An internet of things data classified storage method and device

Abstract

The application provides an Internet of Things data classified storage method and device, belongs to the field of Internet of Things, and is used for solving the problem that the storage strategy determination of Internet of Things data is difficult to implement in related technologies. In the method and device, the storage share and storage strategy threshold of the to-be-stored data are determined by using a plurality of data identifiers of the to-be-stored data, and then the selected storage strategy of the to-be-stored data is determined according to the storage share and the storage strategy threshold in combination with a pre-constructed strategy evaluation model.

Description

Technical Field

[0001] This application relates to the field of the Internet of Things (IoT), and more particularly to an IoT data classification and storage method and apparatus. Background Technology

[0002] With the development of IoT technology, the scale of IoT is becoming increasingly large and its structure increasingly complex, and the data generated by IoT is becoming increasingly abundant. Therefore, the classification and storage of IoT data has become an important task in the IoT field. Current IoT data classification and storage solutions generally involve configuring the database structure and data identifiers for IoT data, and then configuring the association between the data identifiers and database identifiers to determine the storage strategy for IoT data. However, in practical IoT applications, the database structure and data identifiers are not static. When the database structure and data identifiers change, it is necessary to reconfigure the association between the data identifiers and database identifiers. This reconfiguration process is labor-intensive and costly. Summary of the Invention

[0003] This application provides a method and apparatus for classifying and storing IoT data, which can intelligently determine the storage strategy for IoT data.

[0004] Firstly, this application provides a method for classifying and storing Internet of Things (IoT) data. The method includes:

[0005] Obtain the data identification information of the data to be stored, wherein the data identification information includes multiple data identifiers;

[0006] The storage condition information of the data to be stored is determined based on the data identification information, and the storage condition information includes the number of storage copies and the storage policy threshold.

[0007] The selected storage strategy for the data to be stored is determined based on the storage condition information;

[0008] Determining the selected storage strategy for the stored data based on the storage condition information includes:

[0009] An optional storage strategy is determined based on the pre-acquired available database identifiers and the number of storage copies, wherein the optional storage strategy includes the number of storage copies and the number of available database identifiers;

[0010] Substituting data identification information and optional storage strategies into a pre-built strategy evaluation model yields an optional strategy score. The strategy evaluation model is used to determine the optional strategy score based on the available database identifiers in the optional storage strategies.

[0011] The selected storage strategy is determined by identifying an optional storage strategy whose score is not lower than the storage strategy threshold.

[0012] By adopting the above technical solution, the storage strategy of the data to be stored can be intelligently determined based on the data identification information, without the need to configure the correspondence between data identifiers and database identifiers one by one. It is highly intelligent and has a low cost.

[0013] Furthermore, the step of substituting the data identification information and optional storage strategies into the pre-built strategy evaluation model to obtain optional strategy scores includes:

[0014] The call frequency data, data volume, and importance coefficient are determined based on the data identification information.

[0015] The reliability coefficient and retrieval speed data are determined based on the optional storage strategy;

[0016] The score of the optional strategy is determined based on the call frequency data, data volume, call speed data, importance coefficient, and reliability coefficient.

[0017] Furthermore, determining the call frequency data, data volume, and importance coefficient based on the data identification information includes:

[0018] All possible identifier groups are determined based on the data identifier information, wherein the possible identifier groups include one or more data identifiers within the data identifier information;

[0019] Based on data access to big data, a call frequency component is determined for each possible identifier group;

[0020] Based on the preset calculation weight of each possible identifier group, the weighted sum of all call frequency components is calculated as the call frequency data.

[0021] Furthermore, determining the call frequency data, data volume, and importance coefficient based on the data identification information includes:

[0022] Determine the importance component of each data identifier in the data identifier information;

[0023] The importance coefficient is calculated based on all importance components. In the formula, z is the importance coefficient. This represents the i-th importance component. p is the number of importance components. This indicates taking the maximum value among all importance components, where e is the natural constant.

[0024] Furthermore, determining the reliability coefficient and call speed data based on the optional storage strategy includes:

[0025] Obtain the reliability component pre-acquired relative to each available database identifier;

[0026] The reliability coefficient is calculated based on the reliability component identified by the available database in the optional storage strategy. In the formula, w is the reliability coefficient. This represents the reliability component of the i-th available database in the optional storage strategy. q represents the number of available database identifiers in the optional storage strategy. This indicates taking the maximum value among all reliability components, where e is a natural constant.

[0027] Furthermore, determining the reliability coefficient and call speed data based on the optional storage strategy includes:

[0028] Get the database call speed relative to each available database identifier;

[0029] The maximum value of the database call speed corresponding to the available database identifier for the optional storage strategy is the call speed data.

[0030] Furthermore, determining the score of the optional strategy based on the call frequency data, data volume, call speed data, importance coefficient, and reliability coefficient includes:

[0031] The first impact score is calculated based on call frequency data, data volume, and call speed data. In the formula, g is the first influence score, f is the call frequency data, b is the data volume, and v is the call speed data;

[0032] The scores for the optional strategies are calculated based on the first impact score, the importance coefficient, and the reliability coefficient. In the formula, y is the optional strategy score, and a is the preset adjustment value. .

[0033] Further, determining an optional storage strategy whose score is not lower than the storage strategy threshold as the selected storage strategy includes:

[0034] Selectable storage strategies whose optional strategy scores are not lower than the storage strategy threshold are identified as alternative storage strategies;

[0035] The alternative storage strategy with the lowest score among the available strategies is selected as the chosen storage strategy.

[0036] Further, determining the storage condition information of the data to be stored based on the data identification information includes:

[0037] Obtain the preset policy threshold component for each data identifier in the relative data identifier information;

[0038] The storage policy threshold is calculated based on the policy threshold components of all data identifiers in the data identifier information. In the formula, h is the storage policy threshold. This represents the policy threshold component of the i-th data identifier in the data identifier information. s represents the number of data identifiers in the data identifier information. This indicates taking the maximum value among all policy threshold components, where e is a natural constant.

[0039] In a second aspect, this application provides a classification and storage device for Internet of Things (IoT) data, used to perform any of the methods described in the first aspect above.

[0040] In summary, this application has at least the following beneficial effects:

[0041] 1. A method and apparatus for classifying and storing IoT data are provided, which can intelligently and rationally determine the storage strategy of IoT data, which is conducive to reducing costs and increasing efficiency in IoT data management.

[0042] 2. The self-designed and trained policy evaluation model and the algorithm for determining the selected storage policy based on the policy evaluation model are reasonable and accurate, which is conducive to determining the selected storage policy more reasonably and accurately.

[0043] It should be understood that the description in the Summary Section is not intended to limit the key or essential features of the embodiments of this application, nor is it intended to restrict the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0044] The above and other features, advantages, and aspects of the embodiments of this application will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0045] Figure 1 A flowchart of an IoT data classification and storage method according to an embodiment of this application is shown. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0047] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0048] This application provides a method and apparatus for classifying and storing IoT data, which can intelligently determine the storage strategy for IoT data.

[0049] Firstly, this application provides a method for classifying and storing IoT data. This method can be executed by a server.

[0050] Figure 1 A flowchart of an IoT data classification and storage method according to an embodiment of this application is shown.

[0051] Reference Figure 1 The method specifically includes the following steps:

[0052] S110: Obtain the data identification information of the data to be stored.

[0053] The data identification information includes multiple data identifiers. Specifically, the data identifiers are constructed based on a pre-built data identifier system. Data identifiers can also be expressed as data tags, data classification tags, etc. Each data identifier represents a certain characteristic of the data. Taking some known data identifiers as examples, for instance, a time identifier reflects the time when the data to be stored was generated, a data volume identifier reflects the size of the data to be stored, a terminal identifier reflects which IoT terminal device generated the data to be stored, a scene identifier reflects the application scene of the data to be stored, and so on. In short, based on the pre-built data identifier system, multiple data identifiers can be assigned to each piece of IoT data when it is generated, or the data identifiers that should be assigned to the IoT data can be determined according to the characteristics of the IoT data, so that the data identification information of the data to be stored can be determined and obtained.

[0054] S120: Determine the storage condition information of the data to be stored based on the data identification information.

[0055] The storage condition information includes the number of storage copies and storage policy thresholds. It should be understood that the data to be stored in this embodiment is ultimately stored in a data system. The database system includes multiple sub-databases. For each piece of data to be stored, it is possible to choose to store a single copy in one sub-database, or to choose to store one or more backups in multiple sub-databases respectively, to improve the storage reliability of the data. Of course, different sub-databases have different performance parameters such as storage reliability and retrieval speed; therefore, even with the same number of storage copies, storing in different sub-databases will have different effects.

[0056] In this step, determining the storage conditions of the data to be stored does not explicitly determine which sub-database the data will be stored in, but only determines the number of copies and the storage policy threshold. The storage policy threshold can be used to subsequently determine which sub-database the data will be stored in.

[0057] In this step, the determination of the number of storage copies specifically includes: a pre-built identification copy number lookup table, the data identification information of the data to be stored includes multiple data identifiers, and the identification copy number lookup table includes the correspondence between the data identifier information and the number of storage copies, that is, after the data identifier information to be stored is determined, its number of storage copies can be determined.

[0058] In this step, determining the storage policy threshold specifically includes: obtaining a preset policy threshold component for each data identifier in the relative data identifier information; and calculating the storage policy threshold based on the policy threshold components of all data identifiers in the data identifier information. In the formula, h is the storage policy threshold. This represents the policy threshold component of the i-th data identifier in the data identifier information. s represents the number of data identifiers in the data identifier information. This indicates taking the maximum value among all policy threshold components, where e is a natural constant.

[0059] A pre-built identifier-policy lookup table is provided, which includes all data identifiers and policy threshold components. After determining the data identifier, the corresponding policy threshold component can be determined by substituting the data identifier into the identifier-policy lookup table. All policy threshold components are constants.

[0060] S130: Determine the selected storage strategy for the data to be stored based on the storage condition information.

[0061] The method in this step includes: determining an optional storage strategy based on the pre-acquired available database identifiers and the number of storage copies, wherein the optional storage strategy includes the number of storage copies and available database identifiers; substituting the data identifier information and the optional storage strategy into a pre-constructed strategy evaluation model to obtain an optional strategy score, wherein the strategy evaluation model is used to determine the optional strategy score based on the available database identifiers in the optional storage strategy; and determining an optional storage strategy with an optional strategy score not lower than the storage strategy threshold as the selected storage strategy.

[0062] Specifically, in this step, the server can obtain the database identifier and storage status of each sub-database, and determine the available database identifier based on the storage status. For a sub-database, if the remaining storage space of the sub-database can support the storage of the data to be stored (the capacity of the remaining storage space is not less than the amount of data to be stored), then the sub-database is considered an available database for the data to be stored, and the corresponding database identifier of the sub-database is the available database identifier for the data to be stored. In this way, all available databases and their corresponding available database identifiers for the data to be stored can be determined. Combined with the number of storage copies, all possible storage strategies can be determined. The possible storage strategies are all combinations of a number of storage copies selected from all available database identifiers. For example, available database identifiers include... , , If the number of storage copies is two, then all available storage strategies are from , , Choose any two available database identifiers. There are three storage strategies to choose from: , , .

[0063] In this step, the process of substituting data identification information and optional storage strategies into a pre-built strategy evaluation model to obtain optional strategy scores includes: determining call frequency data, data volume, and importance coefficient based on the data identification information; determining reliability coefficient and call speed data based on the optional storage strategy; and determining the optional strategy score based on the call frequency data, data volume, call speed data, importance coefficient, and reliability coefficient.

[0064] In one example, the method for determining the call frequency based on data identification information specifically includes: determining all possible identification groups based on data identification information, wherein the possible identification groups include one or more data identifiers within the data identification information; determining a call frequency component relative to each possible identification group based on the data call big data; and calculating the weighted sum of all call frequency components based on the preset calculation weight of each possible identification group, which is the call frequency data.

[0065] For a specific method of determining possible identifier groups based on data identifier information, let the data identifier information include... Data identifier ( If the integer is positive, then when determining possible identifier groups based on the data identifier information, first determine the number of data identifiers in the possible identifier group. The number of data identifiers in a possible identifier group is not less than 1 and not greater than 1. Once the number of data identifiers in a possible identifier group is determined, all positive integers can be selected from the data identifier information to obtain all possibilities for the possible identifier group under that number of data identifiers. This allows us to determine all possibilities for the possible identifier group under all possible numbers. For example, suppose the data identifier information includes three data identifiers, namely... , , The possible number of data identifiers in a possible identifier group is 1, 2, or 3. Correspondingly, there are three possible identifier groups with one data identifier. , , There are three possible identifier groups with two data identifiers. , , There are 3 possible identifier groups, one of which is... .

[0066] Of course, in data identification information, the number of data identifiers is relatively large. When the value is large, if we take all the data identifiers in the possible identifier group (i.e., not less than 1 and not greater than 1), we can determine the value of the data identifiers. (All positive integers) may result in excessive computation, so it is also possible to start from... ( Start by taking, and take no less than and not greater than Positive integers, about The determination of this can be compared to Small preset value, or equal to The preset ratio, etc., are generally in Only when the number of identifiers exceeds the threshold will it start from... If no identifier is selected, it can be selected starting from 1; otherwise, it can start from 1. The identifier quantity threshold is preset in the server. In this embodiment, the identifier quantity threshold is 5. When the value is not greater than 5, take values ​​from 1 to 5. When it is greater than 5, take arrive .

[0067] The data retrieval big data refers to the retrieval records of all stored data in the database system. It records the retrieval time, retrieval speed, and retrieval result of each piece of stored data (reserved data that are backups of each other are counted as different copies). The stored data also has data identification information, which includes multiple data identifiers for that piece of stored data. A possible identifier group includes one or more data identifiers. Each stored data retrieval record carrying only the data identifiers of all possible identifier groups can be identified from the data retrieval big data. The retrieval frequency within a preset time period is determined based on these stored data retrieval records. The retrieval frequency component of the possible identifier group can be determined based on the average and / or median retrieval frequency of these stored data. In this embodiment, when calculating the retrieval frequency component of the possible identifier group, the average and standard deviation of all retrieval frequencies are first calculated. Then, the average minus three times the standard deviation and the average plus three times the standard deviation are calculated. The average retrieval frequency within the range of average minus three times the standard deviation to average plus three times the standard deviation is determined as the retrieval frequency component of the possible identifier group. In this way, the retrieval frequency component of each possible identifier group can be determined.

[0068] In one specific implementation, for each possible data identifier, all possible identifier groups can be determined. Therefore, a weight lookup table can be pre-constructed, which includes the preset calculation weights corresponding to the possible identifier groups of the data identifier. This allows a preset calculation weight to be determined relative to each possible identifier group after the data identifier of the data to be stored is determined.

[0069] In another specific implementation, to reduce computational load, a preset computational weight can be determined relative to the number of each possible identifier group. For example, if the identifier quantity threshold is 5, when... When the value is no greater than 5, the number of data identifiers within a possible identifier group can be between 1 and 5. When the value is greater than 5, the number of data identifiers within the possible identifier group can be taken as follows: arrive In this case, five preset calculation weights can be determined, each corresponding to one of the five possible numbers of data identifiers within a possible identifier group. It should be understood that generally, the higher the number of data identifiers within a possible identifier group, the greater the preset calculation weight.

[0070] Once all possible identifier groups are determined, the call frequency data of the data to be stored can be calculated. The call frequency data is equal to the sum of all results obtained by multiplying the call frequency components of all possible identifier groups by their corresponding preset calculation weights, which is the weighted sum of the call frequency components determined based on the preset calculation weights.

[0071] Data volume reflects the size of the data to be stored, measured in bytes (B), but can also be expressed in other units such as KB, MB, GB, TB, etc. The data volume is contained within the data identification information, typically represented by a data identifier within that information.

[0072] The specific method for determining the importance coefficient based on data identification information includes: determining the importance component of each data identifier in the data identification information; and calculating the importance coefficient based on all importance components. In the formula, z is the importance coefficient. This represents the i-th importance component. p is the number of importance components. This indicates taking the maximum value among all importance components, where e is the natural constant.

[0073] In this embodiment, a pre-constructed importance lookup table is provided, which includes the correspondence between each data identifier and its importance component. All importance components are preset constants. Therefore, when the data identifier information of the data to be stored is known, the importance component of each data identifier in the data identifier information can be determined.

[0074] In this step, the method for determining the reliability coefficient based on the optional storage strategy specifically includes: obtaining a pre-acquired reliability component relative to each available database identifier; and calculating the reliability coefficient based on the reliability component of the available database identifier in the optional storage strategy. In the formula, w is the reliability coefficient. This represents the reliability component of the i-th available database in the optional storage strategy. q represents the number of available database identifiers in the optional storage strategy. This indicates taking the maximum value among all reliability components, where e is a natural constant.

[0075] Similarly, a reliability lookup table can be pre-built, which includes the relationship between each database identifier and a reliability component, that is, it also includes the relationship between available database identifiers and reliability components, where the reliability components are all preset constants.

[0076] The method for determining call speed data based on the optional storage strategy specifically includes: obtaining the database call speed pre-acquired relative to each available database identifier; and determining the maximum value of the database call speed corresponding to the available database identifier of the optional storage strategy as the call speed data.

[0077] In this embodiment, based on data access to large datasets, the database access speed is determined according to the average access speed (the amount of data that can be accessed per unit time) of the sub-database over a preset period of time. That is, the database access speed is equal to the average access speed of the corresponding sub-database over a preset period of time. It should be understood that when the performance parameters of the sub-database are relatively fixed, the database access speed corresponding to the database identifier can also be preset based on the performance parameters of the corresponding sub-database.

[0078] In this step, determining the score of the optional strategy based on the call frequency data, data volume, call speed data, importance coefficient, and reliability coefficient includes: calculating a first influence score based on the call frequency data, data volume, and call speed data. In the formula, g is the first impact score, f is the call frequency data, b is the data volume, and v is the call speed data; the optional strategy score is calculated based on the first impact score, importance coefficient, and reliability coefficient. In the formula, y is the optional strategy score, and a is the preset adjustment value. 'a' is a constant.

[0079] In this step, determining an optional storage strategy with a score not lower than the storage strategy threshold as the selected storage strategy includes: determining an optional storage strategy with a score not lower than the storage strategy threshold as a candidate storage strategy; and determining the backup storage strategy with the lowest optional strategy score as the selected storage strategy.

[0080] Based on the above, it can be seen that this method can achieve intelligent and reasonable classification and storage of IoT data, ensure the rationality of the selected storage strategy for each piece of data to be stored, and ensure that the stability, reliability and efficiency of data storage and data retrieval can all meet the requirements.

[0081] It should be understood that this method is essentially a machine learning model. The parameters of the machine learning model can be trained based on the data stored with a selected storage strategy, combined with the stability and reliability of data storage (data corruption rate, call failure rate) and the timeliness of data retrieval (actual call speed). The parameters being trained include preset adjustment values, identification strategy comparison tables, identification number comparison tables, weight comparison tables, identification quantity thresholds, identification importance comparison tables, identification reliability comparison tables, etc. In the long-term use, the parameters of this machine learning model are trained, making the classification results of IoT data more intelligent and reasonable.

[0082] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0083] Secondly, this application provides a classification and storage device for Internet of Things (IoT) data. This device is used to perform the method provided in the first aspect of the embodiments of this application above.

[0084] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the described device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0085] In summary, this application has at least the following beneficial effects:

[0086] 1. A method and apparatus for classifying and storing IoT data are provided, which can intelligently and rationally determine the storage strategy of IoT data, which is conducive to reducing costs and increasing efficiency in IoT data management.

[0087] 2. The self-designed and trained policy evaluation model and the algorithm for determining the selected storage policy based on the policy evaluation model are reasonable and accurate, which is conducive to determining the selected storage policy more reasonably and accurately.

[0088] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. An Internet of Things data classified storage method, characterized by, The method comprises the following steps: obtaining data identification information of data to be stored, the data identification information comprising a plurality of data identifications; determining storage condition information of the data to be stored according to the data identification information, the storage condition information comprising a storage quantity and a storage strategy threshold value; determining a selected storage strategy of the data to be stored according to the storage condition information; the step of determining the selected storage strategy of the data to be stored according to the storage condition information comprises: determining a selectable storage strategy according to the storage quantity and the pre-obtained available database identifications, the selectable storage strategy comprising a storage quantity of available database identifications; substituting the data identification information and the selectable storage strategy into a pre-constructed strategy evaluation model to obtain a selectable strategy score, the strategy evaluation model being used to determine the selectable strategy score according to the available database identifications in the selectable storage strategy; determining one of the selectable storage strategies with a selectable strategy score not lower than the storage strategy threshold value as the selected storage strategy; The determination of the storage strategy threshold comprises: obtaining a preset strategy threshold component of each data identifier in the relative data identifier information; and calculating the storage strategy threshold according to the strategy threshold components of all data identifiers in the data identifier information, , wherein h is the storage strategy threshold, represents the strategy threshold component of the i th data identifier in the data identifier information, , s is the number of data identifiers in the data identifier information, represents taking the maximum value in all strategy threshold components, and e is a natural constant; in the step of determining the selectable storage strategy according to the storage quantity and the pre-obtained available database identifications, an identification strategy comparison table of all data identifications and strategy threshold components is pre-constructed, and after the data identification is determined, the data identification is substituted into the identification strategy comparison table to determine the corresponding strategy threshold component, the strategy threshold component being a constant.

2. The method of claim 1, wherein, the step of substituting the data identification information and the selectable storage strategy into the pre-constructed strategy evaluation model to obtain the selectable strategy score comprises: determining calling frequency data, data quantity and importance coefficient according to the data identification information; determining reliability coefficient and calling speed data according to the selectable storage strategy; determining the selectable strategy score according to the calling frequency data, the data quantity, the calling speed data, the importance coefficient and the reliability coefficient.

3. The method of claim 2, wherein, the step of determining the calling frequency data, the data quantity and the importance coefficient according to the data identification information comprises: determining all possible identification groups according to the data identification information, the possible identification group comprising one or more data identifications in the data identification information; determining a calling frequency component for each possible identification group based on data calling big data; calculating a weighted sum value of all calling frequency components as the calling frequency data based on a pre-designed weight of each possible identification group.

4. The method of claim 2, wherein, the step of determining the calling frequency data, the data quantity and the importance coefficient according to the data identification information comprises: determining an importance component of each data identification in the data identification information; said importance coefficient is calculated from all importance components, where z is the importance coefficient, denotes the i-th importance component, p is the number of importance components, denotes the maximum value taken from all importance components, and e is the natural constant.

5. The method of claim 2, wherein, the step of determining the reliability coefficient and the calling speed data according to the selectable storage strategy comprises: obtaining a pre-obtained reliability component relative to each available database identification; a reliability coefficient is calculated from the reliability components identified by the available databases in the alternative storage strategy, where w is the reliability coefficient, represents the reliability component of the i-th available database in the alternative storage strategy, q is the number of available database identifications in the alternative storage strategy, represents the maximum value among all the reliability components, and e is the natural constant.

6. The method of claim 2, wherein, the step of determining the reliability coefficient and the calling speed data according to the selectable storage strategy comprises: obtaining a pre-obtained database calling speed relative to each available database identification; determining a maximum value of the database calling speed corresponding to the available database identification of the selectable storage strategy as the calling speed data.

7. The method of claim 2, wherein, the step of determining the selectable strategy score according to the calling frequency data, the data quantity, the calling speed data, the importance coefficient and the reliability coefficient comprises: The first influence score is calculated according to the calling frequency data, the data volume and the calling speed data, , wherein g is the first influence score, f is the calling frequency data, b is the data volume, and v is the calling speed data. The optional strategy score is calculated according to the first influence score, the importance coefficient and the reliability coefficient, , wherein y is the optional strategy score, and a is a preset adjustment value, .

8. The method of claim 1, wherein, the step of determining one of the selectable storage strategies with a selectable strategy score not lower than the storage strategy threshold value as the selected storage strategy comprises: determining a selected storage policy from the one or more candidate storage policies, wherein the selected storage policy is determined based on a comparison of a candidate policy score of each candidate storage policy to the storage policy threshold value. determining a selected storage policy from the one or more candidate storage policies, wherein the selected storage policy is determined based on a comparison of a candidate policy score of each candidate storage policy to the storage policy threshold value.

9. A device for classified storage of Internet of Things data, characterized in that for performing a method as claimed in any one of claims 1 to 8.

Images