A method, device and storage medium for efficient data synchronization in an office system.

Abstract

This invention relates to the field of data synchronization technology, specifically to a method, device, and storage medium for efficient data synchronization in an office system. The invention collects several changed fields and their attribute data, time-series data, and concurrent operation data of the current business process within the current synchronization cycle, as well as the data of the current business process itself. Based on the changed field attribute data and concurrent operation data, the business weight of each changed field in the current business process is obtained. Atomic synchronization units are constructed by analyzing the closeness of business linkage between each changed field and other changed fields. The exclusivity and overlap rate of atomic synchronization units containing the same changed fields are analyzed to select conflict-free synchronization units. Based on the numerical distribution of the changed field business weights within the conflict-free synchronization units and the timeliness characteristics of the time-series data, the priority scheduling coefficient of the conflict-free synchronization units is obtained. Combined with the basic risk characteristics in the changed field attribute data, this guides the orderly execution of the final update synchronization.

Description

Technical Field

[0001] This invention relates to the field of data synchronization technology, and specifically to a method, device and storage medium for efficient data synchronization in an office system. Background Technology

[0002] With the continuous advancement of enterprise informatization and digital office practices, office systems have gradually evolved from monolithic applications into distributed office platforms that integrate multiple terminals, nodes, and cross-system collaboration. In practical applications, office systems typically need to interact with data across multiple terminal devices and business systems, such as document editing and version update synchronization, and approval process status synchronization. As enterprises expand and office scenarios become more complex, data synchronization in office systems often needs to simultaneously meet requirements such as high real-time performance, high consistency, low resource consumption, and high reliability. Therefore, how to achieve a highly efficient and low-latency data synchronization mechanism while ensuring data consistency has become a key technical issue in office system architecture design.

[0003] In the data synchronization process of existing office systems, data processing mechanisms generally lack the ability to identify the business scope of data fields, resulting in the inability to perform fine-grained processing based on actual field-level changes. Because different fields play different roles in business processes, when the synchronization mechanism still uses object-level or full-data processing methods, it can lead to redundant transmission of invalid data, decreased synchronization efficiency in weak network environments, and the inability to identify the independence of different fields when multiple users concurrently modify them, thus misjudging non-conflicting operations as conflicts and causing data overwriting. Furthermore, due to the lack of modeling capabilities for the business importance of fields, the synchronization queue cannot achieve dynamic scheduling based on business priorities, and it also lacks atomic processing mechanisms for the scope of field effects, further leading to synchronization blocking and data consistency risks. Summary of the Invention

[0004] To address the technical problems of insufficient modeling capability for field business importance and low efficiency of atomic data synchronization processing in existing office systems, this invention aims to provide a method, device, and storage medium for efficient data synchronization in office systems. The specific technical solution adopted is as follows: This invention proposes a method for efficient data synchronization in an office system, the method comprising: Collect data from the system log of several changed fields and their attributes that have changed when data synchronization is triggered, as well as the time-series data and concurrent operation data of the current business process within the current synchronization period. Based on the modification permissions of the user in the attribute data and the concurrent operation data, the business weight of each changed field in the current business process is obtained. Combined with the analysis of the closeness of the business linkage between each changed field and other changed fields in the current business process, an atomic synchronization unit is constructed. Based on the attribute data, analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field, and filter the atomic synchronization units to obtain conflict-free synchronization units. Based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, the priority scheduling coefficient of each conflict-free synchronization unit is obtained; based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit, data update and synchronization are performed on each changed field.

[0005] Furthermore, based on the attribute data and the modification permissions of the operating users in the concurrent operation data, the business weight of each changed field in the current business process is obtained, including: The attribute data includes the business weight benchmark of each changed field in the current business process; the concurrent operation data includes the modification permission range of each operating user in the current business process; Based on the modification permission range of the corresponding user for each changed field under the current data change operation, the dynamic correction coefficient of each changed field in the current business process is obtained; For each changed field, the business weight benchmark is adjusted based on a dynamic correction coefficient to obtain the business weight of each changed field in the current business process.

[0006] Furthermore, by analyzing the degree of business linkage between each changed field and other changed fields within the current business process, atomic synchronization units are constructed, including: Combine any two different change fields within the current business process into a field pair. The attribute data includes the dependency and linkage coefficient benchmark of each field pair under the current business process. For any pair of fields, based on the balance of the business weights of the two changed fields in the pair, the dependency linkage coefficient benchmark of each pair of fields is adjusted to obtain a dependency linkage coefficient that quantifies the degree of business linkage between each pair of fields. Based on the dependency linkage coefficient of each field pair, the field pairs are combined to obtain atomic synchronization units.

[0007] Furthermore, based on the dependency linkage coefficient of each field pair, the field pairs are combined to obtain atomic synchronization units, including: Field pairs with dependency linkage coefficients greater than or equal to a preset dependency threshold are assigned to the same atomic synchronization unit; the two changed fields in field pairs with dependency linkage coefficients less than the preset dependency threshold are assigned to different atomic synchronization units. Change fields whose business weight is greater than the preset weight threshold are marked as exclusive fields; For atomic synchronization units containing exclusive fields, all atomic synchronization units containing exclusive fields within the current business process are merged into a single exclusive atomic synchronization unit.

[0008] Furthermore, based on the attribute data, the exclusivity and overlap rate of all atomic synchronization units containing the same change field are analyzed, and the atomic synchronization units are filtered to obtain conflict-free synchronization units, including: The attribute data includes a change field ID. Based on each change field ID, all atomic synchronization units that do not have the same change field as other atomic synchronization units in the current synchronization cycle are filtered as conflict-free synchronization units. Based on each change field ID, all atomic synchronization units containing the same change field are grouped into a unit group; any two atomic synchronization units within a unit group are grouped into a unit pair. Based on whether there is an exclusive atomic synchronization unit in the two atomic synchronization units of each unit pair, determine the exclusivity attribute coefficient of each unit pair; Based on the degree of overlap between the modification permission scopes of the two atomic synchronization units corresponding to each unit pair in the current business process, determine the modification permission overlap rate of each unit pair. Based on the ratio of the business weights corresponding to the same changed fields in the two atomic synchronization units of each unit pair to the business weights corresponding to all changed fields in the two atomic synchronization units, the business domain overlap rate of each unit pair is determined. The operation timing overlap rate of each unit pair is determined based on the degree of overlap between the write time windows of the two atomic synchronization units of each unit pair; wherein, the write time window refers to the time window between the earliest and latest times corresponding to all changed fields of the atomic synchronization unit in the current synchronization cycle. The business mutual exclusion degree of each unit pair is obtained by weighted summing of the exclusive attribute coefficient, modification permission overlap rate, business domain overlap rate and operation sequence overlap rate. All atomic synchronization units within all unit pairs whose business mutual exclusion degree is less than or equal to the preset mutual exclusion threshold are selected as conflict-free synchronization units.

[0009] Furthermore, based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, the priority scheduling coefficient for each conflict-free synchronization unit is obtained, including: The time-series data includes the remaining time, total time, and historical timeout rate of the current business process; based on the ratio of the remaining time to the total time of the current business process, and the historical timeout rate, the time urgency index of the current business process is obtained. The maximum value of the business weight in all change fields within each conflict-free synchronization unit is used as the business importance index for each conflict-free synchronization unit. The priority scheduling coefficient of each conflict-free synchronization unit is obtained by weighted summing of the business importance index and the timing urgency index of the current business process.

[0010] Furthermore, based on the priority scheduling coefficient and the basic risk characteristics in the changed field attribute data within the conflict-free synchronization unit, data update and synchronization are performed on each changed field, including: The maximum value of the conflict risk baseline in all changed field attribute data within each conflict-free synchronization unit is used as the conflict risk indicator for each conflict-free synchronization unit. Based on the priority scheduling coefficient and conflict risk index of each conflict-free synchronization unit, real-time reliable transmission link or batch asynchronous transmission link matching is performed on each conflict-free synchronization unit to obtain the transmission link matching result of each conflict-free synchronization unit. Based on the priority scheduling coefficient of each conflict-free synchronization unit, all conflict-free synchronization units are sorted to obtain the primary synchronization transmission queue; based on the transmission link matching result of each conflict-free synchronization unit, the primary synchronization transmission queue is sorted again to obtain the final synchronization transmission queue. Based on the final synchronization transmission queue, data updates and synchronization are performed on each changed field.

[0011] Furthermore, based on the priority scheduling coefficient and conflict risk index of each conflict-free synchronization unit, real-time reliable transmission link or batch asynchronous transmission link matching is performed on each conflict-free synchronization unit to obtain the transmission link matching result of each conflict-free synchronization unit, including: Conflict-free synchronization units with priority scheduling coefficients greater than or equal to the preset scheduling threshold and conflict risk indicators greater than or equal to the preset risk threshold are matched with real-time reliable transmission links; other conflict-free synchronization units that do not meet the above requirements are matched with batch asynchronous transmission links.

[0012] This invention also proposes a high-efficiency data synchronization device for an office system, the device comprising: The change data acquisition module is used to collect data on several changed fields and their attributes in the system log during the current synchronization period of the current business process when data changes occur under the conditions that trigger data synchronization, as well as the time-series data and concurrent operation data of the current business process. The atomic synchronization unit construction module is used to obtain the business weight of each changed field in the current business process based on the modification permissions of the operation user in the attribute data and the concurrent operation data, and to construct an atomic synchronization unit by combining the analysis of the closeness of the business linkage between each changed field and other changed fields in the current business process. The conflict-free synchronization unit filtering module is used to analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field based on the attribute data, and to filter the atomic synchronization units to obtain conflict-free synchronization units. The data update and synchronization module is used to obtain the priority scheduling coefficient of each conflict-free synchronization unit based on the numerical distribution of the business weight of the changed fields and the timeliness characteristics of the time-series data within each conflict-free synchronization unit; and to perform data update and synchronization for each changed field based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit.

[0013] The present invention also proposes a data-efficient synchronization storage medium for an office system, wherein the storage medium stores computer program code, and the computer program code, when run on a computer, enables the computer to implement any of the steps of the data-efficient synchronization method for an office system described above.

[0014] The present invention has the following beneficial effects: This invention collects data on several changed fields and their attributes from the system log within the current synchronization period of the current business process where data changes occur when data synchronization is triggered. It also collects the time-series data and concurrent operation data of the current business process, providing a unified and computable data foundation for subsequent analysis. Based on the attribute data of the changed fields and the concurrent operation data of the current business process, the business weight of each changed field in the current business process is obtained, quantifying the business impact of each changed field. By combining the analysis of the closeness of the business linkage between each changed field and other changed fields within the current business process, atomic synchronization units are constructed to balance business integrity and synchronization efficiency, providing the smallest complete business execution unit for subsequent data synchronization management.

[0015] This study analyzes the exclusivity and overlap rate of all atomic synchronization units containing the same modified fields, quantifies the risk of concurrent conflicts between atomic synchronization units, and filters out conflict-free synchronization units accordingly. This achieves strong data consistency in concurrent scenarios by identifying conflict risks, providing a standardized basis for subsequent synchronization concurrency management.

[0016] Based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, the priority scheduling coefficient of each conflict-free synchronization unit is obtained. Combined with the basic risk characteristics in the attribute data of the changed fields within each conflict-free synchronization unit, the final update synchronization is executed in an orderly manner. This enables the system to allocate appropriate execution strategies and resources to different update tasks, thereby ensuring that the limited system processing capacity prioritizes and reliably guarantees the flow of core business. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for efficient data synchronization in an office system, provided as an embodiment of the present invention. Detailed Implementation

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific solution for an efficient data synchronization method, device, and storage medium for an office system provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an efficient data synchronization method for an office system according to an embodiment of the present invention. The method includes: Step S1: Collect data from the system log of several changed fields and their attributes in the current synchronization period for the current business process that has undergone data changes when the data synchronization condition is triggered, as well as the time sequence data and concurrent operation data of the current business process.

[0021] In distributed office systems, the core of accurate data updates and synchronization lies in analyzing the changed data and the business processes involved in the updates. Therefore, this invention provides a unified and computable data foundation for subsequent analysis.

[0022] When conditions trigger data synchronization (such as field writing, process node changes, terminal reconnection, or cross-system synchronization requests), the system extracts the changed business process corresponding to the changed business entity (e.g., if the business entity is a leave application, its corresponding business process includes leave application submission, direct supervisor approval, HR approval, and leave process completion), locates the relevant changed data (e.g., leave days, leave reason, and approval status), and obtains the corresponding changed fields. Here, a field refers to the smallest unit of a data record in a business process. For example, in a leave application, a field is each independent, modifiable, and changeable data item, such as 3 days (referring to the specific value of the leave days field). The system extracts the time interval from the last synchronization time to the current synchronization time of the changed current business process as the current synchronization period. It also collects several changed fields and their attribute data of the business entities that changed within the current synchronization period, as well as the time-series data and concurrent operation data of the current business process, through system logs.

[0023] In this embodiment of the invention, the attribute data of several changed fields includes: changed field ID, original data before change synchronization, final data after change synchronization, operating user, timestamp (corresponding time node) when each changed field is changed, business weight benchmark of each changed field in the current business process, dependency and linkage coefficient benchmark of each changed field with other changed fields in the current business process, and conflict risk benchmark of each changed field in the current business process.

[0024] The current business process's time-series data includes: current business process ID, remaining timeout, total timeout, and historical timeout rate. Remaining timeout refers to the time between the current synchronization time and the deadline for completion of the current business process. Total timeout refers to the total time from start to finish of the current business process. Historical timeout rate is the ratio in the system logs of the number of times the current business process was not processed within the total timeout to the total number of times the current business process was triggered. These parameters can be directly extracted from the office system.

[0025] The concurrent operation data of the current business process includes: the modification permission scope of each user in the current business process and the concurrent modification records of each user within the current synchronization period. The modification permission scope refers to the total number of fields that a user is allowed to modify in the current business process, and the concurrent modification records refer to the records of different users simultaneously modifying business entities within the current synchronization period (such as in a leave application, the applicant modifies the number of leave days at the same time as the direct supervisor approves the application).

[0026] In one specific implementation of this invention, before the conditions for triggering data synchronization are met, the system extracts the field change records of all business entities, the association records between each changed field and each business process, the concurrent conflict records of each operating user in the same business process, and the timeliness records of each business process from the system's historical synchronization log.

[0027] As a specific example, this embodiment uses any field, field pair, business process, and business entity as examples for ease of description. Specifically, any business entity is taken as the target business entity, any business process in the target business entity is taken as the target business process, and any field pair and any field in the target business process are respectively denoted as the target field pair and the target field.

[0028] The frequency percentage of each field triggering process changes in each business process is used as the business weight benchmark for that field in that business process. More specifically, the ratio of the number of times the target field undergoes data changes in the target business process to the total number of times all fields undergo data changes in the target business process is used as the frequency percentage of the target field triggering process changes in the target business process.

[0029] Any two different fields of the same business entity within the same business process are combined into a field pair. The co-occurrence change frequency ratio between each field pair is used as the baseline for the dependency linkage coefficient of each field pair. More specifically, the ratio of the number of times the two fields in the target field pair undergo simultaneous data changes within the target business process to the total number of times the two fields in all field pairs undergo simultaneous data changes within the target business process is used as the co-occurrence change frequency ratio of the target field pair within the target business process. Here, one field pair corresponds to one number of simultaneous data changes within the target business process, and the total number refers to the cumulative number of changes corresponding to all field pairs within the target business process.

[0030] The proportion of conflicts that occur to each field in scenarios where each user concurrently modifies the target field is used as the conflict risk benchmark for each field. More specifically, the ratio of the number of times a target field causes conflicts due to concurrent modifications by all different users in each business process of the target business entity to the total number of concurrent modifications by all different users in each business process of the target business entity is used as the proportion of conflicts that occur to the target field in scenarios where each user concurrently modifies the target field.

[0031] Concurrent modification conflicts refer to situations where a field is modified or changed simultaneously by multiple users. For example, consider a business entity like a leave application form, with corresponding business processes including leave application submission, direct supervisor approval, HR approval, and leave process completion. If a user in the HR approval process modifies the applicant's leave days, and the applicant also modifies their leave days, this change to the leave days field will cause a conflict in this scenario. The total number of concurrent modifications by all different users across all business processes of the target business entity represents the sum of all instances where different users simultaneously modify or change a field across all business processes within the target business entity.

[0032] For fields without historical data, the business weight baseline is initialized to 1, the dependency linkage coefficient baseline is initialized to 1, and the conflict risk baseline is initialized to 0. The strictest atomic packaging and exclusive isolation are implemented, and the baseline will be adjusted as the operation logs accumulate.

[0033] Step S2: Based on the modification permissions of the user in the attribute data and the concurrent operation data, obtain the business weight of each changed field in the current business process, and construct an atomic synchronization unit by analyzing the degree of business linkage between each changed field and other changed fields in the current business process.

[0034] In distributed office systems, different fields of the same business entity have fundamentally different business functions and impacts within various business processes. Traditional synchronization mechanisms fail to identify the business impact of each changed field within the changed business process, as well as the degree of business linkage between each changed field and other changed fields within the changed business process, resorting only to object-level full synchronization. Extreme synchronization modes can easily lead to data transmission redundancy or business logic distortion, causing problems such as data overwriting and process logic disorder. Furthermore, static dependency rules are difficult to adapt to dynamic process requirements. Therefore, this embodiment relies on the linkage data between each changed field within the business entity and the current changed business process to quantitatively analyze the business impact of each changed field within the current business process. Combined with the degree of business linkage between each changed field and other changed fields within the current business process, atomic synchronization units are constructed. The degree of business linkage between changed fields is evaluated to balance business integrity and synchronization efficiency, providing the smallest complete business execution unit for subsequent data synchronization management.

[0035] Based on the distribution characteristics of the modification permission scope of each user in the current business process in the concurrent operation data described in step S1 and the business weight benchmark of each change field in the current business process in the attribute data, analyze the business weight of each change field in the current business process.

[0036] Based on the dependency linkage coefficient benchmark between each changed field and other changed fields in the current business process in the attribute data described in step S1, and combined with the business weight of the corresponding two changed fields in the current business process, the dependency linkage coefficient between each changed field and other changed fields in the current business process is analyzed. Based on the relationship between the dependency linkage coefficient between each changed field and other changed fields in the current business process and the preset dependency threshold, an atomic synchronization unit is constructed.

[0037] Step S3: Based on the attribute data, analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field, and filter the atomic synchronization units to obtain conflict-free synchronization units.

[0038] Because existing distributed office systems' synchronization mechanisms cannot quantify the risk of concurrent conflicts between atomic synchronization units, relying solely on a one-size-fits-all serial or unlimited concurrent mode can easily lead to low synchronization efficiency or data consistency anomalies. This embodiment addresses this by balancing concurrent conflict risk and data consistency by analyzing the exclusivity and overlap rate of any two different atomic synchronization units containing the same changed field. This quantifies the risk of concurrent conflicts between atomic synchronization units and, based on this, filters out conflict-free synchronization units. This achieves strong data consistency in concurrent scenarios by identifying conflict risks, providing a standardized basis for subsequent synchronization concurrency management.

[0039] Based on the change field ID in the attribute data described in step S1, all atomic synchronization units containing the same change field are located. The concurrency conflict risk between any two atomic synchronization units is quantified using four dimensions: exclusive attribute coefficient, modification permission overlap rate, business domain overlap rate, and operation sequence overlap rate. This is used as the business mutual exclusion degree between the two atomic synchronization units. A higher business mutual exclusion degree indicates a greater risk of concurrency conflict between the two atomic synchronization units, and a higher probability of business anomalies; conversely, a lower business mutual exclusion degree indicates a lower risk of concurrency conflict between the two atomic synchronization units, and a lower probability of business anomalies. Combined with a preset mutual exclusion threshold, atomic synchronization units are filtered to obtain conflict-free synchronization units.

[0040] Step S4: Based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, obtain the priority scheduling coefficient of each conflict-free synchronization unit; based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit, perform data update and synchronization for each changed field.

[0041] Because existing distributed office systems often employ a crude approach to data update and synchronization, failing to differentiate between business criticality and timeliness, high-priority tasks become congested and system resources are underutilized, this embodiment aims to achieve intelligent allocation of system processing capacity while ensuring data consistency and maximizing business value throughput. It obtains priority scheduling coefficients based on the numerical distribution of business weights of changed fields within each conflict-free synchronization unit and the timeliness characteristics of time-series data. Furthermore, it guides the orderly execution of final update synchronization by incorporating fundamental risk characteristics from the attribute data of changed fields within each conflict-free synchronization unit. This allows the system to allocate appropriate execution strategies and resources to different update tasks, ensuring that limited system processing capacity prioritizes and reliably guarantees the flow of core business operations.

[0042] Based on the remaining time-to-date, total time-to-date, and historical timeout rate of the current business process in the time-series data described in step S1, the time-to-date characteristics of the current business process are analyzed. Combined with the numerical distribution of the business weights of the change fields within each conflict-free synchronization unit, the priority scheduling coefficient for each conflict-free synchronization unit is obtained. A larger priority scheduling coefficient corresponds to a higher priority for the conflict-free synchronization unit; a smaller priority scheduling coefficient corresponds to a lower priority for the conflict-free synchronization unit.

[0043] Based on the conflict risk benchmark of the attribute data described in step S1, analyze the basic risk characteristics of each conflict-free synchronization unit, combine the priority scheduling coefficient of each conflict-free synchronization unit, determine the transmission resource consumption and transmission order of each conflict-free synchronization unit, and update and synchronize the data of each changed field according to the transmission order of each conflict-free synchronization unit.

[0044] In this embodiment of the invention, the changed field ID of the attribute data described in step S1 is accurately written into the target business database, and all data in the corresponding field attribute data is updated synchronously. After all conflict-free synchronization units are written, the time-series data of the business entity in the current business process is updated immediately. According to the numerical distribution of the business weight of the changed field in each conflict-free synchronization unit, standardized business events associated with the changed field are triggered sequentially, including business process jumps, collaborative user notifications, permission status updates, cross-system data pushes, etc., to achieve seamless connection between data synchronization and business processes.

[0045] After this synchronization is completed, the office system archives the full-link synchronization data according to the atomic synchronization unit dimension, including change records, calculation results, conflict handling and transmission effect data, and stores them in the system's historical synchronization log, forming a self-optimizing closed loop of synchronization rules.

[0046] In summary, this invention collects data on several changed fields and their attributes from the system log within the current synchronization period of the current business process that undergoes data changes when data synchronization is triggered, along with the time-series data and concurrent operation data of the current business process. Based on the attribute data of the changed fields and the concurrent operation data of the current business process, the business weight of each changed field within the current business process is obtained. By analyzing the closeness of the business linkage between each changed field and other changed fields within the current business process, atomic synchronization units are constructed. The exclusivity and overlap rate of all atomic synchronization units containing the same changed field are analyzed, and conflict-free synchronization units are selected. Based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the time-series data's timeliness characteristics, the priority scheduling coefficient of each conflict-free synchronization unit is obtained. This, combined with the basic risk characteristics in the attribute data of the changed fields within each conflict-free synchronization unit, guides the orderly execution of the final update synchronization.

[0047] Preferably, in some implementations of the present invention, the business weight of each changed field in the current business process is obtained based on the modification permissions of the user in the attribute data and the concurrent operation data, including: The attribute data includes the business weight benchmark of each changed field in the current business process; the concurrent operation data includes the modification permission range of each operating user in the current business process; Based on the modification permission range of the corresponding user for each changed field under the current data change operation, the dynamic correction coefficient of each changed field in the current business process is obtained; For each changed field, the business weight benchmark is adjusted based on a dynamic correction coefficient to obtain the business weight of each changed field in the current business process.

[0048] The scope of modification permissions for each changed field under the current data change operation represents the coverage of that user's modification permissions within the current business process. Higher coverage means a greater weighting of the user's modification operation on the business process, resulting in a larger dynamic correction coefficient for each changed field within the current business process; conversely, lower coverage means a smaller weighting of the user's influence on the business process, resulting in a smaller dynamic correction coefficient for each changed field within the current business process.

[0049] The business weight of each changed field in the current business process represents the actual business core level of that changed field within the current business process. The higher the business weight, the higher the actual business core level of the changed field in the current business process, and the greater its impact on the current business process; conversely, the lower the business weight, the lower the actual business core level of the changed field in the current business process, and the smaller its impact on the current business process.

[0050] More specifically, for any changed field i among all changed fields, the formula for obtaining its dynamic correction coefficient in the current business process is: in, This represents the dynamic adjustment coefficient of the changed field i in the current business process; This represents the total number of fields that the corresponding user can modify in the current business process for the changed field i. This represents the total number of fields that can be modified by all users in the current business process.

[0051] In this context, the total number of fields that a user can modify within the current business process represents the modification permission range for that user; the total number of fields that all users can modify within the current business process represents the sum of their modification permission ranges. The modification permission range refers to the total number of fields that a user is allowed to modify within the current business process.

[0052] The "1+" format is designed to provide a baseline for the business weight of each changed field within the current business process, ensuring that the dynamic adjustment coefficient amplifies rather than weakens or resets it. Without the "1+" format, when a user has extremely low modification permissions (e.g., can only modify one field), the dynamic adjustment coefficient would approach zero. This would lead to the business weight being incorrectly flattened to zero after the dynamic adjustment coefficient, regardless of how core the changed field is (how high the business weight baseline is), violating common business logic. Therefore, using a combination of baseline value and percentage ensures that the original core status of the changed field is never downgraded.

[0053] Furthermore, for any changed field i among all changed fields, the formula for obtaining its business weight in the current business process is: in, This represents the business weight value of the changed field i in the current business process; This represents the business weight benchmark of the changed field i in the current business process; This represents the dynamic adjustment coefficient of the changed field i in the current business process; This represents the original business weight value of the changed field i in the current business process; It is a normalization function for maximum and minimum values.

[0054] The business weight values ​​are converged to the interval [0, 1] using a maximum-minimum normalization function. The maximum and minimum values ​​of the original business weights can be determined by considering the range of values ​​for all changed fields within the current business process, and then normalization calculations can be performed.

[0055] Specifically, in this embodiment, the formula for the maximum and minimum value normalization function is: in, This represents the original business weight value of any changed field that needs to be normalized in the current business process. This represents the normalized business weight value after the original business weight value has been obtained. This represents the minimum value of all original business weights for all changed fields in the historical data under the current business process; This represents the maximum value of all original business weight values ​​for all changed fields in the historical data under the current business process; It is a constant set to prevent the removal of zero.

[0056] It should be understandable that, using The method of normalizing data using the maximum and minimum value normalization function is a well-known technique, and will only be briefly introduced here without further elaboration.

[0057] Preferably, in some implementations of the embodiments of the present invention, an atomic synchronization unit is constructed by combining analysis of the closeness of business linkage between each changed field and other changed fields in the current business process, including: Combine any two different change fields within the current business process into a field pair. The attribute data includes the dependency and linkage coefficient benchmark of each field pair under the current business process. For any pair of fields, based on the balance of the business weights of the two changed fields in the pair, the dependency linkage coefficient benchmark of each pair of fields is adjusted to obtain a dependency linkage coefficient that quantifies the degree of business linkage between each pair of fields. Based on the dependency linkage coefficient of each field pair, the field pairs are combined to obtain atomic synchronization units.

[0058] Since static dependency linkage coefficient benchmarks cannot accurately reflect the actual linkage tightness of each field pair in a dynamic business environment, introducing dynamic business weights can correct and "enhance" the business influence of static dependency linkage coefficient benchmarks. Two changed fields may only have occasional linkage in the past (low dependency linkage coefficient benchmark), but if they both have extremely high business weights in the current business process (i.e., both are core key fields), this means that once they change simultaneously, they will definitely have a decisive impact on the current business process. By stacking business weights to increase the linkage tightness of such high-risk combinations, the linkage tightness of such combinations can be forcibly increased and packaged into an atomic synchronization unit. In addition, historical statistics often show a large number of non-core auxiliary fields (such as "update time" and "view count") frequently co-occurring. Since their business weights are extremely low, weighting and suppressing their dependency linkage coefficient benchmarks can prevent the system from misjudging unimportant field pairs as having high linkage tightness and forcibly packaging them into an atomic synchronization unit, avoiding unnecessary synchronization blocking.

[0059] The dependency linkage coefficient is a quantitative value of the degree of business linkage between each field. The larger the dependency linkage coefficient, the stronger the business linkage between the corresponding field pair; the smaller the dependency linkage coefficient, the weaker the business linkage between the corresponding field pair.

[0060] More specifically, for any pair of fields z, the formula for calculating the dependency linkage coefficient is: in, The coefficient representing the dependence of the field on z; The baseline for the dependency linkage coefficient of the representative field on z; The representative field represents the business weight of the change field e in z; The representative field represents the business weight of the change field f in z; This is a function for normalizing the maximum and minimum values. Field pair z consists of modified field e and modified field f.

[0061] In this embodiment The application method of the maximum and minimum value normalization function is the same as that described above in the business weight formula.

[0062] Furthermore, based on the dependency linkage coefficient of each field pair, the field pairs are combined to obtain atomic synchronization units, including: Field pairs with dependency linkage coefficients greater than or equal to a preset dependency threshold are assigned to the same atomic synchronization unit; the two changed fields in field pairs with dependency linkage coefficients less than the preset dependency threshold are assigned to different atomic synchronization units. Change fields whose business weight is greater than the preset weight threshold are marked as exclusive fields; For atomic synchronization units containing exclusive fields, all atomic synchronization units containing exclusive fields within the current business process are merged into a single exclusive atomic synchronization unit.

[0063] Some change fields in business entities are often critical data in business processes, such as contract amounts, approval status, and user permissions. The accuracy of this data directly determines financial security, legal compliance, and the continuation of processes. Once data anomalies occur, they can lead to serious business consequences, resulting in decision-making errors, financial losses, or operational disruptions. Marking data as exclusive will force serialization and the highest level of transaction isolation for its changes, ensuring that only one update operation can be executed at any given time. This sacrifices limited concurrency to ensure absolute reliability of core data, priority for resource allocation, and effective reduction of system complexity. Ultimately, it achieves precise protection by "focusing on the most critical data with the strictest control and avoiding the greatest business risks with the least performance cost."

[0064] Exclusive fields determine the direction of business processes. If they are not merged into the same atomic synchronization unit, critical data scattered across different atomic synchronization units will cause fragmentation of the critical data's state, making it impossible to guarantee that scheduled resources will prioritize each exclusive field, thus increasing business risks. Therefore, for atomic synchronization units containing exclusive fields, all atomic synchronization units containing exclusive fields within the current business process are merged a second time into a single exclusive atomic synchronization unit to ensure that core business changes are transactionally indivisible.

[0065] In one specific implementation of this invention, the mean and standard deviation of all dependency linkage coefficients between each changed field and other changed fields within the current business process are calculated, and the value of [mean + 2 standard deviations] is used as a preset dependency threshold. Specifically, to avoid low statistical reference value due to small data volume, this embodiment also sets a minimum threshold restriction strategy, that is, the maximum value between [mean + 2 standard deviations] and the base value of the dependency threshold is used as the preset dependency threshold. The base value of the dependency threshold can be 0.6, which is an empirical value obtained through experiments based on a large amount of historical data.

[0066] In one specific implementation of this invention, the mean and standard deviation of the business weights of all changed fields within the current business process are calculated, and the value of [mean + 2 times standard deviation] is used as a preset weight threshold. Specifically, to avoid low statistical value due to small data volume, this embodiment also sets a minimum threshold restriction strategy, that is, the maximum value between [mean + 2 times standard deviation] and the base value of the weight threshold is used as the preset weight threshold. The base value of the weight threshold can be 0.6, which is an empirical value obtained through experiments based on a large amount of historical data.

[0067] Preferably, in some implementations of the present invention, based on the attribute data, the exclusivity and overlap rate of all atomic synchronization units containing the same change field are analyzed, and the atomic synchronization units are screened to obtain conflict-free synchronization units, including: The attribute data includes a change field ID. Based on each change field ID, all atomic synchronization units that do not have the same change field as other atomic synchronization units in the current synchronization cycle are filtered as conflict-free synchronization units. Based on each change field ID, all atomic synchronization units containing the same change field are grouped into a unit group; any two atomic synchronization units within a unit group are grouped into a unit pair. Based on whether there is an exclusive atomic synchronization unit in the two atomic synchronization units of each unit pair, determine the exclusivity attribute coefficient of each unit pair; Based on the degree of overlap between the modification permission scopes of the two atomic synchronization units corresponding to each unit pair in the current business process, determine the modification permission overlap rate of each unit pair. Based on the ratio of the business weights corresponding to the same changed fields in the two atomic synchronization units of each unit pair to the business weights corresponding to all changed fields in the two atomic synchronization units, the business domain overlap rate of each unit pair is determined. The operation timing overlap rate of each unit pair is determined based on the degree of overlap between the write time windows of the two atomic synchronization units of each unit pair; wherein, the write time window refers to the time window between the earliest and latest times corresponding to all changed fields of the atomic synchronization unit in the current synchronization cycle. The business mutual exclusion degree of each unit pair is obtained by weighted summing of the exclusive attribute coefficient, modification permission overlap rate, business domain overlap rate and operation sequence overlap rate. All atomic synchronization units within all unit pairs whose business mutual exclusion degree is less than or equal to the preset mutual exclusion threshold are selected as conflict-free synchronization units.

[0068] In the data synchronization scenario of a distributed office system, the risk of concurrency conflicts between atomic synchronization units essentially stems from multiple users simultaneously operating on the same modified field. If no atomic synchronization units have the same modified field, then there is no risk of concurrency conflicts between them. Therefore, all atomic synchronization units that do not have the same modified field as any other atomic synchronization units within the current synchronization period are selected as conflict-free synchronization units. Specifically, for each atomic synchronization unit within the current synchronization period, if the unit does not have the same modified field as any other atomic synchronization unit, then that unit is selected as a conflict-free synchronization unit.

[0069] For any pair of units within any unit group, since changes to exclusive units directly impact business process flow, inherent conflict risks exist. Therefore, the exclusivity attribute coefficient is used to ensure that pair of units involving exclusive fields receive a higher conflict risk rating. If both atomic synchronization units in a pair have exclusive atomic synchronization units, the exclusivity attribute coefficient for the corresponding pair is set to 1, indicating that the corresponding atomic synchronization unit is assigned the highest conflict risk level. If neither atomic synchronization unit in a pair has exclusive atomic synchronization units, the exclusivity attribute coefficient for the corresponding pair is set to 0, indicating that the corresponding atomic synchronization unit does not have a mandatory exclusivity feature affecting concurrent conflict risk. This allows subsequent evaluations to objectively quantify the overlap of features such as modification permissions, business domains, and operation sequence.

[0070] The overlap rate of modification permissions for a unit pair represents the degree of overlap in the modification permission ranges of all users corresponding to all changed fields within the two atomic synchronization units of the unit pair in the current business process. The higher the overlap rate, the greater the probability of interference between different users operating simultaneously. The unit permission range of each atomic synchronization unit is obtained by taking the set of all modification permission ranges (the union of all modification permission ranges) of all users corresponding to all changed fields within each atomic synchronization unit of the unit pair as the unit permission range of each atomic synchronization unit. The total number of overlapping fields in the unit permission ranges of the two atomic synchronization units is used as the numerator, and the total number of fields in the union of the unit permission ranges of the two atomic synchronization units is used as the denominator. The resulting ratio is the overlap rate of modification permissions for the unit pair, which ranges from 0 to 1.

[0071] For example, obtain the union of all fields that each user is authorized to modify in the current business process corresponding to the changed fields within atomic synchronization unit A in the unit pair. This union is taken as the unit permission range of atomic synchronization unit A, denoted as Sa. Similarly, obtain the unit permission range of the other atomic synchronization unit B in the unit pair, denoted as Sb. Count the number of intersection fields N1 of Sa and Sb (the number of changed fields that can be modified by both atomic synchronization units), and count the number of union fields N2 of Sa and Sb (the number of changed fields in the union of the changed fields that can be modified by both atomic synchronization units). Modification permission overlap rate = N1 / N2. The closer the modification permission overlap rate is to 1, the more similar the unit permission ranges of the two atomic synchronization units are, and the more likely unauthorized or interference-related concurrent conflicts will occur within the same business domain.

[0072] The business domain overlap rate of a unit pair represents the degree of overlap in the impact of two atomic synchronization units on different business domains within a business entity. A higher overlap rate indicates a greater probability of business conflict. The business domain overlap rate is calculated by using the business weights or the sum of business weights corresponding to the same changed fields within the two atomic synchronization units as the numerator, calculating the sum of business weights corresponding to all changed fields within each atomic synchronization unit, and using the smaller value of the sum of business weights corresponding to all changed fields within the two atomic synchronization units as the denominator. The business domain overlap rate ranges from [0, 1]. Specifically, when there are two or more identical changed fields within the two atomic synchronization units, the sum of the business weights corresponding to multiple identical changed fields is used as the numerator; when there is only one identical changed field, the calculation is performed directly. For example, if atomic synchronization unit A modifies changed fields f1 (business weight 0.5) and f2 (business weight 0.2), the total weight of the two changed fields f1 and f2... =0.7; Atomic synchronization unit B modified the change fields f1 (business weight 0.5) and f3 (business weight 0.1), and the total weight of the two change fields f1 and f3 is 0.7. =0.6. The intersection field is f1, and the intersection weight is... =0.5. Business domain overlap rate = / =0.5 / 0.6≈0.83.

[0073] The operation timing overlap rate of a cell pair represents the degree of overlap between the write time windows of the two atomic synchronization units in the cell pair. The higher the overlap rate, the closer the write timing of the two atomic synchronization units is, and the greater the probability of concurrent conflicts. The write time window refers to the time window between the earliest and latest times corresponding to all changed fields of an atomic synchronization unit within the current synchronization cycle.

[0074] Specifically, for continuous operations with time windows, the formula for the operation timing overlap rate is: in, The operation timing overlap rate of the unit pair; Representative unit pairs of atomic synchronization units The latest time corresponding to the internal change field; Representative unit pairs of atomic synchronization units The latest time corresponding to the internal change field; Representative unit pairs of atomic synchronization units The earliest time corresponding to the internal change field; Representative unit pairs of atomic synchronization units The earliest time corresponding to the internal change field; Representative unit pairs of atomic synchronization units The time window; Representative unit pairs of atomic synchronization units The time window; The value range of is [0, 1]. For continuous operations with time windows, the time window of the two atomic synchronization units in a unit pair cannot be zero.

[0075] In this embodiment of the invention, for single-point timestamp operations without a clear time window, a negative exponential mapping calculation is performed based on the absolute value of the difference between the timestamps of the write operations of the two atomic synchronization units: in, The operation timing overlap rate of the unit pair; Represents an exponential function with the natural constant as its base; Representative unit pairs of atomic synchronization units The write timestamp; Representative unit pairs of atomic synchronization units The write timestamp; The preset time decay constant has a value ranging from 0.1 to 0.5. The value range of is [0, 1].

[0076] By weighted summing the exclusivity coefficient, modification permission overlap rate, business domain overlap rate, and operation timing overlap rate of a unit pair, the risk of concurrent conflicts between two atomic synchronization units in the unit pair can be quantified, yielding the business mutual exclusion degree of the unit pair. A higher business mutual exclusion degree indicates a greater risk of concurrent conflicts between the two corresponding atomic synchronization units, and a higher probability of business anomalies; conversely, a lower business mutual exclusion degree indicates a lower risk of concurrent conflicts between the two corresponding atomic synchronization units, and a lower probability of business anomalies.

[0077] Specifically, the formula for obtaining business mutual exclusion is: in, The mutual exclusion of services between unit pairs; The exclusive attribute coefficient representing the unit pair; The weight representing the exclusivity attribute coefficient; The overlap rate of modification permissions for representative unit pairs; The weight representing the overlap rate of modification permissions; The business domain overlap rate of the representative unit pair; Weights representing the overlap rate of business domains; The operation timing overlap rate of the unit pair; The weight represents the overlap rate of operation timing.

[0078] Weights and The weights are dynamically and adaptively adjusted based on the historical conflict identification accuracy: For each weight, a sample set of historical conflict cases and non-conflict concurrent cases is used, with "whether a conflict occurred" as the dependent variable and O, P, C, and T as independent variables. The contribution of each factor to the conflict is fitted through a logistic regression model, and the normalized result is directly used as the corresponding weight. The final calculated business mutual exclusion degree M ranges from [0, 1]. The larger the value, the stronger the business mutual exclusion between the two atomic synchronization units of the unit pair, and the higher the probability of causing business anomalies.

[0079] More specifically, the system continuously collects historical concurrent cases, including the four dimensions of O, P, C, and T corresponding to each concurrent operation, as well as the label indicating whether a real conflict occurred (1 for conflict, 0 for non-conflict). This data forms the training sample set. Using "whether a conflict occurred" as the dependent variable and O, P, C, and T as independent variables, a logistic regression model is used periodically (e.g., daily or weekly) to fit the model. After training, the regression coefficients for each dimension are obtained. , , , These coefficients reflect the direction and intensity of each dimension's influence on the occurrence of conflict, and their absolute values ​​can be regarded as the "original contribution" of each dimension. Taking the absolute value of each coefficient as the contribution, a normalization formula is used... =| | / (| |+| |+| |+| |) Calculate the initial weights for each dimension, ensuring that the sum of the four weights is 100%. Here, j is any one of O, P, C, and T.

[0080] The system evaluates the overall accuracy of the conflict identification model under the current weight configuration. Accuracy is calculated by comparing the model's predictions of historical samples with the actual labels. If the accuracy consistently exceeds the set target threshold (e.g., 90%), it indicates that the current weight allocation is reasonable, and the system will reduce the learning rate or maintain stability. If the accuracy falls below the target threshold, it indicates that the weight allocation has failed to adequately capture key dimensions, and the system will increase the learning rate to accelerate the weight update.

[0081] For regression coefficients that are negative (indicating a negative correlation between that dimension and the conflict), the system still calculates the contribution based on their absolute values, but records the anomaly for analysis. In the initial stages of the system or when the sample size is insufficient, preset empirical weights are used. =0.4, =0.2, =0.2, =0.2 as the starting value.

[0082] It should be understood that the method of using logistic regression models to fit the contribution of each factor to the conflict is a well-known technique, and will only be briefly introduced here without further elaboration.

[0083] In this embodiment of the invention, the mean and standard deviation of the business mutual exclusion degree of all unit pairs are calculated, and the value of [mean + 2 times standard deviation] is used as the preset mutual exclusion threshold. Specifically, to avoid the low reference value of the statistical quantity due to the small amount of data, this embodiment also sets a minimum threshold limit strategy, that is, the maximum value between [mean + 2 times standard deviation] and the basic value of the mutual exclusion threshold is used as the preset mutual exclusion threshold. The basic value of the mutual exclusion threshold can be 0.2, which is an empirical value obtained based on a large amount of historical data.

[0084] All atomic synchronization units within all unit pairs with a business mutual exclusion degree less than or equal to a preset mutual exclusion threshold are filtered as conflict-free synchronization units. All atomic synchronization units within all unit pairs with a business mutual exclusion degree greater than the preset mutual exclusion threshold are marked as conflict isolation units. These conflict isolation units constitute a conflict isolation set, fully preserving all data of the atomic synchronization units, including detailed calculations of the business mutual exclusion degree for each unit pair and markers of conflict reasons (exclusivity attribute, overlapping modification permissions, overlapping business domains, and overlapping operation sequences). The corresponding operations will be explained later in the section on data update and synchronization operations for changed fields.

[0085] Preferably, in some implementations of the present invention, the priority scheduling coefficient of each conflict-free synchronization unit is obtained based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, including: The time-series data includes the remaining time, total time, and historical timeout rate of the current business process; based on the ratio of the remaining time to the total time of the current business process, and the historical timeout rate, the time urgency index of the current business process is obtained. The maximum value of the business weight in all change fields within each conflict-free synchronization unit is used as the business importance index for each conflict-free synchronization unit. The priority scheduling coefficient of each conflict-free synchronization unit is obtained by weighted summing of the business importance index and the timing urgency index of the current business process.

[0086] The time urgency metric indicates the time urgency of the current business process synchronization task; the closer to the deadline, the higher the urgency. The business importance metric indicates the importance of each conflict-free synchronization unit to the current business process. The higher the priority scheduling coefficient obtained by combining the business importance metric of each conflict-free synchronization unit and the time urgency metric of the current business process, the higher the priority of the corresponding conflict-free synchronization unit; the lower the priority scheduling coefficient, the lower the priority of the corresponding conflict-free synchronization unit.

[0087] Specifically, the formula for obtaining the urgency index of the current business process is: in, The urgency metric representing the current business process, when When the calculated value is less than 0, the value is directly taken as 0; Represents the remaining timeframe of the current business process; Represents the total timeframe of the current business process; This represents the historical timeout rate of the current business process.

[0088] This indicates the percentage of the remaining timeframe of the current business process relative to the total timeframe of the current business process. This represents the current on-time probability of the current business process based on historical on-time rates. This represents the probability of timeout for the current business process, quantifying its urgency. The higher the probability of timeout, the more urgent the current business process.

[0089] Specifically, for any conflict-free synchronization unit D, the formula for obtaining its priority scheduling coefficient in the current business process is: in, This represents the priority scheduling coefficient of the conflict-free synchronization unit D in the current business process. An indicator representing the urgency of the current business process; The weights of the time urgency indicators representing the current business process; The business importance metric representing conflict-free synchronization unit D; The weights representing the importance index of conflict-free synchronization unit D services; This represents the time compensation factor. It should be noted that although the corresponding urgency index has the same value in the current business process, if there are multiple business processes undergoing changes within the current business entity, incorporating this dimension as a feature can assign an urgency level to all changed fields within the corresponding business process. This allows all units within the corresponding business process to be moved forward as a whole in the system's global queue, thereby ensuring that system resources are tilted towards business processes that are about to expire, guaranteeing the end-to-end business closure timeliness.

[0090] In one specific implementation of this invention, , The dynamic weights are adaptively calculated using the coefficient of variation method. The calculation formula is: in, The weights of the time urgency indicators representing the current business process; The coefficient of variation of the time urgency index representing the current business process; The coefficient of variation represents the importance index of conflict-free synchronization unit D services.

[0091] The coefficient of variation refers to the ratio of the standard deviation to the mean of the corresponding indicator data for the current business process in the system log over the last 10 changes. If the current business process has fewer than 10 changes in the system log, the default value is used. =1, =1.

[0092] It should be noted that, The method of obtaining and The method for obtaining the coefficient of variation is the same. and coefficient of variation When all values ​​are 0, to avoid the inability to use the above formula for calculation, this embodiment will use weights. and Set them to the same weight, i.e., all of them are 0.5.

[0093] It should be understood that the method of adaptively calculating dynamic weights using the coefficient of variation method is a well-known technique, and will only be briefly introduced here without further elaboration.

[0094] In one specific implementation of this invention, the time compensation factor The system calculates the waiting time of each conflict-free synchronization unit by normalizing the waiting time to prevent low-priority, non-urgent data from lingering in the queue and causing synchronization stalls. The system records the timestamp of each conflict-free synchronization unit entering the waiting queue and calculates its waiting time in real time (current time minus enqueue time). Then, the waiting time is compared with a preset maximum tolerable waiting time limit (e.g., 5 minutes based on the business SLA). =min(waited time / maximum tolerable waiting time, 1.0), and normalize the value so that the compensation factor falls within the range [0, 1]. This mechanism ensures that even tasks with low initial priority are not delayed indefinitely, but rather receive sufficient scheduling opportunities after waiting for a reasonable period. This fundamentally avoids the systemic stagnation problem caused by older tasks starving due to "a higher priority new task always arriving in the queue," thus guaranteeing the fairness and fluidity of the synchronization queue.

[0095] Preferably, in some implementations of the present invention, based on the priority scheduling coefficient and the basic risk characteristics in the changed field attribute data within the conflict-free synchronization unit, data update and synchronization are performed on each changed field, including: The maximum value of the conflict risk baseline in all changed field attribute data within each conflict-free synchronization unit is used as the conflict risk indicator for each conflict-free synchronization unit. Based on the priority scheduling coefficient and conflict risk index of each conflict-free synchronization unit, real-time reliable transmission link or batch asynchronous transmission link matching is performed on each conflict-free synchronization unit to obtain the transmission link matching result of each conflict-free synchronization unit. Based on the priority scheduling coefficient of each conflict-free synchronization unit, all conflict-free synchronization units are sorted to obtain the primary synchronization transmission queue; based on the transmission link matching result of each conflict-free synchronization unit, the primary synchronization transmission queue is sorted again to obtain the final synchronization transmission queue. Based on the final synchronization transmission queue, data updates and synchronization are performed on each changed field.

[0096] Based on the priority scheduling coefficient of each conflict-free synchronization unit, all conflict-free synchronization units are prioritized from high to low to obtain a primary synchronization transmission queue. According to the matching results, two queues are obtained: a real-time reliable transmission link queue and a batch asynchronous transmission link queue. The conflict-free synchronization units in the real-time reliable transmission link queue are rearranged according to the priority order in the primary synchronization transmission queue to ensure the real-time transmission of high-priority tasks; the conflict-free synchronization units in the batch asynchronous transmission link queue are also rearranged according to the priority order in the primary synchronization transmission queue. The real-time reliable transmission link queue is forced to precede the batch asynchronous transmission link queue to achieve priority isolation. After a second sorting, the final synchronization transmission queue is obtained.

[0097] Data updates are performed on each changed field according to the order of the final synchronization transmission queue. It should be understood that the final synchronization transmission queue consists of {a real-time reliable transmission link queue and a batch asynchronous transmission link queue}.

[0098] In this embodiment of the invention, for atomic synchronization units within a conflict isolation set, the system pushes the conflict flag, the complete data of the atomic synchronization unit, and the conflict cause to the corresponding modifier's terminal. After manual verification and confirmation of the final field data, the synchronization process is re-triggered to complete the data writing.

[0099] Furthermore, based on the priority scheduling coefficient and conflict risk index of each conflict-free synchronization unit, real-time reliable transmission link or batch asynchronous transmission link matching is performed on each conflict-free synchronization unit to obtain the transmission link matching result of each conflict-free synchronization unit, including: Conflict-free synchronization units with priority scheduling coefficients greater than or equal to the preset scheduling threshold and conflict risk indicators greater than or equal to the preset risk threshold are matched with real-time reliable transmission links; other conflict-free synchronization units that do not meet the above requirements are matched with batch asynchronous transmission links.

[0100] The real-time reliable transmission link enables a data packet fragmentation confirmation mechanism. The collision-free synchronization unit matched with the real-time reliable transmission link is an important and urgent synchronization task with a high risk of conflict, and priority is given to ensuring the transmission success rate and real-time performance. The collision-free synchronization unit matched with the batch asynchronous transmission link is a non-important and non-urgent synchronization task. The batch transmission reduces redundancy overhead and does not occupy core link resources.

[0101] In one specific implementation of this invention, the preset scheduling threshold is set to 0.6, and the preset risk threshold is set to 0.5. These values ​​are empirical values ​​obtained through experiments based on a large amount of historical data.

[0102] Based on the same inventive concept, the present invention also proposes a data synchronization device for an office system, the device comprising: The change data acquisition module is used to collect data on several changed fields and their attributes in the system log during the current synchronization period of the current business process when data changes occur under the conditions that trigger data synchronization, as well as the time-series data and concurrent operation data of the current business process. The atomic synchronization unit construction module is used to obtain the business weight of each changed field in the current business process based on the modification permissions of the operation user in the attribute data and the concurrent operation data, and to construct an atomic synchronization unit by combining the analysis of the closeness of the business linkage between each changed field and other changed fields in the current business process. The conflict-free synchronization unit filtering module is used to analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field based on the attribute data, and to filter the atomic synchronization units to obtain conflict-free synchronization units. The data update and synchronization module is used to obtain the priority scheduling coefficient of each conflict-free synchronization unit based on the numerical distribution of the business weight of the changed fields and the timeliness characteristics of the time-series data within each conflict-free synchronization unit; and to perform data update and synchronization for each changed field based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit.

[0103] The present invention also proposes a data-efficient synchronization storage medium for an office system, wherein the storage medium stores computer program code, and the computer program code, when run on a computer, enables the computer to implement any of the steps of the data-efficient synchronization method for an office system described above.

Claims

1. A method for efficient data synchronization in an office system, characterized in that, The method includes: Collect data from the system log of several changed fields and their attributes that have changed when data synchronization is triggered, as well as the time-series data and concurrent operation data of the current business process within the current synchronization period. Based on the modification permissions of the user in the attribute data and the concurrent operation data, the business weight of each changed field in the current business process is obtained. Combined with the analysis of the closeness of the business linkage between each changed field and other changed fields in the current business process, an atomic synchronization unit is constructed. Based on the attribute data, analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field, and filter the atomic synchronization units to obtain conflict-free synchronization units. Based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data, the priority scheduling coefficient of each conflict-free synchronization unit is obtained; based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit, data update and synchronization are performed on each changed field; The step of combining analysis of the closeness of business linkage between each changed field and other changed fields in the current business process to construct an atomic synchronization unit includes: Combine any two different change fields within the current business process into a field pair. The attribute data includes the dependency and linkage coefficient benchmark of each field pair under the current business process. For any pair of fields, based on the balance of the business weights of the two changed fields in the pair, the dependency linkage coefficient benchmark of each pair of fields is adjusted to obtain a dependency linkage coefficient that quantifies the degree of business linkage between each pair of fields. Based on the dependency linkage coefficient of each field pair, the field pairs are combined to obtain atomic synchronization units, including: classifying field pairs with dependency linkage coefficients greater than or equal to a preset dependency threshold into the same atomic synchronization unit; classifying the two changed fields in field pairs with dependency linkage coefficients less than the preset dependency threshold into different atomic synchronization units; marking changed fields with business weights greater than a preset weight threshold as exclusive fields; and for atomic synchronization units containing exclusive fields, merging all atomic synchronization units containing exclusive fields in the current business process into the same exclusive atomic synchronization unit. The step of analyzing the exclusivity and overlap rate of all atomic synchronization units containing the same change field based on the attribute data, and filtering the atomic synchronization units to obtain conflict-free synchronization units includes: The attribute data includes a change field ID. Based on each change field ID, all atomic synchronization units that do not have the same change field as other atomic synchronization units in the current synchronization cycle are filtered as conflict-free synchronization units. Based on each change field ID, all atomic synchronization units containing the same change field are grouped into a unit group; any two atomic synchronization units within a unit group are grouped into a unit pair. Based on whether there is an exclusive atomic synchronization unit in the two atomic synchronization units of each unit pair, determine the exclusivity attribute coefficient of each unit pair; Based on the degree of overlap between the modification permission scopes of the two atomic synchronization units corresponding to each unit pair in the current business process, determine the modification permission overlap rate of each unit pair. Based on the ratio of the business weights corresponding to the same changed fields in the two atomic synchronization units of each unit pair to the business weights corresponding to all changed fields in the two atomic synchronization units, the business domain overlap rate of each unit pair is determined. The operation timing overlap rate of each unit pair is determined based on the degree of overlap between the write time windows of the two atomic synchronization units of each unit pair; wherein, the write time window refers to the time window between the earliest and latest times corresponding to all changed fields of the atomic synchronization unit in the current synchronization cycle. The business mutual exclusion degree of each unit pair is obtained by weighted summing of the exclusive attribute coefficient, modification permission overlap rate, business domain overlap rate and operation sequence overlap rate. All atomic synchronization units within all unit pairs whose business mutual exclusion degree is less than or equal to the preset mutual exclusion threshold are selected as conflict-free synchronization units.

2. The method for efficient data synchronization in an office system according to claim 1, characterized in that, The step of obtaining the business weight of each changed field in the current business process based on the modification permissions of the user in the attribute data and the concurrent operation data includes: The attribute data includes the business weight benchmark of each changed field in the current business process; the concurrent operation data includes the modification permission range of each operating user in the current business process; Based on the modification permission range of the corresponding user for each changed field under the current data change operation, the dynamic correction coefficient of each changed field in the current business process is obtained; For each changed field, the business weight benchmark is adjusted based on a dynamic correction coefficient to obtain the business weight of each changed field in the current business process.

3. The method for efficient data synchronization in an office system according to claim 1, characterized in that, The step of obtaining the priority scheduling coefficient for each conflict-free synchronization unit based on the numerical distribution of the business weights of the changed fields within each conflict-free synchronization unit and the timeliness characteristics of the time-series data includes: The time-series data includes the remaining time, total time, and historical timeout rate of the current business process; based on the ratio of the remaining time to the total time of the current business process, and the historical timeout rate, the time urgency index of the current business process is obtained. The maximum value of the business weight in all change fields within each conflict-free synchronization unit is used as the business importance index for each conflict-free synchronization unit. The priority scheduling coefficient of each conflict-free synchronization unit is obtained by weighted summing of the business importance index and the timing urgency index of the current business process.

4. The method for efficient data synchronization in an office system according to claim 1, characterized in that, The step of updating and synchronizing data for each changed field based on the priority scheduling coefficient and the basic risk characteristics in the changed field attribute data within the conflict-free synchronization unit includes: The maximum value of the conflict risk baseline in all changed field attribute data within each conflict-free synchronization unit is used as the conflict risk indicator for each conflict-free synchronization unit. Based on the priority scheduling coefficient and conflict risk index of each conflict-free synchronization unit, real-time reliable transmission link or batch asynchronous transmission link matching is performed on each conflict-free synchronization unit to obtain the transmission link matching result of each conflict-free synchronization unit. Based on the priority scheduling coefficient of each conflict-free synchronization unit, all conflict-free synchronization units are sorted to obtain the primary synchronization transmission queue; based on the transmission link matching result of each conflict-free synchronization unit, the primary synchronization transmission queue is sorted again to obtain the final synchronization transmission queue. Based on the final synchronization transmission queue, data updates and synchronization are performed on each changed field.

5. The method for efficient data synchronization in an office system according to claim 4, characterized in that, The step of matching real-time reliable transmission links or batch asynchronous transmission links for each conflict-free synchronization unit based on its priority scheduling coefficient and conflict risk index, and obtaining the transmission link matching result for each conflict-free synchronization unit, includes: Conflict-free synchronization units with priority scheduling coefficients greater than or equal to the preset scheduling threshold and conflict risk indicators greater than or equal to the preset risk threshold are matched with real-time reliable transmission links; other conflict-free synchronization units that do not meet the above requirements are matched with batch asynchronous transmission links.

6. A high-efficiency data synchronization device for an office system, characterized in that, For implementing the steps of the efficient data synchronization method for an office system as described in any one of claims 1 to 5, the device includes: The change data acquisition module is used to collect data on several changed fields and their attributes in the system log during the current synchronization period of the current business process when data changes occur under the conditions that trigger data synchronization, as well as the time-series data and concurrent operation data of the current business process. The atomic synchronization unit construction module is used to obtain the business weight of each changed field in the current business process based on the modification permissions of the operation user in the attribute data and the concurrent operation data, and to construct an atomic synchronization unit by combining the analysis of the closeness of the business linkage between each changed field and other changed fields in the current business process. The conflict-free synchronization unit filtering module is used to analyze the exclusivity and overlap rate of all atomic synchronization units containing the same change field based on the attribute data, and to filter the atomic synchronization units to obtain conflict-free synchronization units. The data update and synchronization module is used to obtain the priority scheduling coefficient of each conflict-free synchronization unit based on the numerical distribution of the business weight of the changed fields and the timeliness characteristics of the time-series data within each conflict-free synchronization unit; and to perform data update and synchronization for each changed field based on the priority scheduling coefficient and the basic risk characteristics in the attribute data of the changed fields within the conflict-free synchronization unit.

7. A high-efficiency data synchronization storage medium for an office system, characterized in that, The storage medium stores computer program code, which, when executed on a computer, causes the computer to perform the steps of the efficient data synchronization method for an office system as described in any one of claims 1 to 5.

Images