A method and system for providing terminal data services across cloud platforms
By using a data coordination module to determine and manage terminal data on a shared platform, and by monitoring and analyzing the data retrieval process in real time, data sharing is optimized, solving the problems of resource waste and redundant processing in multi-cloud environments, and achieving efficient cross-cloud platform data services.
Patent Information
- Application Number
- CN202611222449.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-12
- Publication Date
- 2026-11-10
AI Technical Summary
In a multi-cloud environment, the terminal data mounting and sharing mechanism lacks effective coordination between its static nature and the dynamic nature of business processing on various cloud platforms, resulting in a waste of computing resources and storage bandwidth. Furthermore, it cannot utilize the processing results of the platforms that have already made the request for data updates or additions, thus reducing the cost of data acquisition and processing.
The data coordination module determines whether terminal data can be mounted and shared, generates required object data, and manages it uniformly on the mounting and sharing platform. It monitors the data retrieval process in real time, generates and analyzes retrieval tracking data, optimizes the data sharing process, and grants access to retrieve processed data.
It reduces redundant computation overhead, shortens end-to-end latency, makes full use of the time difference between cloud platforms, improves data acquisition efficiency, and reduces storage overhead and network transmission burden.
Smart Images

Figure CN123074777A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of terminal data service technology, specifically to a method and system for providing terminal data services across cloud platforms. Background Technology
[0002] With the rapid popularization of cloud computing technology and the widespread application of multi-cloud deployment strategies, the massive amounts of data generated by various terminal devices such as IoT sensors, mobile smart terminals, and industrial data acquisition nodes need to be efficiently uploaded, stored, and shared and scheduled among multiple cloud platforms. Currently, a common cross-cloud data service architecture is to deploy a unified scheduling and storage platform. Each terminal device uploads data to this platform for centralized storage and management. The scheduling and storage platform is connected to multiple cloud platforms and acts as the central hub for data flow. It is responsible for distributing terminal data to the corresponding cloud platforms according to their business needs, so as to support the operation of their upper-layer business applications or computing tasks. In a multi-cloud sharing scenario, when the data of a terminal device is frequently requested by multiple cloud platforms, the existing technology usually adopts the method of mounting the terminal data to the shared data platform, so that each cloud platform can pull it through network mounting protocols such as NFS, CIFS or object storage mounting. The life cycle of the mounted data usually extends until all registered requesting platforms have completed data consumption. During this period, the data content remains unchanged and will not be updated due to any external processing. However, in practical applications, the above-mentioned mounting and sharing mechanism lacks effective coordination between the static nature of the mounted data and the dynamic nature of the business processing of each cloud platform. After pulling the raw mounted data, each cloud platform will process it according to its own business logic, such as performing data filtering, format normalization, dimension alignment, feature calculation, semantic annotation, etc., to form a dataset that meets the needs of its own business. These processing processes are customized for specific demand platform business scenarios. Therefore, the processed data is usually more in line with the internal data model and interface specifications of the demand platform. However, this does not mean that the data is not more applicable to other demand platforms. Different cloud platforms may have the same business background, data usage methods, and quality requirements. The processed data may be easier for some demand platforms to process independently. However, in the existing technology, the processing results produced by the cloud platform that completes data consumption first will not be fed back or written back to the shared mounting point. The cloud platform that needs the terminal data later can only obtain the raw data without any business processing. It has to repeat similar or even the same preprocessing steps, resulting in a waste of computing resources and storage bandwidth. Furthermore, in a multi-cloud environment, the demand for data from the same terminal by different cloud platforms always has a sequential order. This natural time difference provides potential opportunities for optimizing shared data. Specifically, the cloud platform that acquires data earlier has already completed specific business processing. Its processing results may include several highly general intermediate steps, such as data deduplication, outlier removal, and unified unit conversion. These intermediate results have certain reference value for subsequent demand platforms. Even if they cannot completely replace its complete business processing flow, they can at least reduce some repetitive work. However, in the current technical solution, the mounted data remains in its original static state throughout its entire lifecycle. It is impossible to use the processing results of the demand platform to update or supplement the mounted data. This keeps the shared data in its initial original form and cannot gradually accumulate valuable processing information as the number of times it is consumed increases, thereby reducing the data acquisition and processing costs of subsequent demand platforms. To address the above problems, this invention proposes a solution. Summary of the Invention
[0003] The purpose of this invention is to provide a terminal data service method and system across cloud platforms in order to solve the problems mentioned in the background art.
[0004] This invention provides a method for providing terminal data services across cloud platforms, comprising the following steps: Step 1: After receiving terminal data collected by any authorized terminal device according to its preset collection cycle, the data coordination module determines whether the terminal data needs to be mounted and shared. If it is determined that mounting and sharing is required, the module synchronously generates the demand object data of the terminal data. The demand object data contains several demand platforms and their corresponding demand business items and demand cycles. Step 2: After receiving any terminal data and its request object data from any authorized terminal device, the mounting and sharing platform mounts and shares the terminal data according to the preset mounting and sharing rules of the authorized terminal device based on the request object data. During the mounting and sharing period, the mounting status of any terminal data is valid. Step 3: The tracking and acquisition module monitors the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is pulled by any demand platform, the pull tracking data of the terminal data for this pull is generated and stored. After any demand platform pulls the terminal data, it will process it according to the preset processing rules and output the corresponding processed data. Step 4: After the number of retrieved tracking data for all terminal data of any authorized terminal device stored in the tracking and acquisition module reaches a preset fixed number, the data analysis module analyzes all retrieved tracking data of all terminal data of the stored authorized terminal device to determine all demand sets of the authorized terminal device. Step 5: After receiving all the demand sets of any authorized terminal device, the mounting and sharing platform optimizes the mounting and sharing process of the terminal data after receiving any terminal data and demand object data of the authorized terminal device. During the optimization process, it opens pull permissions to all demand platforms within the demand object data, and after several demand platforms process the terminal data and output corresponding processed data, it retrieves the data. Based on all the retrieved processed data, it opens pull permissions to several demand platforms.
[0005] Furthermore, in step one, each authorized terminal device collects terminal data once every corresponding preset collection cycle and transmits it to the data coordination module for storage in real time. The data coordination module also pre-stores all the demand platforms of several authorized terminal devices, as well as the demand cycles of all demand business items deployed in all the demand platforms of each authorized terminal device. The demand platforms are determined from each authorized cloud platform, and the demand business items are determined from all the business items deployed in the authorized cloud platform after it is determined to be a demand platform.
[0006] Furthermore, in step one, the determination of whether the terminal data needs to be mounted and shared is as follows: Obtain the preset collection period of the credit terminal device corresponding to the terminal data, and synchronously obtain all the demand platforms of the credit terminal device stored therein, as well as the demand period of all demand business items deployed by the credit terminal device in all its demand platforms. Based on the collection cycle and the demand cycles of all demand business items deployed within all demand platforms, determine all demand platforms and demand business items that need to retrieve the terminal data. Count the total number of demand platforms that need to retrieve the terminal data. If the determined total number is greater than or equal to a preset demand quantity threshold, it is determined that the terminal data needs to be mounted and shared. At this time, demand object data for the terminal data is generated based on the determined demand platforms, demand business items, and demand business item demand cycles. Otherwise, it is determined that the terminal data does not need to be mounted and shared. At this time, a direct retrieval command is sent to all demand platforms that need to retrieve the terminal data. After receiving the direct retrieval command, the corresponding demand platform directly retrieves the terminal data from the data coordination module.
[0007] Furthermore, in step two, the mounting and sharing rules for the terminal data are as follows: Based on the data of all demand platforms, demand business items, and demand cycles of the demand business items within the demand object data, the terminal data retrieval permission is granted to all demand platforms. The duration of the demand cycle with the largest value among all demand business items is selected as the effective mounting and sharing duration of the terminal data. Starting from the moment the terminal data is mounted and shared, after the effective mounting and sharing duration, the mounting status of the terminal data is changed to invalid mounting, and the terminal data is automatically canceled from mounting and sharing. After granting the terminal data retrieval permission to any demand platform, the demand platform can directly retrieve the terminal data from the mounting and sharing platform.
[0008] Furthermore, in step three, the content of the pull tracking data obtained from a single pull of the terminal data is as follows: The system obtains the request platform and the request business item for this retrieval of terminal data. After the request platform processes the terminal data according to the preset processing rules of the request business item based on the terminal data and outputs the processed data, the system obtains the processed data, the processing duration, and the retrieval time. Based on the obtained processed data, processing duration, retrieval time, request platform, and request business item, the system generates retrieval tracking data for one retrieval of terminal data.
[0009] Furthermore, the content of the set of all requirements for any authorized terminal device is as follows: S11: The terminal data of the authorized terminal device stored in the data analysis module are sequentially labeled as A1, A2, ..., Aa, where a≥1, according to the order of storage. S12: Obtain all pull tracking data of terminal data A1 from the tracking and acquisition module and extract the data. During the extraction process, extract the demand platform and demand business item from each pull tracking data. Use the demand platform as the key and the demand business item as the value to construct the demand dictionary corresponding to the pull tracking data. Add the constructed demand dictionaries corresponding to all pull tracking data into an empty set to obtain the demand set of terminal data A1. S13: Obtain the demand sets of terminal data A2, A3, ..., Aa respectively according to S12, remove duplicates from the demand sets of terminal data A1, A2, ..., Aa, and mark all the remaining demand sets after deduplication as C1, C2, ..., Cc respectively, where c≥1; S14: Traverse the demand sets of terminal data A1, A2, ..., Aa, obtain all terminal data that correspond to the demand set consistent with demand set C1, and re-mark them as D1, D2, ..., Dd, where 1≤d≤a; S15: Copy the demand set C1 once to obtain a replica set of demand set C1, and bind all demand dictionaries in the replica set according to all the pull tracking data of terminal data D1 and their corresponding demand dictionaries; S16: Copy the demand set C1 again d-1 times to obtain d-1 replica sets of demand set C1. Bind the d-1 replica sets to the terminal data D2, D3, ..., Dd respectively according to S15. S17: First, determine all the requirement dictionaries that need to be data bound in the requirement set C1 according to the preset filtering and binding rules, and filter to obtain the corresponding dependency requirement sets that should be bound. After filtering, perform data binding on the corresponding requirement dictionaries in the requirement set C1. S18: Perform data binding on several requirement dictionaries within the requirement sets C2, C3, ..., Cc according to S13 to S17, and transfer all requirement sets that have completed data binding to the mounting and sharing platform for storage.
[0010] Furthermore, in S15, the specific details of data binding for all requirement dictionaries within the replica set are as follows: SS21: According to the order of storage, the pull tracking data of all pulls of terminal data D1 are sequentially marked as E1, E2, ..., Ee, where e is the total number of times terminal data D1 is pulled when the mounting status is valid; SS22: Extract the processing data sequentially from the retrieved tracking data E1, E2, ..., Ee, and label them as F1, F2, ..., Fe; SS23: Create a non-transformable variable G1 for processing data F1, with the initial value of the non-transformable variable G1 being 0; The processing data F1 is selected as the conversion source. Based on the processing data F2, it is selected whether to perform a preset gain direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. If it is selected to perform a preset pull gain direction determination on the conversion source, and the pull gain direction is determined to be positive after the determination is completed, then the demand dictionary corresponding to the pull tracking data F2 is selected as the auxiliary demand dictionary of the demand dictionary of the pull tracking data F1. SS24: According to SS23, based on the processed data F3, F4, ..., Fe, select whether to perform a preset gain determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. After all the executions are completed, obtain all the subordinate demand dictionaries of the demand dictionary corresponding to the pull tracking data F1, and classify them into an empty set to obtain the subordinate demand set of the demand dictionary corresponding to the pull tracking data F1. SS25: After step SS24 is completed, obtain the value J1 of the non-convertible variable G1 at this time, and count the number J2 and J3 of positive and negative gain directions after step SS24 is completed; Using formula Calculate the binding tendency index K1 of the demand dictionary corresponding to the pull tracking data E1 relative to the terminal data D1, and associate the binding tendency index K1 with the demand dictionary in the replica set that is consistent with the demand dictionary corresponding to the pull tracking data E1. Simultaneously associate the demand dictionary with the auxiliary demand set corresponding to the pull tracking data E1. N is the total number of judgments, which is the sum of J1, J2, and J3. α1, α2, and α3 are the preset variable proportion coefficients, positive and negative proportion coefficients, respectively. SS26: Create non-convertible variables for processing data F2, F3, ..., Fe respectively, select processing data F2, F3, ..., Fe as the transformation source respectively, and perform data binding on all requirement dictionaries in the replica set according to SS21 to SS25 respectively.
[0011] Furthermore, in SS23, the selection based on the processing data F2 regarding whether to perform a preset gain-direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1 is as follows: Select the processed data F2 as the target source of the conversion source, and obtain the demand dictionaries corresponding to the pull tracking data E1 and E2, which contain the conversion source and the target source respectively. Use the two demand dictionaries as search conditions to search in the data analysis module to see if there is a pre-stored conversion list. The conversion dictionaries with indices 0 and 1 in the list are consistent with the demand dictionaries corresponding to the pull tracking data E1 and E2 respectively. If they exist, select to perform a preset pull gain direction judgment on the conversion source. If they do not exist, select to perform an auto-increment operation on the value of the non-convertible variable G1. If a preset pull gain direction determination is performed on the selected conversion source, then the pull gain direction determination is performed on it, and the determination content is as follows: Obtain the conversion rules corresponding to the conversion list pre-stored in the data analysis module, convert the conversion source according to the conversion rules, and after the conversion is completed, obtain the conversion time of the conversion source relative to the target source, that is, the conversion time H2 of converting the processed data F1 to the processed data F2; The pull gain value I1 of the conversion source relative to the target source is calculated using the formula I1=H1+H2-(H3+H4)-P1, where H1 is the processing time within the pull tracking data including the conversion source, H3 is the pull interval time obtained by subtracting the absolute value of the pull times in the pull tracking data E1 and E2, H4 is the processing time within the pull tracking data E1, and P1 is a preset time compensation constant. If I1≥0, the pull gain direction of the conversion source relative to the target source is determined to be positive, and the demand dictionary corresponding to the pull tracking data F2 is selected as the auxiliary demand dictionary of the pull tracking data F1. Otherwise, no processing is performed.
[0012] Furthermore, in S17, the data binding of several requirement dictionaries within the requirement set C1 is as follows: SS31: Label all requirement dictionaries contained in requirement set C1 as L1, L2, ..., Ll, l≥1, where l is the total number of requirement dictionaries in requirement set C1; SS32: After the data binding of the d replica sets of the demand set C1 is completed, obtain the binding tendency index of the d replica sets that are associated with the demand dictionary L1. Use the discrete point filtering algorithm to calculate all the obtained binding tendency indices for data processing, obtain the average value of all remaining binding tendency indices after data processing, and use the average value as the mounting peer average value of the demand dictionary L1. SS33: Compare the mean value of the mounting peers with a preset mounting peer threshold. If the mean value of the mounting peers is greater than or equal to the mounting peer threshold, it is determined that the requirement dictionary L1 needs to be data bound. At this time, obtain the auxiliary requirement sets that are associated with the requirement dictionary L1 in each of the d replica sets, and obtain a total of d auxiliary requirement sets. Count the number of times each auxiliary requirement dictionary appears in these d auxiliary requirement sets, and calculate the ratio of the number of times each auxiliary requirement dictionary appears to d. The proportion value of each auxiliary requirement dictionary is compared with the preset proportion threshold one by one. All auxiliary requirement dictionaries with a proportion value greater than or equal to the preset proportion threshold are retained. All retained auxiliary requirement dictionaries are put into an empty set. This set is used as the dependent requirement set that requirement dictionary L1 is bound to. Data binding is performed between requirement dictionary L1 and requirement dictionary L1 in requirement set C1. Conversely, if the average value of the mount peers is less than the mount peer threshold, no action is taken. SS34: Determine whether data binding is required for the requirement dictionaries L2, L3, ..., L1 within the requirement set C1 according to SS31 to SS33, and perform data binding on several requirement dictionaries within the requirement set C1 based on the determination results.
[0013] A cross-cloud platform terminal data service system, characterized in that it includes: The data coordination module is used to receive and store the terminal data periodically transmitted back by each authorized terminal device. Specifically, each authorized terminal device collects terminal data once every corresponding preset collection cycle and transmits it to the data coordination module in real time. The data coordination module is also used to determine whether the terminal data needs to be mounted and shared after receiving the terminal data periodically returned by each authorized terminal device. The authorized terminal device refers to various terminal nodes that have been authorized to access the system after identity authentication and permission verification. The data coordination module pre-stores all the demand platforms for each authorized terminal device, as well as the demand cycles for all demand business items deployed on each authorized terminal device relative to its respective demand platforms. The mounting and sharing platform is used to mount and share the terminal data according to the preset mounting and sharing rules based on the data of the requested object after receiving terminal data and its request object. The tracking and acquisition module is used to monitor the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is detected to be pulled, the module obtains the requesting platform and the requesting business item for pulling the terminal data. After the requesting platform processes the terminal data according to the preset processing rules of the requesting business item based on the terminal data and outputs the processed data, the module obtains the processed data, the processing time, and the time of this pull. Based on the obtained processed data, processing time, and pull time, the module generates and stores the pull tracking data for one pull of the terminal data. The data analysis module is used to analyze all the retrieved tracking data of all terminal data of the stored credit terminal device after the number of retrieved tracking data of any credit terminal device stored in the tracking and acquisition module reaches a preset fixed number, and to determine all the demand sets of the credit terminal device. The mounting and sharing platform is also used to optimize the mounting and sharing process of terminal data after receiving all the demand sets of the authorized terminal device and any terminal data and demand object data of the authorized terminal device received subsequently. During the optimization process, pull permissions are granted to all demand platforms in the demand object data, and after several demand platforms process the terminal data and output corresponding processed data, the data is obtained. Based on all the obtained processed data, pull permissions are granted to several demand platforms.
[0014] Compared with existing technologies, it has the following advantages: This invention receives and stores the periodically transmitted terminal data from various authorized terminal devices through a data coordination module. It then determines whether the periodically transmitted terminal data should be mounted and shared. Based on the determination result, it selectively transmits several terminal data to a mounting and sharing platform, which then mounts and shares the data. This method allows terminal data with high demand and frequently retrieved by multiple authorized cloud platforms to be placed in the same mounting and sharing platform, where unified mounting, sharing, and permission distribution are handled, reducing the concurrent retrieval pressure on the data coordination module. Simultaneously, terminal data with low demand, needed only by a few authorized cloud platforms, is retained in the data coordination module, which directly manages its retrieval permissions, avoiding the waste of shared storage resources caused by low-frequency data mounting. Through this distribution strategy of centralized sharing of high-frequency data and on-demand direct connection of low-frequency data, the overall system storage overhead and network transmission burden are significantly reduced while ensuring the data acquisition efficiency of each demanding platform. This invention uses a tracking and acquisition module to monitor the data retrieval process of each terminal mounted and shared on the mounting and sharing platform in real time. Each time data is retrieved, the system obtains the processed data and processing time output by the corresponding requesting platform after processing the terminal data. Combined with the retrieval time, the corresponding requesting platform, and the requesting business item, corresponding retrieval tracking data is generated. This data is then analyzed by a data analysis module. During the analysis, based on parameters such as the processed data, processing time, and retrieval time within each retrieval tracking data set, data binding operations are performed sequentially to quantitatively assess whether there is dependency and reuse value between different requesting platforms. Subsequently, when the terminal data of each authorized terminal device is determined to need to be mounted and shared, the system not only opens the original data to each requesting platform... In addition to granting access to terminal data, the system also grants access to the corresponding processed data to demand platforms that have been identified as having a positive dependency relationship through data binding within the demand set. This allows subsequent demand platforms that need to pull terminal data to directly obtain the processing results that have already undergone preprocessing operations such as cleaning, transformation, and aggregation by the first platform, provided that their own business allows it, without having to repeat the same processing steps from the original terminal data. This significantly reduces the overhead of repeated calculations on the same terminal data by various authorized cloud platforms, shortens the end-to-end latency from data retrieval to final usability, and fully leverages the time difference advantage between cross-cloud platforms, enabling the mounted shared data to continuously accumulate high-value intermediate processing results as the number of times it is consumed increases. Attached Figure Description
[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system block diagram of the present invention. Detailed Implementation
[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0017] Please see Figure 1 This application provides a method for providing terminal data services across cloud platforms, the method comprising the following steps: Step 1: After receiving terminal data collected by any authorized terminal device according to its preset collection cycle, the data coordination module determines whether the terminal data needs to be mounted and shared. If it is determined that mounting and sharing is required, the module synchronously generates the demand object data of the terminal data. The demand object data contains several demand platforms and their corresponding demand business items and demand cycles. Step 2: After receiving any terminal data and its request object data from any authorized terminal device, the mounting and sharing platform mounts and shares the terminal data according to the preset mounting and sharing rules of the authorized terminal device based on the request object data. During the mounting and sharing period, the mounting status of any terminal data is valid. Step 3: The tracking and acquisition module monitors the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is pulled by any demand platform, the pull tracking data of the terminal data for this pull is generated and stored. After any demand platform pulls the terminal data, it will process it according to the preset processing rules and output the corresponding processed data. Step 4: After the number of retrieved tracking data for all terminal data of any authorized terminal device stored in the tracking and acquisition module reaches a preset fixed number, the data analysis module analyzes all retrieved tracking data of all terminal data of the stored authorized terminal device to determine all demand sets of the authorized terminal device. Step 5: After receiving all the demand sets of any authorized terminal device, the mounting and sharing platform optimizes the mounting and sharing process of the terminal data after receiving any terminal data and demand object data of the authorized terminal device. During the optimization process, it opens pull permissions to all demand platforms in the demand object data, and after several demand platforms process the terminal data and output the corresponding processed data, it retrieves the data. Based on all the retrieved processed data, it opens pull permissions to several demand platforms. Please see Figure 2 This application provides a cross-cloud platform terminal data service system, which includes a data coordination module, a mounting and sharing platform, a tracking and collection module, and a data analysis module; The data coordination module is used to receive and store the terminal data periodically transmitted back by each authorized terminal device. Specifically, each authorized terminal device collects terminal data once every corresponding preset collection cycle and transmits it to the data coordination module in real time. The data coordination module is also used to determine whether the terminal data needs to be mounted and shared after receiving the terminal data periodically returned by each authorized terminal device. The authorized terminal device refers to various terminal nodes that have been authenticated and authorized by the system and have been authorized to access the system, including but not limited to IoT devices, edge computing nodes, mobile terminals, server terminals and other device entities with data generation and transmission capabilities. The terminal data refers to structured or unstructured data with business value collected or generated by each authorized terminal device during operation, including but not limited to device operation status data, environmental perception data, business log data, user operation behavior data, etc. The preset collection period for each authorized terminal device is determined based on the business needs of the authorized cloud platform that needs to use the terminal data of each authorized terminal device, as well as the attributes of the collected terminal data itself. In this application, the collection periods for each authorized terminal device are not the same. The data coordination module pre-stores all the demand platforms for each authorized terminal device, as well as the demand cycles for all demand business items deployed on each of its respective demand platforms. It's important to note that the pre-stored demand cycles for each authorized terminal device relative to any single demand platform are different. This is because the demand cycles for multiple demand business items deployed on the same authorized cloud platform may be consistent. During data mounting and sharing, a single pull operation is performed from the mounting and sharing platform to obtain the terminal data. This allows for data reuse across multiple demand business items within the platform, eliminating the need to initiate separate pull requests for each business item. Therefore, when storing the mapping relationships, the data coordination module only needs to randomly retain one demand business item with consistent demand cycles as a representative. In this application, the demand platform is determined from each credit cloud platform, and the demand business items are determined from all business items deployed within the demand platform after the demand platform is determined. The "trusted cloud platform" refers to various cloud service platforms that have been authenticated and authorized to access the system after system identity verification and permission checks, including but not limited to public cloud platforms, private cloud platforms, hybrid cloud platforms, and industry-specific cloud platforms. The "trusted cloud platform" is the consumer of terminal data, and it has several business items deployed and running internally. Each business item relies on the terminal data uploaded by the trusted terminal device as the input data source to support the normal execution of its own business logic. The business item refers to a business function unit deployed and running on the credit cloud platform, possessing independent data processing logic, and capable of independently providing corresponding functional services to the outside world. During operation, the business function unit relies on terminal data uploaded by the credit terminal devices as input data source to support its own business logic execution such as calculation, monitoring, analysis, alarm, statistics, and display. Each business item has a preset retrieval period for the terminal data of each type of credit terminal device it depends on, which is used to indicate that the business item needs to obtain the corresponding terminal data from the data coordination module or the mounted sharing platform once every retrieval period. Each credit cloud platform has several pre-set processing rules based on any business item deployed within it. Each processing rule is used to process the terminal data of a credit terminal device required by the business item. The processing rule refers to a set of pre-set data processing logics for any business item within the credit cloud platform to transform the original terminal data into the input data format required by its own business logic. Each processing rule corresponds to one or more processing operations, including but not limited to data cleaning, format conversion, field mapping, data aggregation, anomaly removal, unit conversion, and threshold comparison, for terminal data provided by a credit terminal device required by the business item. Different or the same processing rules can be preset for terminal data provided by different credit terminal devices for the same business item. Only one processing rule is preset for the same type of terminal data provided by the same credit terminal device for the same business item. The content of any demand platform for any authorized terminal device and a demand service item of the authorized terminal device relative to the demand platform is determined as follows: The terminal data required for all business items deployed in all credit cloud platforms are traversed. If, among all credit cloud platforms, there is a business item carried in a certain credit cloud platform that requires the terminal data of the credit terminal device as an input data source, then the credit cloud platform is determined to be the demand platform of the credit terminal device, and the business item is simultaneously determined to be a demand business item of the demand platform relative to the credit terminal device. The determination of whether data from any terminal needs to be mounted and shared is as follows: Obtain the preset collection period of the credit terminal device corresponding to the terminal data, and synchronously obtain all the demand platforms of the credit terminal device stored therein, as well as the demand period of all demand business items deployed by the credit terminal device in all its demand platforms. Based on the collection cycle and the demand cycle of all demand business items deployed in all demand platforms, determine all demand platforms and demand business items that need to pull the terminal data, count the total number of demand platforms that need to pull the terminal data, and if the total number is greater than or equal to a preset demand quantity threshold, determine that the terminal data needs to be mounted and shared. At this time, generate demand object data for the terminal data according to the determined demand platforms, demand business items and demand cycle of the demand business items that need to pull the terminal data, and transmit the terminal data and its demand object data together to the mounting and sharing platform. Conversely, if the terminal data does not need to be mounted and shared, a direct pull command is sent to all the demand platforms that need to pull the terminal data. After receiving the direct pull command, each demand platform directly obtains the terminal data from the data coordination module. It should be noted that any fetching period is an integer multiple of the collection period of the corresponding authorized terminal device; The mounting and sharing platform is used to mount and share the terminal data according to preset mounting and sharing rules based on the data of the requesting object after receiving each terminal data and its requesting object. The mounting and sharing content is as follows: Based on the data of all demand platforms, demand business items, and demand cycles of the demand business items within the demand object data, the terminal data retrieval permission is granted to all demand platforms. The duration of the demand cycle with the largest value among all demand business items is selected as the effective mounting and sharing duration of the terminal data. Starting from the moment the terminal data is mounted and shared, after the effective mounting and sharing duration, the mounting status of the terminal data is changed to invalid mounting. When the mounting status of the terminal data is changed to invalid mounting, the retrieval permission of the terminal data by all demand platforms is closed, and the terminal data is automatically canceled from mounting and sharing. After granting access to the terminal data to any requesting platform, the requesting platform can directly retrieve the terminal data from the mounted sharing platform; In this application, the mounting status is divided into two states: valid mounting and invalid mounting. When the mounting status of any terminal data is valid mounting, it means that it is mounted and shared, and the corresponding demand platform can pull it. When it is invalid mounting, it means that it has been canceled from mounting and sharing, and the corresponding demand platform does not have the right to pull it. The tracking and acquisition module is used to monitor the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is detected to be pulled, the module obtains the requesting platform and the requesting business item for pulling the terminal data. After the requesting platform processes the terminal data according to the preset processing rules of the requesting business item based on the terminal data and outputs the processed terminal data, the module obtains the processed data, the processing duration, and the time of this pull. The processing duration refers to the time spent from the start of processing the terminal data to the output of the processed data, and the pull time refers to the time when the terminal data is detected to be pulled. It should be noted that if any demand platform pulls the terminal data once based on any demand business item, then this pull corresponds to the demand platform and the demand business item. The tracking and acquisition module generates and stores the pull tracking data for one pull of terminal data based on the acquired processing data, processing time, pull time, demand platform, and demand business item; It should be noted that after the demand platform completes the processing, it will send the processed data, the corresponding processing time, and the retrieval time back to the tracking and collection module. Alternatively, the tracking and collection module can send a request to the demand platform to obtain the processing results after detecting a retrieval event. The data analysis module is used to analyze all the pulled tracking data of all terminal data of the stored credit terminal device after the number of pulled tracking data of any credit terminal device stored in the tracking and acquisition module reaches a preset fixed number, and to determine all the demand sets of the credit terminal device. The fixed number is configured by the operation and maintenance management personnel based on the frequency of credit terminal business interaction, data fluctuation characteristics, and demand identification accuracy requirements, and serves as the minimum sample threshold for starting batch data analysis. The quantity here refers to the sum of all retrieved and tracked data from all terminals of the authorized terminal device; Analyzing all the pull tracking data of all terminal data of any stored authorized terminal device, the following content is generated for the set of all requirements of the authorized terminal device: S11: The terminal data of the authorized terminal device stored in the order of storage are sequentially marked as A1, A2, ..., Aa, where a≥1. It should be noted that the marking here is performed on the terminal data of the authorized terminal device stored in the data coordination module. S12: Obtain all pull tracking data from terminal data A1 in the tracking and acquisition module and extract the data. During the extraction process, extract the demand platform and demand business item from each pull tracking data, use the demand platform as the key and the demand business item as the value, and construct the demand dictionary corresponding to the pull tracking data. Add all the required dictionaries corresponding to the pulled tracking data constructed into an empty set to obtain the required set of terminal data A1. S13: Obtain the demand sets of terminal data A2, A3, ..., Aa respectively according to S12. Deduplicate the demand sets of terminal data A1, A2, ..., Aa. During the deduplication process, for multiple demand sets that contain the same demand dictionary, only one demand set is randomly retained. Mark all the remaining demand sets after deduplication as C1, C2, ..., Cc respectively, where c≥1, and c is the total number of remaining demand sets after deduplication. Among them, the requirement dictionaries contained in multiple requirement sets are consistent, referring to any requirement dictionary in any requirement set, and there is one and only one such dictionary in the other requirement sets. It should be noted here that a set is chosen as the basis for storing the required fields because a set is unordered and does not record which field was added first or last. S14: Traverse the demand sets of terminal data A1, A2, ..., Aa, obtain all terminal data whose corresponding demand sets are consistent with demand set C1, and re-mark them as D1, D2, ..., Dd, where 1≤d≤a. When re-marking, mark them in ascending order of the value of the previous marking index. S15: Copy the demand set C1 once to obtain a replica set of demand set C1, and bind all demand dictionaries in the replica set according to all the pull tracking data of terminal data D1 and their corresponding demand dictionaries; The specific details of data binding are as follows: SS21: According to the order in which the data is stored in the data analysis module, the pull tracking data of all pulls of terminal data D1 are sequentially marked as E1, E2, ..., Ee, where e is the total number of times terminal data D1 was pulled when the state was mounted and shared; SS22: Extract the processing data sequentially from the retrieved tracking data E1, E2, ..., Ee, and label them as F1, F2, ..., Fe; SS23: Create a non-transformable variable G1 for processing data F1, with the initial value of the non-transformable variable G1 being 0; The processing data F1 is selected as the conversion source. Based on the processing data F2, it is selected whether to perform a preset gain direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. If it is selected to perform a preset pull gain direction determination on the conversion source, and if the pull gain direction is determined to be positive after the determination is completed, then the demand dictionary corresponding to the pull tracking data E2 is selected as the auxiliary demand dictionary of the demand dictionary of the pull tracking data E1. In this application, the pull gain direction is divided into two types: positive and negative. Specifically, the selection based on the processing data F2—whether to perform a preset gain-direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1—is as follows: Select the processed data F2 as the target source of the conversion source, and obtain the demand dictionaries corresponding to the pull tracking data E1 and E2, which contain the conversion source and the target source respectively. Use the two demand dictionaries as search conditions to search in the data analysis module to see if there is a pre-stored conversion list. The conversion dictionaries with indices 0 and 1 in the list are consistent with the demand dictionaries corresponding to the pull tracking data E1 and E2 respectively. If they exist, select to perform a preset gain direction determination on the conversion source. If they do not exist, select to perform an increment operation of 1 on the value of the non-convertible variable G1. If a preset pull gain direction determination is selected to be performed on the conversion source, then the pull gain direction determination is performed on it, and the determination content is as follows: Obtain the conversion rules corresponding to the conversion list pre-stored in the data analysis module, convert the conversion source according to the conversion rules, and after the conversion is completed, obtain the conversion time of the conversion source relative to the target source, that is, the conversion time H2 of converting the processed data F1 to the processed data F2; The pull gain value I1 of the conversion source relative to the target source is calculated using the formula I1=H1+H2-(H3+H4)-P1, where H1 is the processing time within the pull tracking data including the conversion source, H3 is the pull interval time obtained by subtracting the absolute value of the pull times within the pull tracking data E1 and E2, H4 is the processing time within the pull tracking data E1, and P1 is a preset time compensation constant used to avoid time delays caused by external factors such as network. If I1≥0, it means that fetching the demand platform and demand business items in the demand dictionary corresponding to the trace data E2 is more efficient or on par with fetching the corresponding terminal data from the mounting platform, without waiting for the demand platform and demand business items in the demand dictionary corresponding to the trace data E1 to fetch terminal data from the mounting shared platform, process and convert it, and then upload it. In this case, it is determined that the fetching gain of the conversion source relative to the target source is positive, and the demand dictionary corresponding to the fetch trace data F2 is selected as the auxiliary demand dictionary of the fetch trace data F1. Otherwise, it is determined that the fetching gain of the conversion source relative to the target source is negative, and no processing is performed. If it does not exist, then increment the value of the non-convertible variable G1 by 1; SS24: According to SS23, based on the processed data F3, F4, ..., Fe, select whether to perform a preset gain determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. After all the execution is completed, obtain all the subordinate demand dictionaries of the demand dictionary corresponding to the selected pull tracking data F1 and classify them into an empty set to obtain the subordinate demand set corresponding to pull tracking data F1. It should be noted here that once the selection is made to perform a preset gain direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1, the corresponding operation can be executed. That is, to perform a preset gain direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. SS25: Perform data binding on the requirement dictionary in the replica set that matches the requirement dictionary corresponding to pull trace data E1, as follows: After step SS24 is completed, obtain the value J1 of the non-convertible variable G1 at this time, and count the number of positive and negative pull gain directions J2 and J3 respectively. Using formula Calculate the binding tendency index K1 of the demand dictionary corresponding to the pull tracking data E1 relative to the terminal data D1, and associate the binding tendency index K1 with the demand dictionary in the replica set that is consistent with the demand dictionary corresponding to the pull tracking data E1. Simultaneously, associate the demand dictionary in the replica set that is consistent with the demand dictionary corresponding to the pull tracking data E1 with the auxiliary demand set corresponding to the pull tracking data E1. Here, N is the total number of judgments, which is the sum of J1, J2, and J3. α1, α2, and α3 are the preset variable proportion coefficients, positive and negative proportion coefficients, respectively. Preferred, the values of α1, α2, and α3 are 1, 1, and 0.5, respectively. It should be noted that in the above formula for calculating the mounting equivalence value K1, J1, J2, and J3 are all count variables obtained by statistically analyzing historical data, and they themselves do not have physical dimensions. Therefore, the intermediate variables such as N, J2 / N, and (J3+ɑ3×J1) / N, which are linear combinations or ratios of them, are also dimensionless pure numerical values. The preset weighting coefficients ɑ1, ɑ2, and ɑ3 are all empirical constants set by those skilled in the art based on the actual operating load of the system, and they are also dimensionless. In summary, the mounting equivalence value K1 calculated in this application is a dimensionless evaluation index used to measure the dependence of other demand platforms on the terminal data after processing based on demand business items. Its numerical value is only used for relative comparison and does not involve the dimensional conversion of specific physical quantities. SS26: Create non-convertible variables for processing data F2, F3, ..., Fe respectively, select processing data F2, F3, ..., Fe as the transformation source respectively, and perform data binding on all requirement dictionaries in the replica set according to SS21 to SS25 respectively; S16: Copy the demand set C1 again d-1 times to obtain d-1 replica sets of demand set C1. According to S15, bind all demand dictionaries in the d-1 replica sets according to the terminal data D2, D3, ..., Dd respectively; S17: First, determine all the requirement dictionaries that need to be data bound within the requirement set C1 according to the preset filtering and binding rules, and then filter to obtain the corresponding dependency requirement sets that should be bound. After filtering, perform data binding on the corresponding requirement dictionaries within the requirement set C1. The filtering and binding content is as follows: SS31: Label all requirement dictionaries contained in requirement set C1 as L1, L2, ..., Ll, l≥1, where l is the total number of requirement dictionaries in requirement set C1; SS32: After the data binding of the d replica sets of the demand set C1 is completed, obtain the binding tendency index of the d replica sets that are associated with the demand dictionary L1. Use the discrete point filtering algorithm to calculate all the obtained binding tendency indices for data processing, and obtain the average value of all remaining binding tendency indices after data processing. Use the average value as the mounting peer average value of the demand dictionary L1. In this application, the discrete point filtering algorithm adopts the median absolute deviation method (MAD). It should be noted here that since all d replica sets are copies of the demand set C1, each of the d replica sets contains a demand dictionary L1; SS33: Compare the mean value of the mount peers with the preset mount peer threshold. If the mean value of the mount peers is greater than or equal to the mount peer threshold, then it is determined that the requirement dictionary L1 needs to be data bound. Obtain the subordinate requirement sets that are associated with and bound to the requirement dictionary L1 in each of the d replica sets, resulting in a total of d subordinate requirement sets; count the number of times each subordinate requirement dictionary appears in these d subordinate requirement sets, and calculate the ratio of the number of times each subordinate requirement dictionary appears to d. The proportion of each auxiliary requirement dictionary is compared with the preset proportion threshold one by one. All auxiliary requirement dictionaries with a proportion value greater than or equal to the preset proportion threshold are retained. All retained auxiliary requirement dictionaries are grouped into an empty set, which is used as the set of dependent requirements that requirement dictionary L1 is bound to. Conversely, if the average value of the mount peers is less than the mount peer threshold, no action is taken. SS34: Determine whether data binding is required for the requirement dictionaries L2, L3, ..., L1 in the requirement set C1 according to SS31 to SS33 respectively, and perform data binding on several requirement dictionaries in the requirement set C1 based on the determination results; S18: According to S13 to S17, perform data binding on several demand dictionaries in demand sets C2, C3, ..., Cc respectively, and transmit all demand sets of the authorized terminal devices that have completed data binding to the mounting and sharing platform for storage; It should be noted that the data analysis module only needs to perform one analysis operation when no new credit cloud platform is added, or when no new business items are deployed on any credit cloud platform. It should be noted here that before analyzing all the pull tracking data of all terminal data of the credit terminal device, for each terminal data, according to all the demand platforms and demand business items contained in all the pull tracking data of the terminal data, the preset processing rules of the demand business items based on the terminal data are obtained from all the demand platforms. After the acquisition is completed, they are uniformly displayed to the operation and maintenance management personnel for browsing and viewing. The operation and maintenance personnel determine several conversion lists based on the terminal data. Each conversion list contains two conversion dictionaries. Each conversion list corresponds to a conversion rule written by the operation and maintenance personnel, which is used to convert the conversion dictionary with index 0 in the conversion list into the conversion dictionary with index 1. After receiving all the demand sets of the authorized terminal device, the mounting and sharing platform optimizes the mounting and sharing process of the terminal data after receiving any terminal data and demand object data of the authorized terminal device. During the optimization process, it opens pull permissions to all demand platforms in the demand object data, and after several demand platforms process the terminal data and output corresponding processed data, it retrieves the data and opens pull permissions to several demand platforms based on all the retrieved processed data. The specific steps for optimizing the mounting and sharing process of any subsequent received terminal data from the aforementioned authorized terminal device are as follows: The first step is to obtain all the demand platforms and the demand business items corresponding to each demand platform contained in the demand object data corresponding to the terminal data. Then, retrieve the matching demand set from all the demand sets of the authorized terminal device stored in the shared platform. The matching demand set satisfies the condition that the key of each demand dictionary in the set is the all demand platforms and the value of each demand set is the demand business item corresponding to each demand platform. Extract all demand platforms that are used as keys from all demand dictionaries that are data-bound within the demand set, grant these demand platforms access to pull terminal data, and temporarily store the demand period with the largest value within the demand period of these demand platforms. The second step involves binding any data-bound requirement dictionary within the requirement set. The requirement platform used as a key in the requirement dictionary is designated as a primary platform. From the dependent requirement set corresponding to the requirement dictionary, the requirement platforms used as keys in each subordinate requirement dictionary are extracted and designated as secondary platforms of the primary platform. Pull permissions from the primary platform are granted to all secondary platforms. After granting these permissions, when a primary platform pulls terminal data and outputs processed data, which is then transmitted to the mounting and sharing platform, the mounting and sharing platform mounts and shares it. During the mounting and sharing process, pull permissions for the processed data are granted to all secondary platforms of the primary platform. Each secondary platform can then pull the processed data. Similarly, the effective mounting duration of the processed data is determined based on the requirement cycle of all secondary platform objects. Step 3: After completing Step 1 and Step 2 above, obtain all remaining demand platforms in the demand set that are keys and have not yet been granted access to pull data. Grant access to pull terminal data to all demand platforms and obtain the demand period with the largest value within the demand period of these demand platforms. Compare the value of the demand period with the demand period temporarily stored in Step 1 and use the demand period with the larger value as the effective duration of the terminal data mounting.
[0018] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.
[0019] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.
Claims
1. A method for providing terminal data services across cloud platforms, characterized in that, Includes the following steps: Step 1: After receiving terminal data collected by any authorized terminal device according to its preset collection cycle, the data coordination module determines whether the terminal data needs to be mounted and shared. If it is determined that mounting and sharing is required, the module synchronously generates the demand object data of the terminal data. The demand object data contains several demand platforms and their corresponding demand business items and demand cycles. Step 2: After receiving any terminal data and its request object data from any authorized terminal device, the mounting and sharing platform mounts and shares the terminal data according to the preset mounting and sharing rules of the authorized terminal device based on the request object data. During the mounting and sharing period, the mounting status of any terminal data is valid. Step 3: The tracking and acquisition module monitors the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is pulled by any demand platform, the pull tracking data of the terminal data for this pull is generated and stored. After any demand platform pulls the terminal data, it will process it according to the preset processing rules and output the corresponding processed data. Step 4: After the number of retrieved tracking data for all terminal data of any authorized terminal device stored in the tracking and acquisition module reaches a preset fixed number, the data analysis module analyzes all retrieved tracking data of all terminal data of the stored authorized terminal device to determine all demand sets of the authorized terminal device. Step 5: After receiving all the demand sets of any authorized terminal device, the mounting and sharing platform optimizes the mounting and sharing process of the terminal data after receiving any terminal data and demand object data of the authorized terminal device. During the optimization process, it opens pull permissions to all demand platforms within the demand object data, and after several demand platforms process the terminal data and output corresponding processed data, it retrieves the data. Based on all the retrieved processed data, it opens pull permissions to several demand platforms.
2. The method for providing terminal data services across cloud platforms according to claim 1, characterized in that, In step one, each authorized terminal device collects terminal data once every corresponding preset collection cycle and transmits it to the data coordination module for storage in real time. The data coordination module also pre-stores all the demand platforms of several authorized terminal devices, as well as the demand cycles of all demand business items deployed in all the demand platforms of each authorized terminal device. The demand platforms are determined from each authorized cloud platform, and the demand business items are determined from all the business items deployed in the authorized cloud platform after it is determined to be a demand platform.
3. The method for providing terminal data services across cloud platforms according to claim 2, characterized in that, In step one, the determination of whether the terminal data needs to be mounted and shared is as follows: Obtain the preset collection period of the credit terminal device corresponding to the terminal data, and synchronously obtain all the demand platforms of the credit terminal device stored therein, as well as the demand period of all demand business items deployed by the credit terminal device in all its demand platforms. Based on the collection cycle and the demand cycles of all demand business items deployed within all demand platforms, determine all demand platforms and demand business items that need to retrieve the terminal data. Count the total number of demand platforms that need to retrieve the terminal data. If the determined total number is greater than or equal to a preset demand quantity threshold, it is determined that the terminal data needs to be mounted and shared. At this time, demand object data for the terminal data is generated based on the determined demand platforms, demand business items, and demand business item demand cycles. Otherwise, it is determined that the terminal data does not need to be mounted and shared. At this time, a direct retrieval command is sent to all demand platforms that need to retrieve the terminal data. After receiving the direct retrieval command, the corresponding demand platform directly retrieves the terminal data from the data coordination module.
4. The method for providing terminal data services across cloud platforms according to claim 1, characterized in that, In step two, the mounting and sharing rules for the terminal data are as follows: Based on the data of all demand platforms, demand business items, and demand cycles of the demand business items within the demand object data, the terminal data retrieval permission is granted to all demand platforms. The duration of the demand cycle with the largest value among all demand business items is selected as the effective mounting and sharing duration of the terminal data. Starting from the moment the terminal data is mounted and shared, after the effective mounting and sharing duration, the mounting status of the terminal data is changed to invalid mounting, and the terminal data is automatically canceled from mounting and sharing. After granting the terminal data retrieval permission to any demand platform, the demand platform can directly retrieve the terminal data from the mounting and sharing platform.
5. The method for providing terminal data services across cloud platforms according to claim 1, characterized in that, In step three, the content of the pull tracking data obtained from a single pull of the terminal data is as follows: The system obtains the request platform and the request business item for this retrieval of terminal data. After the request platform processes the terminal data according to the preset processing rules of the request business item based on the terminal data and outputs the processed data, the system obtains the processed data, the processing duration, and the retrieval time. Based on the obtained processed data, processing duration, retrieval time, request platform, and request business item, the system generates retrieval tracking data for one retrieval of terminal data.
6. The method for providing terminal data services across cloud platforms according to claim 5, characterized in that, The content of the complete set of requirements for any authorized terminal device is as follows: S11: The terminal data of the authorized terminal device stored in the data analysis module are sequentially labeled as A1, A2, ..., Aa, where a≥1, according to the order of storage. S12: Obtain all pull tracking data of terminal data A1 from the tracking and acquisition module and extract the data. During the extraction process, extract the demand platform and demand business item from each pull tracking data. Use the demand platform as the key and the demand business item as the value to construct the demand dictionary corresponding to the pull tracking data. Add the constructed demand dictionaries corresponding to all pull tracking data into an empty set to obtain the demand set of terminal data A1. S13: Obtain the demand sets of terminal data A2, A3, ..., Aa respectively according to S12, remove duplicates from the demand sets of terminal data A1, A2, ..., Aa, and mark all the remaining demand sets after deduplication as C1, C2, ..., Cc respectively, where c≥1; S14: Traverse the demand sets of terminal data A1, A2, ..., Aa, obtain all terminal data that correspond to the demand set consistent with demand set C1, and re-mark them as D1, D2, ..., Dd, where 1≤d≤a; S15: Copy the demand set C1 once to obtain a replica set of demand set C1, and bind all demand dictionaries in the replica set according to all the pull tracking data of terminal data D1 and their corresponding demand dictionaries; S16: Copy the demand set C1 again d-1 times to obtain d-1 replica sets of demand set C1. Bind the d-1 replica sets to the terminal data D2, D3, ..., Dd respectively according to S15. S17: First, determine all the requirement dictionaries that need to be data bound in the requirement set C1 according to the preset filtering and binding rules, and filter to obtain the corresponding dependency requirement sets that should be bound. After filtering, perform data binding on the corresponding requirement dictionaries in the requirement set C1. S18: Perform data binding on several requirement dictionaries within the requirement sets C2, C3, ..., Cc according to S13 to S17, and transfer all requirement sets that have completed data binding to the mounting and sharing platform for storage.
7. The method for providing terminal data services across cloud platforms according to claim 6, characterized in that, In S15, the specific details of data binding for all requirement dictionaries within the replica set are as follows: SS21: According to the order of storage, the pull tracking data of all pulls of terminal data D1 are sequentially marked as E1, E2, ..., Ee, where e is the total number of times terminal data D1 is pulled when the mounting status is valid; SS22: Extract the processing data sequentially from the retrieved tracking data E1, E2, ..., Ee, and label them as F1, F2, ..., Fe; SS23: Create a non-transformable variable G1 for processing data F1, with the initial value of the non-transformable variable G1 being 0; The processing data F1 is selected as the conversion source. Based on the processing data F2, it is selected whether to perform a preset gain direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. If it is selected to perform a preset pull gain direction determination on the conversion source, and the pull gain direction is determined to be positive after the determination is completed, then the demand dictionary corresponding to the pull tracking data F2 is selected as the auxiliary demand dictionary of the demand dictionary of the pull tracking data F1. SS24: According to SS23, based on the processed data F3, F4, ..., Fe, select whether to perform a preset gain determination on the conversion source or to increment the value of the non-convertible variable G1 by 1. After all the executions are completed, obtain all the subordinate demand dictionaries of the demand dictionary corresponding to the pull tracking data F1, and classify them into an empty set to obtain the subordinate demand set of the demand dictionary corresponding to the pull tracking data F1. SS25: After step SS24 is completed, obtain the value J1 of the non-convertible variable G1 at this time, and count the number J2 and J3 of positive and negative gain directions after step SS24 is completed; Using formula Calculate the binding tendency index K1 of the demand dictionary corresponding to the pull tracking data E1 relative to the terminal data D1, and associate the binding tendency index K1 with the demand dictionary in the replica set that is consistent with the demand dictionary corresponding to the pull tracking data E1. Simultaneously associate the demand dictionary with the auxiliary demand set corresponding to the pull tracking data E1. N is the total number of judgments, which is the sum of J1, J2, and J3. α1, α2, and α3 are the preset variable proportion coefficients, positive and negative proportion coefficients, respectively. SS26: Create non-convertible variables for processing data F2, F3, ..., Fe respectively, select processing data F2, F3, ..., Fe as the transformation source respectively, and perform data binding on all requirement dictionaries in the replica set according to SS21 to SS25 respectively.
8. The method for providing terminal data services across cloud platforms according to claim 7, characterized in that, In SS23, the selection based on the processing data F2 regarding whether to perform a preset gain-direction determination on the conversion source or to increment the value of the non-convertible variable G1 by 1 is as follows: Select the processed data F2 as the target source of the conversion source, and obtain the demand dictionaries corresponding to the pull tracking data E1 and E2, which contain the conversion source and the target source respectively. Use the two demand dictionaries as search conditions to search in the data analysis module to see if there is a pre-stored conversion list. The conversion dictionaries with indices 0 and 1 in the list are consistent with the demand dictionaries corresponding to the pull tracking data E1 and E2 respectively. If they exist, select to perform a preset pull gain direction judgment on the conversion source. If they do not exist, select to perform an auto-increment operation on the value of the non-convertible variable G1. If a preset pull gain direction determination is performed on the selected conversion source, then the pull gain direction determination is performed on it, and the determination content is as follows: Obtain the conversion rules corresponding to the conversion list pre-stored in the data analysis module, convert the conversion source according to the conversion rules, and after the conversion is completed, obtain the conversion time of the conversion source relative to the target source, that is, the conversion time H2 of converting the processed data F1 to the processed data F2; The pull gain value I1 of the conversion source relative to the target source is calculated using the formula I1=H1+H2-(H3+H4)-P1, where H1 is the processing time within the pull tracking data including the conversion source, H3 is the pull interval time obtained by subtracting the absolute value of the pull times in the pull tracking data E1 and E2, H4 is the processing time within the pull tracking data E1, and P1 is a preset time compensation constant. If I1≥0, the pull gain direction of the conversion source relative to the target source is determined to be positive, and the demand dictionary corresponding to the pull tracking data F2 is selected as the auxiliary demand dictionary of the pull tracking data F1. Otherwise, no processing is performed.
9. A method for providing terminal data services across cloud platforms according to claim 7, characterized in that, In S17, the data binding of several requirement dictionaries within the requirement set C1 is as follows: SS31: Label all requirement dictionaries contained in requirement set C1 as L1, L2, ..., Ll, l≥1, where l is the total number of requirement dictionaries in requirement set C1; SS32: After the data binding of the d replica sets of the demand set C1 is completed, obtain the binding tendency index of the d replica sets that are associated with the demand dictionary L1. Use the discrete point filtering algorithm to calculate all the obtained binding tendency indices for data processing, obtain the average value of all remaining binding tendency indices after data processing, and use the average value as the mounting peer average value of the demand dictionary L1. SS33: Compare the mean value of the mounting peers with a preset mounting peer threshold. If the mean value of the mounting peers is greater than or equal to the mounting peer threshold, it is determined that the requirement dictionary L1 needs to be data bound. At this time, obtain the auxiliary requirement sets that are associated with the requirement dictionary L1 in each of the d replica sets, and obtain a total of d auxiliary requirement sets. Count the number of times each auxiliary requirement dictionary appears in these d auxiliary requirement sets, and calculate the ratio of the number of times each auxiliary requirement dictionary appears to d. The proportion value of each auxiliary requirement dictionary is compared with the preset proportion threshold one by one. All auxiliary requirement dictionaries with a proportion value greater than or equal to the preset proportion threshold are retained. All retained auxiliary requirement dictionaries are put into an empty set. This set is used as the dependent requirement set that requirement dictionary L1 is bound to. Data binding is performed between requirement dictionary L1 and requirement dictionary L1 in requirement set C1. Conversely, if the average value of the mount peers is less than the mount peer threshold, no action is taken. SS34: Determine whether data binding is required for the requirement dictionaries L2, L3, ..., L1 within the requirement set C1 according to SS31 to SS33, and perform data binding on several requirement dictionaries within the requirement set C1 based on the determination results.
10. A terminal data service system across cloud platforms, characterized in that, include: The data coordination module is used to receive and store the terminal data periodically transmitted back by each authorized terminal device. Specifically, each authorized terminal device collects terminal data once every corresponding preset collection cycle and transmits it to the data coordination module in real time. The data coordination module is also used to determine whether the terminal data needs to be mounted and shared after receiving the terminal data periodically returned by each authorized terminal device. The authorized terminal device refers to various terminal nodes that have been authorized to access the system after identity authentication and permission verification. The data coordination module pre-stores all the demand platforms for each authorized terminal device, as well as the demand cycles for all demand business items deployed on each authorized terminal device relative to its respective demand platforms. The mounting and sharing platform is used to mount and share the terminal data according to the preset mounting and sharing rules based on the data of the requested object after receiving terminal data and its request object. The tracking and acquisition module is used to monitor the terminal data of each terminal with a valid mounting status in the mounting and sharing platform in real time. For any terminal data with a valid mounting status, each time the terminal data is detected to be pulled, the module obtains the requesting platform and the requesting business item for pulling the terminal data. After the requesting platform processes the terminal data according to the preset processing rules of the requesting business item based on the terminal data and outputs the processed data, the module obtains the processed data, the processing time, and the time of this pull. Based on the obtained processed data, processing time, and pull time, the module generates and stores the pull tracking data for one pull of the terminal data. The data analysis module is used to analyze all the retrieved tracking data of all terminal data of the stored credit terminal device after the number of retrieved tracking data of any credit terminal device stored in the tracking and acquisition module reaches a preset fixed number, and to determine all the demand sets of the credit terminal device. The mounting and sharing platform is also used to optimize the mounting and sharing process of terminal data after receiving all the demand sets of the authorized terminal device and any terminal data and demand object data of the authorized terminal device received subsequently. During the optimization process, pull permissions are granted to all demand platforms in the demand object data, and after several demand platforms process the terminal data and output corresponding processed data, the data is obtained. Based on all the obtained processed data, pull permissions are granted to several demand platforms.
