Embodiments of the present application disclose a method and device for processing data and a storage medium. The method comprises: obtaining at least one storage capability value according to a steady state value of each two first characteristic parameters, so as to solve the tediousness of manually processing the first characteristic parameters, and make the processing of the first characteristic parameters simpler; for any storage capability value in the at least one storage capability value, when the storage capability value is greater than a preset storage capability value, determining the storage capability value as a to-be-processed parameter value, obtaining at least one to-be-processed parameter value, thereby improving the efficiency of processing the first characteristic parameters; obtaining a storage performance value according to the electrical characteristic parameters and the magnetic characteristic parameters, the storage performance value being used to confirm whether the performance of the MTJ reaches an expectation, thereby improving the accuracy and efficiency of processing the first characteristic parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of storage, and particularly relates to a data processing method and device and storage medium. BACKGROUND

[0002] Computers are widely used in various fields of people's production and life. The computer includes a memory, and the performance of the memory plays a crucial role in the running speed, processing speed and computing power of the computer. Among them, the magnetic random access memory (MRAM) is a kind of memory, and the MRAM has the advantages of avoiding data loss in the case of device power failure, fast reading speed, high erasing and writing resistance, and low power consumption.

[0003] The storage unit in a typical MRAM is a magnetic tunnel junction (MTJ), and the MRAM storage function is realized through the flipping characteristics and reading and writing capabilities of the MTJ. In order to accurately confirm the performance of the MRAM, the characteristic parameters of the MTJ can be extracted, and the extracted characteristic parameters can be analyzed.

[0004] However, the current way of extracting the characteristic parameters of the MTJ is manual extraction, and the processing method of the extracted characteristic parameters is also manual processing, which causes the problems of tedious processing of the characteristic parameters, low efficiency and inaccurate extraction of the parameters, and cannot provide reliable reference for the optimization of the performance of the MRAM. SUMMARY

[0005] In view of the above technical problems, the embodiments of the present application provide a data processing method, device and storage medium to solve the problems of tedious data processing, low efficiency and inaccurate parameter extraction.

[0006] The first aspect of the embodiments of the present application provides a data processing method, comprising:

[0007] obtaining a first set of characteristic parameters of a storage unit which is a magnetic tunnel junction (MTJ);

[0008] obtaining a steady state value corresponding to each first characteristic parameter in the first set of characteristic parameters;

[0009] obtaining at least one storage capacity value according to the steady state values of each two first characteristic parameters;

[0010] For any storage capacity value in the at least one storage capacity value, when the storage capacity value is greater than a preset storage capacity value, determining the storage capacity value as a to-be-processed parameter value, and obtaining at least one to-be-processed parameter value;

[0011] Extracting electrical characteristic parameters and / or magnetic characteristic parameters from the at least one parameter value to be processed;

[0012] A storage performance value is obtained according to the electrical characteristic parameter and / or the magnetic characteristic parameter, and the storage performance value is used to confirm whether the performance of the MTJ meets expectations.

[0013] In some embodiments, obtaining the steady-state value corresponding to each first characteristic parameter in the first characteristic parameter set includes:

[0014] Clustering each of the first characteristic parameters to obtain a first characteristic group and a second characteristic group, wherein the steady state represented by each characteristic in the first steady state group is higher than the steady state represented by each characteristic in the second steady state group;

[0015] Calculating steady-state parameters corresponding to each feature in the first feature group and the second feature group, wherein the first feature group corresponds to a first steady-state parameter group, the second feature group corresponds to a second steady-state parameter group, and a steady-state represented by each steady-state parameter in the first steady-state parameter group is higher than a steady-state represented by each steady-state parameter in the second steady-state parameter group;

[0016] calculating a total distance between each steady-state parameter in the first steady-state parameter group and a first reference steady-state parameter, and taking the first reference steady-state parameter with the smallest total distance as a high steady-state value, where the first reference steady-state parameter is any steady-state parameter in the first steady-state parameter group;

[0017] The total distance between each steady-state parameter in the second steady-state parameter group and a second reference steady-state parameter is calculated, and the second reference steady-state parameter with the smallest total distance is taken as a low steady-state value, where the second reference steady-state parameter is any steady-state parameter in the second steady-state parameter group.

[0018] In some embodiments, obtaining at least one storage capacity value according to the steady-state values ​​of every two first characteristic parameters includes:

[0019] Calculating the difference between the high steady-state value and the low steady-state value to obtain a steady-state difference value;

[0020] The steady-state difference value is divided by the low steady-state value to obtain the storage capacity value.

[0021] In some embodiments, the at least one parameter value to be processed includes at least one set of electrical characteristic parameters and / or magnetic characteristic parameters, and extracting target electrical characteristic parameters and / or target magnetic characteristic parameters from the at least one parameter value to be processed includes:

[0022] Presetting initial target electrical characteristic parameters and / or initial target magnetic characteristic parameters, and calculating the initial target electrical characteristic parameters and / or initial target magnetic characteristic parameters to obtain initial fitting values;

[0023] calculating squares of differences between the actual parameter values and the initial fitting values, and summing all the squares of the differences as an error value;

[0024] comparing the error value with a preset error value, and if the error value is less than the preset error value, confirming that the initial target electrical characteristic parameter and / or initial target magnetic characteristic parameter is the required data characteristic.

[0025] In some embodiments, the obtaining the electrical characteristic parameter and / or the magnetic characteristic parameter comprises:

[0026] The electrical characteristic parameter comprises an interference value and a distribution width value, and the magnetic characteristic parameter comprises an interference value and a distribution width value, the interference value representing an ability to resist external interference, and the distribution width value representing a read-write ability of the storage.

[0027] The distribution width value is divided by the interference value to obtain the storage performance value.

[0028] A second aspect of the embodiments of the present application provides a data processing device, which comprises:

[0029] A first characteristic parameter set obtaining module is configured to obtain a first characteristic parameter set of a magnetic tunnel junction (MTJ) storage unit.

[0030] A steady state value obtaining module is configured to obtain a steady state value corresponding to each first characteristic parameter in the first characteristic parameter set.

[0031] A storage capacity value obtaining module is configured to obtain at least one storage capacity value according to the steady state values of each two first characteristic parameters.

[0032] A to-be-processed parameter value obtaining module is configured to, for any storage capacity value in the at least one storage capacity value, determine the storage capacity value as a to-be-processed parameter value when the storage capacity value is greater than a preset storage capacity value, and obtain at least one to-be-processed parameter value.

[0033] A data extracting module is configured to extract an electrical characteristic parameter and / or a magnetic characteristic parameter from the at least one to-be-processed parameter value.

[0034] A confirming module is configured to obtain a storage performance value according to the electrical characteristic parameter and / or the magnetic characteristic parameter, and the storage performance value is used to confirm whether the performance of the MTJ reaches an expectation.

[0035] In some embodiments, the steady state value obtaining module comprises:

[0036] a clustering module configured to cluster each of the first characteristic parameters to obtain a first characteristic group and a second characteristic group, wherein each characteristic in the first characteristic group represents a higher steady state than each characteristic in the second characteristic group;

[0037] a steady state parameter module configured to calculate a steady state parameter corresponding to each characteristic in the first characteristic group and the second characteristic group, wherein the first characteristic group corresponds to a first steady state parameter group, the second characteristic group corresponds to a second steady state parameter group, and each steady state parameter in the first steady state parameter group represents a higher steady state than each steady state parameter in the second steady state parameter group;

[0038] a first reference steady state module configured to calculate a sum distance between each steady state parameter in the first steady state parameter group and a first reference steady state parameter, and to determine a first reference steady state parameter with a minimum sum distance as a high steady state value, wherein the first reference steady state parameter is any steady state parameter in the first steady state parameter group;

[0039] a second reference steady state module configured to calculate a sum distance between each steady state parameter in the second steady state parameter group and a second reference steady state parameter, and to determine a second reference steady state parameter with a minimum sum distance as a low steady state value, wherein the second reference steady state parameter is any steady state parameter in the second steady state parameter group.

[0040] In some embodiments, the method further comprises a calculating module configured to calculate a difference between the high steady state value and the low steady state value to obtain a steady state difference value, and to divide the steady state difference value by the low steady state value to obtain the storage capacity value.

[0041] In some embodiments, the obtaining parameter value module further comprises:

[0042] a filtering module configured to preset an initial target electrical characteristic parameter and / or an initial target magnetic characteristic parameter, and to calculate an initial fitting value based on the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter;

[0043] a calculating module configured to calculate a square of a difference between the actual parameter value and the initial fitting value, and to calculate a sum of squares of all the squares as an error value;

[0044] a comparing module configured to compare the error value with a preset error value, and to determine that the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter is the required data characteristic if the error value is less than the preset error value.

[0045] A third aspect of the embodiments of the present application provides a computer readable storage medium having computer executable instructions stored thereon, which, when executed by a computing device, can be used to implement the method according to any one of the preceding embodiments.

[0046] The embodiment of the present application obtains at least one storage capability value according to the steady state values of each two first characteristic parameters, solves the tediousness of manually processing the first characteristic parameters, makes the processing of the first characteristic parameters more simple, and for any one of the at least one storage capability value, when the storage capability value is greater than a preset storage capability value, determines the storage capability value as a to-be-processed parameter value, obtains at least one to-be-processed parameter value, thereby improving the efficiency of processing the first characteristic, obtains a storage performance value according to the electrical characteristic parameter and the magnetic characteristic parameter, and the storage performance value is used to confirm whether the performance of the MTJ reaches an expectation, thereby improving the accuracy and efficiency of processing the first characteristic parameter. BRIEF DESCRIPTION OF DRAWINGS

[0047] The features and advantages of the present application will be more clearly understood through reference to the following detailed description when taken in conjunction with the accompanying drawings, which are presented as illustrative and not restrictive of the present application, wherein:

[0048] Figure 1 is a curve schematic diagram of a method for processing data provided by the present application;

[0049] Figure 2 is a flow schematic diagram of a method for processing data provided by the present application. DETAILED DESCRIPTION

[0050] In the following detailed description, many specific details are set forth in order to provide a thorough understanding of the relevant disclosure. However, it will be apparent to one ordinarily skilled in the art that the present application can be practiced without these details. It will be understood by those of ordinary skill in the art that the terms “system,” “device,” “unit,” and / or “module” are used in the present application as a method for distinguishing between different levels of different components, elements, parts, or assemblies in a sequential arrangement. However, these terms can be replaced by other expressions if other expressions can achieve the same purpose.

[0051] It should be understood that when a device, unit, or module is referred to as being “on”, “connected to”, or “coupled to” another device, unit, or module, it can be directly on, connected or coupled to, or in communication with the other device, unit, or module, or there can be intervening devices, units, or modules, unless the context clearly indicates otherwise. For example, the term “and / or” used in the present application includes any one and all combinations of the relevant listed items.

[0052] The terms used in this application are only for describing specific embodiments and are not intended to limit the scope of this application. As shown in the specification and claims of this application, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified features, wholes, steps, operations, elements and / or components, and such expressions do not constitute an exclusive list, and other features, wholes, steps, operations, elements and / or components may also be included.

[0053] Currently, Magnetic Random Access Memory (MRAM), as a new type of non-volatile memory, is widely used in various electronic products. Its advantages are that it can avoid data loss when the device is powered off, and it has fast reading speed, erase and write resistance, and low power consumption.

[0054] The magnetic tunnel junction (MTJ), a core component of magnetic random access memory (MRAM), consists of a structure consisting of two ferromagnetic layers sandwiching a tunneling barrier. MRAM storage functionality is achieved through the MTJ's switching characteristics and read / write capabilities. The switching characteristics and read / write capabilities of the MTJ are characterized through various testing methods, including the application of an external magnetic field. The test data is then analyzed to obtain key characteristic parameters, which are crucial for characterizing the device performance, material properties, and film stack process flow of the MTJ.

[0055] However, the current method for extracting the characteristic parameters of MTJs is manual extraction, and the extracted characteristic parameters are also processed manually. This results in cumbersome and inefficient processing of characteristic parameters and inaccurate extraction parameters, making it impossible to provide a reliable reference for optimizing MRAM performance.

[0056] Figure 1 For example Figure 2 The data processing method provided in the figure is used to obtain data from a magnetic storage unit and process the obtained data. Specifically, based on the normal cumulative distribution function and clustering algorithm, it can batch extract characteristic parameters from the characteristic test data of all MTJ devices with hysteresis characteristics and process the characteristic parameters. This method can also process the characteristic data on the left and right curves of the hysteresis loop separately.

[0057] The hysteresis characteristic refers to the device characteristic Y changing along a closed loop and lagging behind the change of the external test variable X when the external test variable X changes periodically. The main characteristics of the hysteresis loop YX curve obtained by the test are:

[0058] The curve has two different steady-state values, and can jump from one steady-state value to another steady-state value around a certain value of the test variable. The curve has two branches, and the two branches form a closed curve because the test variable changes in opposite directions.

[0059] As shown in Figure 2 The present application provides a method for processing data, comprising:

[0060] Step S10 obtains a first set of characteristic parameters of a storage unit, which is a magnetic tunnel junction (MTJ). The first set of characteristic parameters is a collection of data obtained through electrical tests and magnetic tests. The electrical tests include resistance-magnetic field tests, resistance-voltage tests, and resistance-current density tests. The magnetic tests include magnetization-magnetic field tests and magneto-optical Kerr signal-magnetic field tests. Correspondingly, the above tests can obtain Y-X curves (data curves of the present application) of different types with hysteresis loop characteristics, i.e., the Y-X curves can be R-H curves, R-V curves, R-J curves, M-H curves, Kerr-H curves, etc. In fact, the present application is described by using data obtained through resistance-magnetic field tests.

[0061] Step S20 obtains a steady-state value corresponding to each first characteristic parameter in the first set of characteristic parameters. The first characteristic parameter file format can be.txt,.csv,.svd,.xlsx, etc. Thus, data obtained through different test methods and their different data formats can be used to achieve a wide range of applications. In addition, the general information storage process refers to writing binary information '0' and '1' into a storage device. The magnetic storage unit (magnetic tunnel junction) is a device for storing information. In the present application, the steady-state value represents '0' and '1', respectively. Through a certain test method, the device characteristic value Y will change between 0 and 1, thereby characterizing the information read-write capability of the magnetic storage unit.

[0062] Step S30 obtains at least one storage capability value according to the steady-state values of each two first characteristic parameters. The greater the storage capability value, the better the distinction between '0' and '1' when writing information, and the better the device can represent '0' and '1' in actual use, i.e., the better the storage capability.

[0063] Step S40 determines, for any one of the at least one storage capability value, that the storage capability value is a to-be-processed parameter value when the storage capability value is greater than a preset storage capability value, to obtain at least one to-be-processed parameter value. When it is determined that the storage capability value is greater than the preset storage capability value, it can be considered that the to-be-processed parameter value obtained by the present application is a normal parameter value, and the storage capability value less than the preset storage capability value is excluded, so that all the to-be-processed parameter values obtained are normal values, that is, the quality of the storage device obtained is normal. When the storage capability value is less than the preset storage capability value, the quality of the storage device obtained is poor, and the next step cannot be normally executed.

[0064] By step S40 screening, at least one normal to-be-processed parameter value is obtained, and the at least one normal to-be-processed parameter value is subjected to the next step, that is, step S50 extracts an electrical characteristic parameter and / or a magnetic characteristic parameter from the at least one to-be-processed parameter value. The to-be-processed parameter value is fitted to obtain a fitting curve, and the electrical characteristic parameter and / or the magnetic characteristic parameter are extracted from the fitting curve. According to the drawing of the fitting curve, data of a storage device with normal characteristics are preliminarily screened out, so as to avoid defective data leading to failure of curve fitting, enhance the executable degree of the flow, and improve the efficiency and accuracy. The test of the electrical characteristic parameter includes the above resistance-magnetic field test, and also includes resistance-voltage test, resistance-current density test, and the test of the magnetic characteristic parameter includes magnetization-magnetic field test, magneto-optical Kerr signal-magnetic field test, etc. Correspondingly, the above test methods can obtain Y-X curves (data) of different types with hysteresis loop characteristics, that is, the Y-X curve can be R-H curve, R-V curve, R-J curve, M-H curve, Kerr-H curve, etc.

[0065] Step S60 obtains a storage performance value according to the electrical characteristic parameter and / or the magnetic characteristic parameter, and the storage performance value is used to confirm whether the performance of the MTJ reaches the expectation. By obtaining the size of the storage performance value, it can be confirmed whether the MTJ reaches the expectation by comparing the performance value with a preset standard. Generally, the smaller the storage performance value is, the better the quality of the storage device is, and therefore the storage performance value needs to be taken as a standard to compare different batches of storage devices with different processes and material characteristics, so as to more reasonably and effectively find the most suitable process, material, experimental condition, etc. The storage performance represents read-write speed, power consumption, and anti-interference ability.

[0066] The step solves the tediousness of manually processing the first characteristic parameters, makes the processing of the first characteristic parameters more simple, and for any one of the at least one storage capacity value, when the storage capacity value is greater than a preset storage capacity value, the storage capacity value is determined as a to-be-processed parameter value, at least one to-be-processed parameter value is obtained, thereby improving the efficiency of processing the first characteristic, and the storage performance value is obtained according to the electrical characteristic parameter and the magnetic characteristic parameter, the storage performance value is used to confirm whether the performance of the MTJ reaches the expectation, and the accuracy of processing the first characteristic parameter is improved.

[0067] Preferably, the obtaining of the steady state value corresponding to each first characteristic parameter in the first characteristic parameter set comprises:

[0068] The first characteristic group and the second characteristic group are obtained by clustering each first characteristic parameter, the steady state of each characteristic in the first steady state group is higher than the steady state of each characteristic in the second steady state group. The steady state parameter corresponding to each characteristic in the first characteristic group and the second characteristic group is calculated, wherein the first characteristic group corresponds to the first steady state parameter group, the second characteristic group corresponds to the second steady state parameter group, and the steady state of each steady state parameter in the first steady state parameter group is higher than the steady state of each steady state parameter in the second steady state parameter group. By using a clustering algorithm to divide all the first characteristic parameters into two groups, two steady state values are calculated, and then the storage capacity value is calculated by using the two steady state values. In fact, due to experimental errors and instability of storage devices in the actual process of obtaining the first characteristic parameters, the data distribution in each group is relatively scattered, so a central data point is needed to represent other data points in the group.

[0069] The sum distance of each steady state parameter in the first steady state parameter group and a first reference steady state parameter is calculated, the first reference steady state parameter with the minimum sum distance is taken as the high steady state value, and the first reference steady state parameter is any steady state parameter in the first steady state parameter group.

[0070] The sum distance of each steady state parameter in the second steady state parameter group and a second reference steady state parameter is calculated, the second reference steady state parameter with the minimum sum distance is taken as the low steady state value, and the second reference steady state parameter is any steady state parameter in the second steady state parameter group.

[0071] Generally, the information storage process refers to writing binary information '0' and '1' into a storage device. The magnetic storage unit (magnetic tunnel junction) is a storage device for storing information. Specifically, a high steady state represents '1', and a low steady state represents '0'. Through a certain test method, the characteristic value Y of the storage device will change between 0 and 1, thereby representing the information writing capability of the magnetic storage unit.

[0072] Preferably, the at least one storage capability value is obtained according to the steady state value of each two first characteristic parameters, comprising:

[0073] The difference between the high steady state value and the low steady state value is calculated to obtain a steady state difference value. The specific implementation process is as follows:

[0074]

[0075] Y0 refers to the centroid, i.e. the first reference steady state parameter or the second reference steady state parameter. Y refers to the steady state parameter value in this group. L refers to the sum distance of each steady state parameter from the first reference steady state parameter in the first reference steady state parameter group or the sum distance of each steady state parameter from the second reference steady state parameter in the second reference steady state parameter group.

[0076] Further, the steady state difference value is divided by the low steady state value to obtain the storage capability value. The calculation process of the storage capability value is as follows:

[0077] Ratio=(Y H -Y L ) / Y L ×100%

[0078] Ratio refers to the storage capability value, Y H refers to the high steady state value, and Y L refers to the low steady state value.

[0079] By judging the condition that the storage capability value Ratio>10% and the low steady state value Y L >0, data without normal R-H characteristics is excluded, i.e. the obtained parameter values to be processed are all normal parameter values, thereby providing necessary conditions for successful fitting in the next step. Due to the instability of the process, the yield is low, and many magnetic tunnel junction devices cannot exhibit normal electrical characteristics, resulting in chaotic data. This judgment condition is used to preliminarily select device data with normal electrical characteristics, avoiding process interruption caused by subsequent curve fitting failure. 10% is a preset value, i.e. an empirical value. Of course, in addition to taking 10%, some parameters with higher requirements need to take a preset value greater than 10%. Therefore, the preset value can be taken in the range of 10%-20%.

[0080] Preferably, the at least one to-be-processed parameter value comprises at least one set of electrical characteristic parameters and / or magnetic characteristic parameters, and the extracting of the target electrical characteristic parameter and / or the target magnetic characteristic parameter from the at least one to-be-processed parameter value comprises:

[0081] The preset initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter is calculated to obtain an initial fitting value. The present application is described by taking electrical characteristic parameters as an example, and is described by taking resistance-magnetic field as an example. The preset initial target electrical characteristic parameter is calculated as follows:

[0082]

[0083] H sw is a critical flipping magnetic field, H sw_sigma is a flipping field distribution width, R i refers to an initial fitting value, that is, a resistance value, and H refers to a magnetic field.

[0084] The square of the difference between the actual parameter value and the initial fitting value is calculated, and the sum of all the squares of the differences is taken as an error value. The error value is calculated as follows:

[0085]

[0086] S refers to an error value, R i (Hsw, Hsw_sigma) refers to an initial fitting value, and R is an actual parameter value.

[0087] To obtain a plurality of sets of R i values, a plurality of sets of initial fitting values R i are obtained by substituting a set of Hsw, Hsw_sigma parameters into a plurality of different magnetic fields H. i The plurality of sets of R

[0088] Each error value in the plurality of sets of error values is compared with a preset error value. If the error value is less than the preset error value, it is confirmed that the current initial target electrical characteristic parameter and / or initial target magnetic characteristic parameter is the required data characteristic. If the error value is greater than the preset error value, it is confirmed that the current preset initial target electrical characteristic parameter is unqualified, and the size of the preset initial target electrical characteristic parameter needs to be adjusted until the error value is less than the preset error value. It should be noted that the above calculation implementation manner is also applicable to magnetic characteristic parameters.

[0089] Preferably, the according to the electrical characteristic parameters and / or the magnetic characteristic parameters comprises:

[0090] The electrical characteristic parameter includes an interference value and a distribution width value, the magnetic characteristic parameter includes an interference value and a distribution width value, the interference value represents the ability to resist external interference, and the distribution width value represents the read-write ability of storage; the distribution width value is divided by the interference value to obtain the storage performance value. Wherein, the interference value is X sw , the distribution width value is X sw_sigma , and the specific calculation process of the storage performance value is:

[0091] XCV=X sw_sigma / X sw *100%

[0092] Wherein, X CV represents the storage performance value, X CV is a characteristic parameter for comprehensively evaluating the performance of a device, and X CV is usually determined by experience. If X CV is small, it can be considered that the performance of the memory device is good. Because if the distribution width value is small and the interference value is also small, it is not an ideal device performance. If the interference value is large, but the distribution width value is also large, it is also not an ideal device performance. Therefore, X CV is used as a standard reference value to compare different batches of memory devices with different processes and material characteristics, so that the most suitable process, material, experimental conditions, etc. can be found more reasonably and effectively. In addition, the distribution width value representing the read-write ability of storage includes read-write speed, power consumption, etc.

[0093] It should be noted that the present application is from high steady state jump to low steady state and from low steady state jump to high steady state. The required external test variable X value is different, that is, the jump position is different. According to the characteristic parameters corresponding to the two processes extracted respectively, the material thickness and experimental conditions can be adjusted to achieve the desired device performance. Wherein, X SW can represent the ability of the memory device (magnetic basic unit) to resist external interference (magnetic field or voltage). Within a certain range, the larger X SW is, the stronger the ability of the memory device to resist external interference in actual application is, the lower the probability of error writing is, and the stronger the data storage retention ability is. For example, in the above embodiment, the resistance-magnetic field test, X sw is H swCritical Switching Field, i.e. the anti-interference ability of the memory device is relatively strong, and the original steady state value can be maintained in a certain magnetic field environment instead of jumping due to a little interference. The magnetic tunnel junction is composed of many small magnetization regions (domains), and the sizes of these domains are not completely the same, so the external variable value that can make the magnetization direction of the magnetic tunnel junction flip is also not the same. When testing the device, under the action of the same external variable X, the magnetization direction of some domains has flipped, some have not flipped, and some are in the process of flipping. sw_sigma i.e. the distribution width, which is used to represent whether the magnetization directions of the domains can be concentrated and unified near a certain external condition value X sw . Overall, the smaller X sw_sigma , i.e. the smaller the flip distribution width, the more concentrated the magnetization directions of the domains flip, and the faster the flip speed of the magnetic tunnel junction, and the lower the write power consumption. In the face of a large amount of test data, the present application is a key tool that can realize the flow and batch processing of data, and can meet the feature data extraction needs of various test schemes. Its efficiency is much higher than that of the prior art, and the obtained results are accurate and reliable.

[0094] The second aspect of the embodiment of the present application provides a data processing device, and the device comprises:

[0095] A first feature parameter set acquisition module is configured to acquire a first feature parameter set of a storage unit that is a magnetic tunnel junction (MTJ);

[0096] A steady state value acquisition module is configured to acquire a steady state value corresponding to each first feature parameter in the first feature parameter set;

[0097] A storage capacity value acquisition module is configured to obtain at least one storage capacity value according to the steady state values of each two first feature parameters;

[0098] A to-be-processed parameter value acquisition module is configured to, for any storage capacity value in the at least one storage capacity value, determine the storage capacity value as a to-be-processed parameter value when the storage capacity value is greater than a preset storage capacity value, and obtain at least one to-be-processed parameter value;

[0099] An extracted data module is configured to extract an electrical feature parameter and / or a magnetic feature parameter from the at least one to-be-processed parameter value;

[0100] A confirmation module is configured to obtain a storage performance value according to the electrical feature parameter and / or the magnetic feature parameter, and the storage performance value is used to confirm whether the performance of the MTJ meets the expectation.

[0101] The module obtains at least one storage capability value according to the steady state value of each two first characteristic parameters, solves the tediousness of manual processing of the first characteristic parameters, makes the processing of the first characteristic parameters more simple, and for any storage capability value in the at least one storage capability value, when the storage capability value is greater than a preset storage capability value, the storage capability value is determined as a to-be-processed parameter value, at least one to-be-processed parameter value is obtained, thereby improving the efficiency of processing the first characteristic, and the storage performance value is obtained according to the electrical characteristic parameter and the magnetic characteristic parameter, the storage performance value is used to confirm whether the performance of the MTJ reaches the expectation, and the accuracy of processing the first characteristic parameter is improved.

[0102] In some embodiments, the obtaining steady state value module comprises:

[0103] The clustering module is configured to cluster each first characteristic parameter to obtain a first characteristic group and a second characteristic group, and a steady state of each characteristic in the first steady state group is higher than a steady state of each characteristic in the second steady state group.

[0104] The steady state parameter module is configured to calculate a steady state parameter corresponding to each characteristic in the first characteristic group and the second characteristic group, wherein the first characteristic group corresponds to a first steady state parameter group, the second characteristic group corresponds to a second steady state parameter group, and a steady state of each steady state parameter in the first steady state parameter group is higher than a steady state of each steady state parameter in the second steady state parameter group.

[0105] The first reference steady state module is configured to calculate a sum distance of each steady state parameter in the first steady state parameter group and a first reference steady state parameter, and take the first reference steady state parameter with the minimum sum distance as a high steady state value, wherein the first reference steady state parameter is any steady state parameter in the first steady state parameter group.

[0106] The second reference steady state module is configured to calculate a sum distance of each steady state parameter in the second steady state parameter group and a second reference steady state parameter, and take the second reference steady state parameter with the minimum sum distance as a low steady state value, wherein the second reference steady state parameter is any steady state parameter in the second steady state parameter group.

[0107] In some embodiments, the computing module is further configured to calculate a difference between the high steady state value and the low steady state value to obtain a steady state difference value, and divide the steady state difference value by the low steady state value to obtain the storage capability value.

[0108] In some embodiments, the obtaining to-be-processed parameter value module further comprises:

[0109] The filtering module is configured to preset an initial target electrical characteristic parameter and / or an initial target magnetic characteristic parameter, and calculate the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter to obtain an initial fitting value.

[0110] squares of all the differences as an error value;

[0111] comparing the error value with a preset error value, and if the error value is less than the preset error value, confirming that the initial target electrical characteristic parameter and / or initial target magnetic characteristic parameter is the required data characteristic.

[0112] A third aspect of the embodiments of the present application provides a computer readable storage medium having computer executable instructions stored thereon, which, when executed by a computing device, can be used to implement the method as described in the foregoing embodiments.

[0113] It should be understood that the above specific embodiments of the present application are merely used for illustrative or explanatory purposes, and do not constitute a limitation on the present application. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present application shall be included in the protection scope of the present application. In addition, the claims of the present application are intended to cover all variations and modifications falling within the scope and boundary of the appended claims, or the equivalent forms of such scope and boundary.

Claims

1. A method of processing data, characterized by, The method comprises: obtaining a first characteristic parameter set of a magnetic tunnel junction (MTJ) storage unit; obtaining a steady-state value corresponding to each first characteristic parameter in the first characteristic parameter set; obtaining at least one storage capacity value according to the steady-state values of each two first characteristic parameters; for any storage capacity value in the at least one storage capacity value, when the storage capacity value is greater than a preset storage capacity value, determining the storage capacity value as a to-be-processed parameter value, and obtaining at least one to-be-processed parameter value; extracting an electrical characteristic parameter and / or a magnetic characteristic parameter from the at least one to-be-processed parameter value; obtaining a storage performance value according to the electrical characteristic parameter and / or the magnetic characteristic parameter, the storage performance value being used to confirm whether the performance of the MTJ reaches an expectation; the obtaining of the steady-state value corresponding to each first characteristic parameter in the first characteristic parameter set comprises: clustering the each first characteristic parameter to obtain a first characteristic group and a second characteristic group, a steady state represented by each characteristic in the first steady state group being higher than a steady state represented by each characteristic in the second steady state group; calculating a steady-state parameter corresponding to each characteristic in the first characteristic group and the second characteristic group, wherein the first characteristic group corresponds to a first steady-state parameter group, the second characteristic group corresponds to a second steady-state parameter group, and a steady state represented by each steady-state parameter in the first steady-state parameter group is higher than a steady state represented by each steady-state parameter in the second steady-state parameter group; calculating a sum distance of each steady-state parameter in the first steady-state parameter group and a first reference steady-state parameter, taking the first reference steady-state parameter with the minimum sum distance as a high steady-state value, and the first reference steady-state parameter being any steady-state parameter in the first steady-state parameter group; calculating a sum distance of each steady-state parameter in the second steady-state parameter group and a second reference steady-state parameter, taking the second reference steady-state parameter with the minimum sum distance as a low steady-state value, and the second reference steady-state parameter being any steady-state parameter in the second steady-state parameter group; the obtaining of the at least one storage capacity value according to the steady-state values of each two first characteristic parameters comprises: calculating a steady-state difference value by subtracting the low steady-state value from the high steady-state value; dividing the steady-state difference value by the low steady-state value to obtain the storage capacity value.

2. The method of processing data according to claim 1, wherein, The at least one to-be-processed parameter value comprises at least one group of electrical characteristic parameters and / or magnetic characteristic parameters, and the extracting of the target electrical characteristic parameter and / or the target magnetic characteristic parameter from the at least one to-be-processed parameter value comprises: presetting an initial target electrical characteristic parameter and / or an initial target magnetic characteristic parameter, and calculating the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter to obtain an initial fitting value; calculating a square of a difference value between an actual parameter value and the initial fitting value, and taking a sum of squares of all difference values as an error value; comparing the error value with a preset error value, and if the error value is less than the preset error value, confirming that the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter is the required data characteristic.

3. The method of processing data according to any of claims 1-2, wherein, the obtaining of the storage performance value according to the electrical characteristic parameter and / or the magnetic characteristic parameter comprises: The electrical characteristic parameter includes an interference value and a distribution width value, the magnetic characteristic parameter includes an interference value and a distribution width value, the interference value represents the ability to resist external interference, and the distribution width value represents the read-write ability of the storage. The distribution width value is divided by the interference value to obtain the storage performance value.

4. A data processing apparatus, characterized by The device includes: A first characteristic parameter set acquisition module is configured to acquire a first characteristic parameter set of a storage unit that is a magnetic tunnel junction (MTJ). A steady state value acquisition module is configured to acquire a steady state value corresponding to each first characteristic parameter in the first characteristic parameter set. A storage capacity value acquisition module is configured to obtain at least one storage capacity value according to the steady state values of each two first characteristic parameters. A to-be-processed parameter value acquisition module is configured to, for any storage capacity value in the at least one storage capacity value, determine the storage capacity value as a to-be-processed parameter value when the storage capacity value is greater than a preset storage capacity value, and obtain at least one to-be-processed parameter value. An extraction data module is configured to extract electrical characteristic parameters and magnetic characteristic parameters from the at least one to-be-processed parameter value. A confirmation module is configured to obtain a storage performance value according to the electrical characteristic parameters and / or the magnetic characteristic parameters, and the storage performance value is used to confirm whether the performance of the MTJ meets an expectation. The steady state value acquisition module includes: A clustering module is configured to cluster each first characteristic parameter to obtain a first characteristic group and a second characteristic group, and a steady state represented by each characteristic in the first steady state group is higher than a steady state represented by each characteristic in the second steady state group. A steady state parameter module is configured to calculate a steady state parameter corresponding to each characteristic in the first characteristic group and the second characteristic group, wherein the first characteristic group corresponds to a first steady state parameter group, the second characteristic group corresponds to a second steady state parameter group, and a steady state represented by each steady state parameter in the first steady state parameter group is higher than a steady state represented by each steady state parameter in the second steady state parameter group. A first reference steady state module is configured to calculate a sum distance of each steady state parameter in the first steady state parameter group and a first reference steady state parameter, and take the first reference steady state parameter with the minimum sum distance as a high steady state value, wherein the first reference steady state parameter is any steady state parameter in the first steady state parameter group. A second reference steady state module is configured to calculate a sum distance of each steady state parameter in the second steady state parameter group and a second reference steady state parameter, and take the second reference steady state parameter with the minimum sum distance as a low steady state value, wherein the second reference steady state parameter is any steady state parameter in the second steady state parameter group. A calculation module is further configured to calculate a difference between the high steady state value and the low steady state value to obtain a steady state difference value, and divide the steady state difference value by the low steady state value to obtain the storage capacity value.

5. The apparatus for processing data according to claim 4, wherein, The to-be-processed parameter value acquisition module further includes: A filtering module is configured to preset an initial target electrical characteristic parameter and an initial target magnetic characteristic parameter, calculate an initial fitting value of the initial target electrical characteristic parameter and / or the initial target magnetic characteristic parameter. Calculate the square of the difference between the actual parameter value and the initial fitting value, and take the sum of all squares of the differences as an error value. Comparing the error value with a preset error value, if the error value is less than the preset error value, it is confirmed that the initial target electrical characteristic parameter and / or initial target magnetic characteristic parameter is the required data characteristic.

6. A storage medium having stored thereon computer-executable instructions for causing a computing device to perform the method of any of claims 1-3 when the computer-executable instructions are executed by the computing device.

Images