A conflict resolution method, apparatus, device, medium, and computer program product

Abstract

The application provides a conflict resolution method, device, equipment, medium and computer program product, and relates to the technical field of hybrid IT architecture. The method comprises the following steps: in the case that it is determined that at least two first attribute values of a target configuration item from different data sources received by a configuration management database exist conflicts, determining that the first attribute values belong to a target attribute type in a plurality of preset attribute types; wherein each first attribute value corresponds to a data source; determining a target conflict resolution strategy of the target configuration item from a plurality of conflict resolution strategies according to the target attribute type; determining a target attribute value from the at least two first attribute values according to the target conflict resolution strategy; and storing the target attribute value into the configuration management database. Through targeted classification of the conflict data, hierarchical management of the conflict data is realized, and the dynamic conflict resolution rule is used to solve the problem that the existing technology relies on static rules and manual intervention, resulting in low management efficiency.

Description

Technical Field

[0001] This invention relates to the field of hybrid IT architecture technology, and in particular to a conflict resolution method, apparatus, device, medium, and computer program product. Background Technology

[0002] The Configuration Management Database (CMDB), as the cornerstone of an enterprise's IT service management system, is a structured database used to store all configuration items (CI) in the information technology environment. Originally a database for the service desk to "record assets," the CMDB has now become a "real-time battle map" for digital transformation. It needs to accurately describe all configuration items (CI) such as servers, databases, and middleware, and also provide dependencies, change paths, and performance snapshots between CIs, providing a trusted source for fault location, capacity planning, auditing, and automated operations and maintenance. The CMDB's configuration items cover various types, including servers, databases, middleware, business systems, networks, and security devices. The data comes from multiple independent systems, naturally leading to problems such as "discontinuity," "redundancy," "lag," and "duplication."

[0003] Existing technologies rely on static rules and human intervention to resolve conflicts, resulting in low governance efficiency, high risk, and dependence on manual intervention. Summary of the Invention

[0004] The purpose of this invention is to provide a conflict resolution method, apparatus, device, medium, and computer program product to solve the problems in the prior art where conflict resolution relies on static rules and manual intervention, resulting in low governance efficiency, high risk, and dependence on manual intervention.

[0005] To achieve the above objectives, embodiments of the present invention provide a conflict resolution method, comprising:

[0006] If it is determined that at least two first attribute values ​​of a target configuration item received by the configuration management database from different data sources conflict, the first attribute value is determined to belong to a target attribute type among multiple preset attribute types; wherein each first attribute value corresponds to one data source;

[0007] Based on the target attribute type, determine the target conflict resolution strategy for the target configuration item from among multiple conflict resolution strategies;

[0008] According to the target conflict resolution strategy, the target attribute value is determined from the at least two first attribute values;

[0009] The target attribute value is stored in the configuration management database.

[0010] Optionally, in the method, the target attribute type includes one or more of the following:

[0011] Unique attribute values; wherein, the unique attribute is an attribute value that is globally unique and cannot be easily changed;

[0012] Soft attribute values; wherein, the soft attribute values ​​are attribute values ​​related to business processes and personnel information;

[0013] Dynamic attribute values; wherein, the dynamic attribute values ​​are attribute values ​​that change over time and have time-sensitive characteristics.

[0014] Optionally, the method, wherein, when the target attribute type is a unique attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0015] Among the at least two first attribute values, a second attribute value is determined; wherein the second attribute value is the attribute value with the smallest difference between the timestamp and the current time among the at least two first attribute values;

[0016] Compare the hash value of the second attribute with the evidence storage hash value corresponding to the target configuration item in the blockchain;

[0017] If the hash value of the second attribute value matches the evidence storage hash value, the second attribute value is determined as the target attribute value; if the hash value of the second attribute value does not match the evidence storage hash value, the target attribute value is determined by a first confirmation instruction sent by the manual review terminal; wherein, the first confirmation instruction is obtained by sending the at least two first attribute values ​​to the manual review terminal.

[0018] Optionally, the method further includes:

[0019] The accuracy of the data source corresponding to the at least two first attribute values ​​is calculated based on the at least two first attribute values ​​and the target attribute value.

[0020] The dynamic weight value of the data source is adjusted according to the accuracy corresponding to the data source; wherein, the dynamic weight value is the weight value corresponding to the data source in the target conflict resolution strategy when the target attribute type is a dynamic attribute value.

[0021] Optionally, the method, wherein, when the target attribute type is a soft attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0022] Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source.

[0023] If, among the plurality of data sources corresponding to at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset absolute advantage threshold, the first attribute value corresponding to the target data source is determined as the target attribute value.

[0024] If, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than a preset absolute advantage threshold, and the difference between the second attribute value and the third attribute value is greater than a preset difference threshold, then the second attribute value is determined as the target attribute value; wherein, the second attribute value is the attribute value with the largest dynamic weight value among the data sources corresponding to the at least two first attribute values; and the third attribute value is the attribute value with the smallest difference from the second attribute value among the at least two attribute values.

[0025] If, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than a preset absolute advantage threshold, and the difference between the second attribute value and the third attribute value is less than or equal to a preset difference threshold, the target attribute value is determined by a second confirmation instruction sent by a manual review terminal; wherein, the second confirmation instruction is obtained by sending the at least two first attribute values ​​and the dynamic weight value of the data source to the manual review terminal.

[0026] Optionally, in the method, the quantitative indicator includes one or more of the following:

[0027] Historical accuracy of the data source, wherein the historical accuracy of the data source is the accuracy of the data source in reporting the attribute value corresponding to the target configuration item in a historical period;

[0028] Attribute value freshness, wherein the attribute value freshness is the freshness of the first attribute value corresponding to the data source;

[0029] Attribute value field completeness, wherein the attribute value field completeness is the completeness ratio of the field of the first attribute value corresponding to the data source;

[0030] Specifically, obtaining the dynamic weight value of each data source based on the quantification index of the data source corresponding to each of the first attribute values ​​includes:

[0031] Based on the historical accuracy of the data source, the freshness of the attribute value, and the completeness of the attribute value field of the data source corresponding to each first attribute value, a health score is obtained for each data source.

[0032] The health scores from multiple data sources are normalized to obtain the dynamic weight value for each data source.

[0033] Optionally, the method, wherein, when the target attribute type is a dynamic attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0034] Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source.

[0035] If, among the plurality of data sources corresponding to the at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset weight threshold, then the first attribute value corresponding to the target data source is determined as the fourth attribute value; if, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than the preset weight threshold, then the fourth attribute value is determined based on the compatibility sensitivity of the target configuration item.

[0036] The fourth attribute value is verified using a sandbox verification environment to determine the target attribute value.

[0037] Optionally, the method, wherein determining the fourth attribute value based on the compatibility sensitivity of the target configuration item, includes:

[0038] If the compatibility sensitivity of the target configuration item is greater than a preset sensitivity threshold, the fifth attribute value with the highest compatibility among the at least two first attribute values ​​shall be determined as the fourth attribute value.

[0039] If the compatibility sensitivity of the target configuration item is less than or equal to the preset sensitivity threshold, the weighted aggregate value of the at least two first attribute values ​​is determined as the fourth attribute value.

[0040] Optionally, the method, wherein verifying the fourth attribute value using a sandbox verification environment to determine the target attribute value includes:

[0041] The fourth attribute value is input into a sandbox verification environment isolated from the production environment for trial operation to obtain the configuration health index; wherein, the configuration health index is used to characterize the degree of adaptation of the fourth attribute value;

[0042] If the configured health index exceeds a preset percentage threshold of the baseline value, the fourth attribute value is determined as the target attribute value; if the configured health index does not exceed a preset percentage threshold of the baseline value, the target attribute value is determined by a third confirmation instruction sent by a manual review terminal; wherein, the third confirmation instruction is obtained by sending the at least two first attribute values ​​and the configured health index to the manual review terminal.

[0043] To achieve the above objectives, embodiments of the present invention also provide a conflict resolution apparatus, comprising:

[0044] The first processing module is configured to determine, when at least two first attribute values ​​of a target configuration item received from different data sources by the configuration management database conflict, that the first attribute value belongs to a target attribute type among a plurality of preset attribute types; wherein each first attribute value corresponds to one of the data sources;

[0045] The first determining module is used to determine the target conflict resolution strategy of the target configuration item from multiple conflict resolution strategies based on the target attribute type.

[0046] The second determining module is used to determine a target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy.

[0047] The second processing module is used to store the target attribute value into the configuration management database.

[0048] To achieve the above objectives, embodiments of the present invention also provide a conflict resolution device, comprising: a processor, a memory, and a program or instructions stored in the memory and executable on the processor; wherein, when the processor executes the program or instructions, it implements the conflict resolution method as described above.

[0049] To achieve the above objectives, embodiments of the present invention also provide a readable storage medium having a program or instructions stored thereon, wherein the program or instructions, when executed by a processor, implement the steps in the conflict resolution method described above.

[0050] To achieve the above objectives, embodiments of the present invention also provide a computer program product, which includes computer instructions that, when executed by a processor, implement the steps of the conflict resolution method described above.

[0051] The beneficial effects of the above-described technical solution of the present invention are as follows:

[0052] In this embodiment of the invention, when conflicts exist between first attribute values ​​of the same configuration item received by the configuration management database from different data sources, a corresponding target conflict resolution strategy is determined based on the target attribute type of the first attribute value. According to the target conflict resolution strategy, a target attribute value is determined from at least two first attribute values ​​and stored in the configuration management database. By determining the attribute type of the conflicting data, the conflicting data can be classified in a targeted manner. Determining the corresponding conflict resolution strategy based on the attribute type enables hierarchical governance of the conflicting data. The use of dynamic conflict resolution rules solves the problems of low governance efficiency, high risk, and reliance on manual intervention in existing technologies that rely on static rules and manual intervention. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the conflict resolution method described in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the conflict resolution system of the conflict resolution method described in the embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram illustrating the conflict resolution strategy for unique attribute values ​​in the conflict resolution method described in this embodiment of the invention.

[0056] Figure 4 This is a schematic diagram of the conflict resolution strategy for soft attribute values ​​in the conflict resolution method described in this embodiment of the invention;

[0057] Figure 5 This is a schematic diagram of the conflict resolution strategy for dynamic attribute values ​​in the conflict resolution method described in this embodiment of the invention;

[0058] Figure 6 This is a schematic diagram of the conflict resolution device described in an embodiment of the present invention. Detailed Implementation

[0059] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0060] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0061] In various embodiments of the present invention, it should be understood that the sequence number of each process described below does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0062] In addition, the terms "system" and "network" are often used interchangeably in this article.

[0063] In the embodiments provided by this invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0064] For ease of understanding, the following describes some aspects of the embodiments of the present invention:

[0065] like Figure 1 As shown, an embodiment of the present invention provides a conflict resolution method, which includes:

[0066] Step S10: If it is determined that at least two first attribute values ​​of the target configuration item received by the configuration management database from different data sources conflict, the first attribute value is determined to belong to a target attribute type among multiple preset attribute types; wherein each first attribute value corresponds to one data source.

[0067] It should be noted that the embodiments of the present invention also provide, for example... Figure 2 The conflict resolution system corresponding to the conflict resolution method shown adopts the idea of ​​attribute ternary classification and hierarchical governance to construct a method and system from conflict detection, intelligent resolution to effect verification. The entire conflict resolution system is divided into four modules: detection module, decision module, verification module, and optimization module. The overall logic is: detection -> classification -> (channel-specific) decision -> verification -> optimization.

[0068] The multi-source input in the detection module serves as the starting point for the conflict resolution system. This data can come from cloud platform APIs, monitoring agents, asset management systems, automated scripts, etc. These sources may report different values ​​for different attributes of the same configuration item. For example, for the operating system version attribute of a server, the cloud platform records the default value "CentOS 8.2" during installation, while the monitoring agent obtains the value "CentOS 7.9" through a query, and the asset management system may record "CentOS 7" based on the purchase order. The conflict resolution system employs a conflict degree assessment algorithm based on information entropy to quantitatively analyze the differences in attribute values ​​of the same configuration item across different data sources. The system calculates the standard deviation entropy value of each source data. When the entropy value exceeds a preset threshold, an attribute conflict is detected. Therefore, during the persistence to the configuration management database, at least two first attribute values ​​of the target configuration item from different data sources may conflict. In this case, the attribute classification engine in step 201 is needed to determine whether the first attribute value belongs to the target attribute type among multiple preset attribute types. Attribute Classification Engine: As the brain of the conflict resolution system, the attribute ternary classification engine is responsible for accurately classifying conflicts. It automatically divides the detected conflicts into three channels based on a predefined rule base: unique, dynamic, and soft.

[0069] Step S20: Based on the target attribute type, determine the target conflict resolution strategy for the target configuration item from among multiple conflict resolution strategies;

[0070] It should be noted that, as Figure 2 As shown, in the decision-making module, different target attribute types are matched with corresponding target conflict resolution strategies.

[0071] Step S30: Determine the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy;

[0072] It should be noted that, as Figure 2As shown, in the decision-making module: Addressing the core issue of conflict resolution, this embodiment of the invention abandons static rule strategies and proposes an innovative dynamic weight decision engine. This engine first calculates the Health Score (HS) for each data source's reporting behavior in response to multiple conflicting attribute values ​​of the same attribute for the same configuration item across different data sources. Then, it normalizes and converts each HS value into a dynamic weight, representing the relative confidence of each data source in the current conflict decision. Finally, based on the dynamic weight distribution pattern and attribute sensitivity, it outputs a resolution strategy through a predefined decision tree. For dynamic attribute values, the resolution strategy needs to be further validated in a sandbox environment using the Configuration Health Index (CHI); for soft attribute values, it enters a manual arbitration and arbitration feedback optimization process.

[0073] exist Figure 2 The verification module in the middle:

[0074] In step 204, the CHI sandbox verification process is conducted. To ensure the feasibility of the conflict resolution strategy, the attribute values ​​obtained through decision-making are sent to a sandbox verification environment isolated from the production environment for trial operation. The conflict resolution system mirrors a portion of the production traffic to the sandbox and calculates the Configuration Health Index (CHI) in real time.

[0075] The blockchain evidence storage module is a subsystem that provides tamper-proof and traceable trusted assurance for configuration item change operations. Through a layered evidence storage strategy, it writes different types of configuration item attribute changes into the blockchain network in the most suitable way, forming a trusted audit and traceability chain.

[0076] Step S40: Store the target attribute value in the configuration management database;

[0077] It should be noted that, ultimately, all authoritative data after governance—whether unique attributes confirmed through the storage chain, dynamic attributes verified through the sandbox, or soft attributes arbitrated by humans—will be written into the trusted CMDB in step 207. This ultimately builds a CMDB system that combines resilience, trustworthiness, and availability.

[0078] exist Figure 2 The optimization module employs a reinforcement learning framework, treating conflict resolution as a Markov decision process. The state space represents conflict feature vectors, the action space represents resolution strategies, and the reward function is designed based on CHI (Contingency Intelligence) changes. Through continuous learning, the conflict resolution system continuously improves the effectiveness of its resolution strategies. This module focuses on optimizing its processing flow, data source quality, and associated context.

[0079] In step 205, the reinforcement learning optimizer: the reinforcement learning component collects the success and failure results of sandbox verification and the final correct value of human arbitration as training data, with the goal of maximizing the long-term CHI value, and dynamically adjusts the weight coefficients (α, β, γ) in the data source health model and the threshold parameters in the decision tree using methods such as policy gradient.

[0080] In this embodiment, when there are conflicts between the first attribute values ​​of the same configuration item received by the configuration management database from different data sources, a corresponding target conflict resolution strategy is determined based on the target attribute type of the first attribute value. According to the target conflict resolution strategy, a target attribute value is determined from at least two first attribute values ​​and stored in the configuration management database. By determining the attribute type of the conflicting data, the conflicting data can be classified in a targeted manner. Determining the corresponding conflict resolution strategy based on the attribute type enables hierarchical governance of the conflicting data. The use of dynamic conflict resolution rules solves the problems of low governance efficiency, high risk, and reliance on manual intervention inherent in existing technologies that rely on static rules and manual intervention.

[0081] Optionally, in the method, the target attribute type includes one or more of the following:

[0082] Unique attribute values; wherein, the unique attribute is an attribute value that is globally unique and cannot be easily changed;

[0083] Soft attribute values; wherein, the soft attribute values ​​are attribute values ​​related to business processes and personnel information;

[0084] Dynamic attribute values; wherein, the dynamic attribute values ​​are attribute values ​​that change over time and have time-sensitive characteristics.

[0085] In this embodiment, unique attribute values, such as Internet Protocol (IP) addresses, Media Access Control (MAC) addresses, and asset numbers, possess global uniqueness and cannot be easily changed. Therefore, conflicts with these attributes indicate serious infrastructure problems, and a "strong consistency verification and protection" strategy is adopted to address such conflicts. These unique attribute values ​​include, but are not limited to, evidence of changes, including configuration item identifiers (IDs), attribute names, values ​​before and after the change, the operator, timestamps, and digital signatures.

[0086] Soft attribute values, such as responsible person, project, and cost center, typically involve business processes and personnel information and cannot be judged correctly or incorrectly using purely technical means. The conflict resolution system first uses HS (Helper, Controller, and Analyst) and dynamic weights to form a recommendation. If a sufficiently clear advantage structure is not formed, the system proceeds to the manual arbitration workflow. The soft attribute values ​​include, but are not limited to, a snapshot of the manual arbitration result, including the final decision value, the arbitrator, digital signature, and arbitration reasons.

[0087] Dynamic attribute values, such as operating system (OS) version, CPU usage, and memory usage, change dynamically and can vary significantly depending on the data source and collection method. The conflict resolution system uses a data source health score (HS) and normalized dynamic weights to weight and compare the reported values ​​from various conflicting data sources. Based on attribute sensitivity, it then selects a strategy of direct adoption, conservative adoption, or weighted aggregation. These dynamic attribute values ​​include, but are not limited to, change operation logs and digital signatures.

[0088] Optionally, the method, wherein, when the target attribute type is a unique attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0089] Among the at least two first attribute values, a second attribute value is determined; wherein the second attribute value is the attribute value with the smallest difference between the timestamp and the current time among the at least two first attribute values;

[0090] Compare the hash value of the second attribute with the evidence storage hash value corresponding to the target configuration item in the blockchain;

[0091] If the hash value of the second attribute value matches the evidence storage hash value, the second attribute value is determined as the target attribute value; if the hash value of the second attribute value does not match the evidence storage hash value, the target attribute value is determined by a first confirmation instruction sent by the manual review terminal; wherein, the first confirmation instruction is obtained by sending the at least two first attribute values ​​to the manual review terminal.

[0092] In this embodiment, such as Figure 3 As shown, in step 301, a unique attribute change application is determined to be unique based on the attribute classification engine and the attribute rule base.

[0093] In step 302, the blockchain history is verified, the hash value of `new_value` is calculated, and after detecting attribute conflicts, the attribute value is verified using its stored hash value in the blockchain. After verification, a new value is written and stored. This step has two important components: First, in step 303, the latest stored value in the blockchain is compared to `new_value`. Similarly, in... Figure 2 In step 202, a strong consistency processor determines whether the value is consistent, that is, whether the changed attribute value is consistent with the data stored in the blockchain. This step involves calculating the attribute value's new_value_hash and comparing it with the latest evidence stored in the blockchain. The hash value of the second attribute value is compared with the evidence hash value corresponding to the target configuration item in the blockchain. If they match, in step 304, a "confirmation record" is generated and uploaded to the blockchain. This generates an operation record containing the operator, unchanged value, and time, which is stored in the blockchain, but the attribute value remains unchanged. That is, if the hash value of the second attribute value matches the evidence hash value, the second attribute value is determined as the target attribute value. The second step: If the changed attribute value is inconsistent with the data stored in the blockchain, in step 307, the field is frozen and an alert is triggered to initiate manual approval. In step 308, a new value is written and notarized. The new value and its operation log after authorization and approval will be written to the blockchain as an immutable record. That is, if the hash value of the second attribute value is inconsistent with the notarized hash value, the target attribute value is determined by a first confirmation instruction sent by the manual review terminal. The first confirmation instruction is obtained after sending the at least two first attribute values ​​to the manual review terminal. In step 306, the value is finally written to the trusted CMDB.

[0094] Optionally, the method further includes:

[0095] The accuracy of the data source corresponding to the at least two first attribute values ​​is calculated based on the at least two first attribute values ​​and the target attribute value.

[0096] The dynamic weight value of the data source is adjusted according to the accuracy corresponding to the data source; wherein, the dynamic weight value is the weight value corresponding to the data source in the target conflict resolution strategy when the target attribute type is a dynamic attribute value.

[0097] In this embodiment, such as Figure 3As shown, in step 305, the reinforcement learning optimizer analyzes the accuracy and failure rate of unique attribute values ​​reported by each data source. For example, if it finds that the error rate of IP addresses reported by a certain monitoring agent is as high as 5%, it calculates the accuracy of the data source corresponding to the at least two first attribute values ​​based on the at least two first attribute values ​​and the target attribute value. It can automatically adjust the weight of this data in dynamic attribute decisions and trigger an alarm to notify operations personnel to check the health status of the agent, that is, adjust the dynamic weight value of the data source according to the accuracy corresponding to the data source.

[0098] Optionally, the method, wherein, when the target attribute type is a soft attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0099] Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source.

[0100] If, among the plurality of data sources corresponding to at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset absolute advantage threshold, the first attribute value corresponding to the target data source is determined as the target attribute value.

[0101] If, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than a preset absolute advantage threshold, and the difference between the second attribute value and the third attribute value is greater than a preset difference threshold, then the second attribute value is determined as the target attribute value; wherein, the second attribute value is the attribute value with the largest dynamic weight value among the data sources corresponding to the at least two first attribute values; and the third attribute value is the attribute value with the smallest difference from the second attribute value among the at least two attribute values.

[0102] If, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than a preset absolute advantage threshold, and the difference between the second attribute value and the third attribute value is less than or equal to a preset difference threshold, the target attribute value is determined by a second confirmation instruction sent by a manual review terminal; wherein, the second confirmation instruction is obtained by sending the at least two first attribute values ​​and the dynamic weight value of the data source to the manual review terminal.

[0103] In this embodiment, such as Figure 4 As shown, in step 401, a soft attribute change request is submitted, meaning the target attribute type is a soft attribute value. The conflict resolution system addresses the same configuration item c. iThe target attribute 'a' receives corresponding attribute value reports from multiple data sources S1, S2, ..., SN. In step 402, an initial credibility score is calculated based on the data source health model. For the i-th data source S... i At time t, for configuration item c i The system calculates the health score based on the attribute value reported by attribute 'a'. The calculation formula is:

[0104] ;

[0105] in, (Accuracy) represents the data source S i Historically, the accuracy rate of reporting similar soft-type properties; (Freshness) indicates the freshness of the data reported by this data source this time, and can preferably be expressed as: in Indicates the current time and data source S i The time difference between the most recent successful synchronization; (Completeness) indicates the completeness ratio of the fields reported by this data source this time; , , These are weighting coefficients, satisfying... Their initial values ​​can be set (e.g., can be set to...). , , The subsequent adjustments will be dynamically made by the reinforcement learning module.

[0106] The conflict resolution system further integrates the HS data from various data sources. i Normalized to dynamic weight value W i The calculation formula is:

[0107] ;

[0108] Where Wi represents the data source S i The relative confidence weights in this conflict decision-making process are calculated, and the sum of the weights of all data sources is 1. This indicates that among the multiple data sources corresponding to the at least two first attribute values, data source S is... j The health score is calculated by obtaining a dynamic weight value for each data source based on the quantitative indicators of the data source corresponding to each of the first attribute values. In this way, the system obtains not a fixed global score for a particular data source, but a dynamic credibility score of "the reporting behavior of a data source for a certain configuration item and a certain attribute value at the current moment," and forms the dynamic weight distribution in this conflict decision-making process accordingly.

[0109] After obtaining the dynamic weights of each data source, the conflict resolution system does not directly use the single highest score as the final result. Instead, it triggers different processing paths based on the weight distribution pattern. According to the scoring judgment, when the score is high—that is, when the dynamic weight value of a certain data source is greater than the preset absolute advantage threshold W_threshold—the system determines that the data source has a significant advantage in this round of conflict, directly adopts its corresponding attribute value as the recommended value, and in step 407, generates an operation log, writes the new value, and writes it to the CMDB. W_threshold is preferably configurable to 0.7. That is, if among the multiple data sources corresponding to at least two first attribute values, the dynamic weight value of the target data source is greater than the preset absolute advantage threshold, the first attribute value corresponding to the target data source is determined as the target attribute value.

[0110] When there is no absolutely dominant data source, but the difference between the highest dynamic weight and the second highest dynamic weight is greater than a preset difference threshold δ, the system determines that the current situation is a relative divergence. In this case, the system uses the attribute value corresponding to the highest dynamic weight as the recommended value and enters the automatic processing flow, recording the details of this divergence, the weight distribution, and the basis for the recommendation in the operation log; where δ is preferably configurable to 0.2. That is, if among the multiple data sources corresponding to at least two first attribute values, there is no target data source whose dynamic weight value is greater than the preset absolute dominance threshold, and the difference between the second attribute value and the third attribute value is greater than the preset difference threshold, the second attribute value is determined as the target attribute value; where the second attribute value is the attribute value with the largest dynamic weight value among the data sources corresponding to the at least two first attribute values; and the third attribute value is the attribute value with the smallest difference from the second attribute value among the at least two attribute values.

[0111] When the dynamic weight values ​​of each data source are relatively small and no clear dominant pattern has formed, the system determines that the current situation is one of serious disagreement and cannot automatically make a high-confidence decision. In cases of low scores, the system proceeds to the manual arbitration work order step. In step 403, an arbitration work order is automatically generated. The system automatically creates a standardized electronic work order, which includes conflict attributes, reported values ​​from each data source, and HS values ​​for each data source. i Details, dynamic weight distribution, and system recommendations are then pushed to designated responsible parties via the integrated ITSM platform. In step 404, the manual arbitration workflow begins. Similarly, in... Figure 2In step 206, an arbitration work order is automatically generated, followed by a manual arbitration workflow. The responsible person reviews the work order information and, based on business rules, organizational processes, and contextual information, makes a final ruling. The ruling is confirmed by the responsible person's electronic signature. The system uses the signed ruling as the final authoritative value and performs two operations: first, it generates a corresponding operation log and writes it to the trusted CMDB in step 406; second, in step 405, the reinforcement learning optimizer processes the ruling, original conflict data, and various data source HS... i The dynamic weight distribution and arbitration reasons are synchronized to the optimization module. Specifically, if none of the target data source's dynamic weight value is greater than a preset absolute advantage threshold among the multiple data sources corresponding to at least two first attribute values, and the difference between the second and third attribute values ​​is less than or equal to a preset difference threshold, the target attribute value is determined through a second confirmation instruction sent by the manual review terminal. This second confirmation instruction is obtained by sending the at least two first attribute values ​​and the dynamic weight value of the data source to the manual review terminal. In this step, the weight coefficients α, β, and γ are not fixed but are continuously optimized by the reinforcement learning module based on historical arbitration results. The system can also combine changes in soft attribute conflict rate, historical arbitration accuracy, field completeness rate, and audit pass status to form a business consistency assessment result, which serves as a feedback signal for subsequent adjustments to the data source health model parameters and arbitration threshold parameters. For example, if a conflict in the "responsible person" attribute is ultimately confirmed manually to have been changed from "Zhang San" to "Li Si", and subsequent audits and business process verifications confirm that the result is correct, the system will not only write "Li Si" into the CMDB, but also increase the weight of the data source that has consistently performed accurately in this type of attribute. Conversely, if a conflict is caused by outdated data in the HR system, the system can automatically lower the health score of the HR system in the corresponding soft attribute and trigger a data quality work order for that data source.

[0112] Optionally, in the method, the quantitative indicator includes one or more of the following:

[0113] Historical accuracy of the data source, wherein the historical accuracy of the data source is the accuracy of the data source in reporting the attribute value corresponding to the target configuration item in a historical period;

[0114] Attribute value freshness, wherein the attribute value freshness is the freshness of the first attribute value corresponding to the data source;

[0115] Attribute value field completeness, wherein the attribute value field completeness is the completeness ratio of the field of the first attribute value corresponding to the data source;

[0116] Specifically, obtaining the dynamic weight value of each data source based on the quantification index of the data source corresponding to each of the first attribute values ​​includes:

[0117] Based on the historical accuracy of the data source, the freshness of the attribute value, and the completeness of the attribute value field of the data source corresponding to each first attribute value, a health score is obtained for each data source.

[0118] The health scores from multiple data sources are normalized to obtain the dynamic weight value for each data source.

[0119] In this embodiment, for the i-th data source S i At time t, for configuration item c i The system calculates the health score based on the attribute value reported by attribute 'a'. The calculation formula is:

[0120] ;

[0121] in, (Accuracy) represents the data source S i The historical accuracy of reporting similar soft attributes (i.e., the historical accuracy of the data source). (Freshness) represents the freshness of the data reported by the data source in this instance (i.e., the freshness of the attribute value), and can preferably be expressed as: ,in Indicates the current time and data source S i The time difference between the most recent successful synchronization; (Completeness) indicates the completeness ratio of the fields reported by the data source this time (i.e., the completeness of the attribute value field). , , These are weighting coefficients, satisfying... Their initial values ​​can be set (e.g., can be set to...). , , The subsequent adjustments will be dynamically made by the reinforcement learning module.

[0122] The system further integrates the HS data from each data source. i Normalized to dynamic weight value W i The calculation formula is:

[0123] ;

[0124] Among them, W i Indicates data source S i In this conflict decision-making process, the relative confidence weight (i.e., the dynamic weight value) is used, and the sum of the weights of all data sources is 1. This indicates that among the multiple data sources corresponding to the at least two first attribute values, data source S is... j Health score.

[0125] Optionally, the method, wherein, when the target attribute type is a dynamic attribute value, determining the target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy includes:

[0126] Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source.

[0127] If, among the plurality of data sources corresponding to the at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset weight threshold, then the first attribute value corresponding to the target data source is determined as the fourth attribute value; if, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than the preset weight threshold, then the fourth attribute value is determined based on the compatibility sensitivity of the target configuration item.

[0128] The fourth attribute value is verified using a sandbox verification environment to determine the target attribute value.

[0129] In this embodiment, in step 501, a dynamic attribute change request is made, meaning the target attribute type is a dynamic attribute value. For example, a data report from a data source (such as a cloud platform, monitoring system, etc.) is triggered, and then the attribute classification engine identifies that the engine belongs to "dynamic" (such as CPU utilization, operating system version, etc.) and routes it to the dynamic processing channel.

[0130] In step 502, the data source health model is invoked to calculate the health score for each conflict source. Similarly, in Figure 2 In step 203, weight calculation is performed using the decision tree engine. The same configuration item c is collected. i Calculate the conflicting attribute values ​​corresponding to the target attribute 'a' across multiple data sources, and then calculate the health score HS for each data source. i :

[0131] ;

[0132] in, (Accuracy) represents the data source S i Historically, the accuracy rate of reporting similar dynamic attributes; (Freshness) indicates the freshness of the data reported by this data source in this instance; (Completeness) represents the completeness ratio of the fields reported by this data source this time. In step 503, HS is assigned as the initial weight, and the HS of each data source is... i Normalized to dynamic weight value W i The calculation formula is:

[0133] ;

[0134] HS from each data source i W obtained after normalization i This serves as the dynamic weight value. Specifically, the dynamic weight value for each data source is obtained based on the quantitative index of the data source corresponding to each of the first attribute values.

[0135] After determining the weight values, in step 504, the decision tree engine executes and begins operation. In step 505, it is determined whether the highest weight is greater than 0.9. First, it is determined whether there is an absolutely dominant data source (i.e., whether the dynamic weight value of this data source is higher than the 0.9 threshold). If so, in step 506, an operation log is generated and a new value is written, directly adopting the value of that source. That is, if among the multiple data sources corresponding to at least two first attribute values, there is a target data source whose dynamic weight value is greater than a preset weight threshold, then the first attribute value corresponding to the target data source is determined as the fourth attribute value.

[0136] If not, determine whether the attribute is extremely sensitive to compatibility (such as operating system version).

[0137] If yes, in step 508, if the determination of whether it is a compatibility-sensitive attribute is yes, the conservative intersection of each source value is taken. If the compatibility is extremely sensitive, a conservative strategy is adopted, and the safest one among the source values ​​(such as the lowest version number) is selected. If no, in step 507, if the determination of whether it is a compatibility-sensitive attribute is no, the weighted average of each source is calculated, and multiple conflicting attribute values ​​are compared or fused according to the dynamic weight of each data source to form an optimal result. That is, if there is no target data source with a dynamic weight value greater than the preset weight threshold among the multiple data sources corresponding to at least two first attribute values, the fourth attribute value is determined according to the compatibility sensitivity of the target configuration item.

[0138] In step 509, the CHI sandbox verification is performed. To ensure the feasibility of the results, the system does not directly write them to the production environment, but instead deploys them in a CHI sandbox verification environment isolated from the production environment. In step 510, a mitigation strategy is deployed and a portion of the production traffic is mirrored. The system mirrors a portion of the production traffic to the sandbox. In step 511, the Configuration Health Index (CHI) is monitored and calculated. The CHI is calculated in real time using the following formula:

[0139] ;

[0140] Wherein, R_cons represents the configuration consistency score, used to characterize the consistency of related fields, rules, and enumeration constraints after the optimization result is written; R_dep represents the dependency matching score, used to characterize the degree of consistency between the optimization result and the upstream and downstream configuration item dependencies, topological relationships, and support matrices; R_exec represents the execution compatibility verification score, used to characterize the pass rate of script execution, interface calls, probe verification, and compatibility checks related to the optimization result; R_obs represents the runtime observation stability score, used to characterize the degree of control over alarms, anomalies, and rollback events during the trial operation of mirrored traffic; P_risk represents the risk penalty item, used to characterize critical compatibility conflicts, serious alarms, or high-risk anomaly events; λ1, λ2, λ3, λ4, and λ5 are weighting coefficients.

[0141] In step 512, the system calculates the baseline health index CHI_base before deployment and the health index CHI_after after deployment, and calculates the health index improvement rate. :

[0142] .

[0143] Then make a judgment Is it greater than or equal to 15%? If the value exceeds 15% (e.g., 15%, which can be set as needed) and no critical risk access control condition is triggered, then in step 513, the policy verification is successful; in step 514, the new value is written to the production trusted CMDB, and is formally written into the production environment's trusted CMDB. That is, the fourth attribute value is verified using a sandbox verification environment to determine the target attribute value.

[0144] Optionally, the method, wherein determining the fourth attribute value based on the compatibility sensitivity of the target configuration item, includes:

[0145] If the compatibility sensitivity of the target configuration item is greater than a preset sensitivity threshold, the fifth attribute value with the highest compatibility among the at least two first attribute values ​​shall be determined as the fourth attribute value.

[0146] If the compatibility sensitivity of the target configuration item is less than or equal to the preset sensitivity threshold, the weighted aggregate value of the at least two first attribute values ​​is determined as the fourth attribute value.

[0147] In this embodiment, it is determined whether the target configuration item is extremely sensitive to compatibility (such as operating system version).

[0148] If yes, in step 508, if the determination of whether it is a compatibility-sensitive attribute is yes, a conservative intersection of the source values ​​is taken. If the compatibility sensitivity is extremely high, a conservative strategy is adopted, selecting the safest source value (such as the lowest version number). That is, if the compatibility sensitivity of the target configuration item is greater than a preset sensitivity threshold, the fifth attribute value with the highest compatibility among the at least two first attribute values ​​is determined as the fourth attribute value. If no, in step 507, if the determination of whether it is a compatibility-sensitive attribute is no, the weighted average value of each source is calculated, and multiple conflicting attribute values ​​are weighted and compared or weighted and fused according to the dynamic weights of each data source to form an optimal result. That is, if the compatibility sensitivity of the target configuration item is less than or equal to the preset sensitivity threshold, the weighted aggregate value of the at least two first attribute values ​​is determined as the fourth attribute value.

[0149] Optionally, the method, wherein verifying the fourth attribute value using a sandbox verification environment to determine the target attribute value includes:

[0150] The fourth attribute value is input into a sandbox verification environment isolated from the production environment for trial operation to obtain the configuration health index; wherein, the configuration health index is used to characterize the degree of adaptation of the fourth attribute value;

[0151] If the configured health index exceeds a preset percentage threshold of the baseline value, the fourth attribute value is determined as the target attribute value; if the configured health index does not exceed a preset percentage threshold of the baseline value, the target attribute value is determined by a third confirmation instruction sent by a manual review terminal; wherein, the third confirmation instruction is obtained by sending the at least two first attribute values ​​and the configured health index to the manual review terminal.

[0152] In this embodiment, such as Figure 5 As shown, in step 509, during CHI sandbox verification, to ensure the feasibility of the results, the system does not directly write the data to the production environment, but instead deploys it in a CHI sandbox verification environment isolated from the production environment. The system mirrors a portion of the production traffic to the sandbox and calculates the Configuration Health Index (CHI) in real time. The calculation formula is as follows:

[0153] ;

[0154] Wherein, R_cons represents the configuration consistency score, used to characterize the consistency of related fields, rules, and enumeration constraints after the optimization result is written; R_dep represents the dependency matching score, used to characterize the consistency between the optimization result and upstream and downstream configuration item dependencies, topological relationships, and support matrices; R_exec represents the execution compatibility verification score, used to characterize the pass rate of script execution, interface calls, probe verification, and compatibility checks related to the optimization result; R_obs represents the runtime observation stability score, used to characterize the degree of control over alarms, anomalies, and rollback events during the mirror traffic trial run; P_risk represents the risk penalty item, used to characterize critical compatibility conflicts, severe alarms, or high-risk anomaly events; λ1, λ2, λ3, λ4, and λ5 are weighting coefficients. The fourth attribute value is input into a sandbox verification environment isolated from the production environment for trial run to obtain the configuration health index.

[0155] In step 512, the system calculates the baseline health index CHI_base before deployment and the health index CHI_after after deployment, and calculates the health index improvement rate:

[0156] .

[0157] Then make a judgment Is it greater than or equal to 15%? If the value exceeds 15% (e.g., 15%, which can be set as needed) and no critical risk access control conditions are triggered, then in step 513, the policy verification is successful; in step 514, the new value is written to the production trusted CMDB, and is formally written into the production environment's trusted CMDB. That is, when the configured health index exceeds the preset percentage threshold of the baseline value, the fourth attribute value is determined as the target attribute value.

[0158] In step 515, policy verification fails; in step 516, the case is recorded and the policy is rolled back; in step 517, a manual arbitration process is triggered. If verification fails, the system automatically rolls back the policy, records the failure case for analysis, and escalates to a manual arbitration process. That is, if the configured health index does not exceed a preset percentage threshold of the baseline value, the target attribute value is determined through a third confirmation instruction sent by a manual review terminal.

[0159] In step 518, the reinforcement learning optimizer feeds the result back to the reinforcement learning optimizer, which is used in step 519 to update the weight parameters of the healthy model. By dynamically adjusting the weight parameters, adaptive optimization is achieved.

[0160] It should be noted that the key points of the embodiments of the present invention are as follows:

[0161] 1. Attribute Classification Engine: Used to classify conflicting configuration item attributes into unique, dynamic, and soft types based on a predefined rule base;

[0162] Layered governance engine: Performs strong consistency processing based on blockchain consensus on unique attributes;

[0163] Initiate manual arbitration work order flow for soft attributes;

[0164] For dynamic attributes, the dynamic weight engine is invoked to generate a resolution strategy, which is then sent to a sandbox environment for verification.

[0165] The blockchain evidence storage module is used to immutably store change records of unique attributes and arbitration results of soft attributes.

[0166] 2. Based on a predefined rule base and categorizing attributes into three categories (unique, dynamic, and soft), detected conflicts are automatically classified into three channels. The classification rules for these attributes are as follows:

[0167] If an attribute is a globally unique identifier, it is classified as a unique type;

[0168] If the attribute is numerical or version-based, affects system stability, and is subject to change, it is classified as dynamic.

[0169] If the attribute is text-based and depends on business rules, it is classified as soft.

[0170] 3. Dynamic attribute governance, the dynamic weight decision engine includes:

[0171] Data source health model: used to calculate the health score (HS) for each data source's current reporting behavior for multiple conflicting reported values ​​of the same configuration target attribute in different data sources;

[0172] Dynamic weight calculation unit: used to normalize each health score HS into a dynamic weight W to characterize the relative confidence of each data source in this conflict decision;

[0173] Manual arbitration workflow: used to automatically generate standardized work orders containing conflicting data, HS details of each data source, dynamic weight distribution and recommendations when no clearly dominant data source is formed, and route them to the preset responsible person. After the responsible person signs the ruling, it is written into the CMDB and stored on the blockchain.

[0174] Decision tree engine: Based on weight distribution and attribute sensitivity, it outputs strategies that adopt high weights, weighted averages, or conservative values.

[0175] Sandbox Validation Unit: Used to deploy the best results to an isolated sandbox environment for dynamic attributes, calculate the configuration health index (CHI), and determine whether to allow the association's trusted CMDB based on the health index improvement rate.

[0176] 4. Data Source Health Model Score: The credibility score (HS) of the data source health model is calculated by weighting the indicators of three core dimensions:

[0177] ;

[0178] in, (Accuracy) represents the data source S i Historically, the accuracy rate of reporting similar dynamic attributes; (Freshness) indicates the freshness of the data reported by this data source in this instance; (Completeness) indicates the completeness ratio of the fields reported by this data source this time.

[0179] 5. The sandbox verification environment operates through the following steps:

[0180] Configuring a cloned production environment;

[0181] Deploy the resolution strategy in the sandbox;

[0182] Mirror a portion of the production traffic to the sandbox;

[0183] Calculate the configuration health of CHI:

[0184] ;

[0185] Wherein, R_cons represents the configuration consistency score, used to characterize the consistency of related fields, rules, and enumeration constraints after the optimization result is written; R_dep represents the dependency matching score, used to characterize the degree of consistency between the optimization result and the upstream and downstream configuration item dependencies, topological relationships, and support matrices; R_exec represents the execution compatibility verification score, used to characterize the pass rate of script execution, interface calls, probe verification, and compatibility checks related to the optimization result; R_obs represents the runtime observation stability score, used to characterize the degree of control over alarms, anomalies, and rollback events during the trial operation of mirrored traffic; P_risk represents the risk penalty item, used to characterize critical compatibility conflicts, serious alarms, or high-risk anomaly events; λ1, λ2, λ3, λ4, and λ5 are weighting coefficients. If the policy is approved, then the policy is approved; otherwise, the policy is rolled back.

[0186] 6. Manual arbitration workflow:

[0187] Automatically generate workflows that include conflict data and data health scores;

[0188] The work order is routed to the designated responsible person, the electronic signature of the responsible person is received, and the decision result and signature are written into the blockchain for evidence storage.

[0189] 7. Reinforcement Learning Module:

[0190] Used to collect sandbox verification results and human arbitration results;

[0191] With the goal of improving the Health Index (CHI), the coefficients in the dynamic weight calculation unit are dynamically adjusted. .

[0192] The advantages of the embodiments of the present invention are as follows:

[0193] This invention introduces the concepts of attribute ternary classification and hierarchical governance, constructing a complete closed-loop system from conflict detection and intelligent resolution to effect verification and continuous optimization. Compared with existing technologies, this invention first calculates the health score (HS) of the current reporting behavior of each data source for multiple conflicting attribute values ​​of the same attribute of the same configuration item in different data sources, and forms dynamic weights through normalization. This enables the system to adaptively evaluate the credibility of data sources under different times, attributes, and source conditions, rather than relying on fixed priorities or static rules.

[0194] For soft attributes, this invention transforms the previously purely manual judgment process into a quantifiable, traceable, and sustainably corrective governance process through "dynamic weight recommendation + manual arbitration + arbitration feedback optimization". For dynamic attributes, this invention adds a CHI sandbox verification step after the optimal results are formed, upgrading the verification method from traditional result comparison to process simulation and effect quantification, thereby bringing operational risks forward to the isolated environment.

[0195] In summary, this invention solves the problem of inconsistent governance granularity for different attribute types through "attribute ternary classification", solves the problem of credibility allocation in multi-data source conflict scenarios through "HS-driven dynamic weight decision", solves the problem of security and validity verification before dynamic attribute write-back through "CHI sandbox verification", and forms a credible, closed-loop, and iterative CMDB multi-source conflict governance system through blockchain notarization and continuous optimization modules.

[0196] like Figure 6 As shown, to achieve the above objectives, embodiments of the present invention also provide a conflict resolution device, comprising:

[0197] The first processing module 601 is configured to determine that the first attribute value belongs to a target attribute type among multiple preset attribute types when at least two first attribute values ​​of a target configuration item received from different data sources by the configuration management database conflict; wherein each first attribute value corresponds to one data source.

[0198] The first determining module 602 is used to determine the target conflict resolution strategy of the target configuration item among multiple conflict resolution strategies based on the target attribute type.

[0199] The second determining module 603 is used to determine a target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy.

[0200] The second processing module 604 is used to store the target attribute value into the configuration management database.

[0201] Optionally, in the aforementioned apparatus, the target attribute type includes one or more of the following:

[0202] Unique attribute values; wherein, the unique attribute is an attribute value that is globally unique and cannot be easily changed;

[0203] Soft attribute values; wherein, the soft attribute values ​​are attribute values ​​related to business processes and personnel information;

[0204] Dynamic attribute values; wherein, the dynamic attribute values ​​are attribute values ​​that change over time and have time-sensitive characteristics.

[0205] Optionally, in the aforementioned apparatus, when the target attribute type is a unique attribute value, the second determining module 603 includes:

[0206] The first determining unit is configured to determine a second attribute value among the at least two first attribute values; wherein the second attribute value is the attribute value with the smallest difference between the timestamp and the current time among the at least two first attribute values;

[0207] The first processing unit is used to compare the hash value of the second attribute value with the evidence storage hash value corresponding to the target configuration item in the blockchain;

[0208] The second determining unit is configured to determine the second attribute value as the target attribute value when the hash value of the second attribute value is consistent with the evidence storage hash value; and to determine the target attribute value through a first confirmation instruction sent by a manual review terminal when the hash value of the second attribute value is inconsistent with the evidence storage hash value; wherein the first confirmation instruction is obtained after sending the at least two first attribute values ​​to the manual review terminal.

[0209] Optionally, the device further includes:

[0210] The first processing module is used to calculate the accuracy of the data source corresponding to the at least two first attribute values ​​based on the at least two first attribute values ​​and the target attribute value.

[0211] The second processing module is used to adjust the dynamic weight value of the data source according to the accuracy corresponding to the data source; wherein, the dynamic weight value is the weight value of the data source in the target conflict resolution strategy when the target attribute type is a dynamic attribute value.

[0212] Optionally, in the aforementioned apparatus, when the target attribute type is a soft attribute value, the second determining module 603 includes:

[0213] The first acquisition unit is configured to acquire the dynamic weight value of each data source based on the quantitative index of the data source corresponding to each first attribute value; wherein the quantitative index is used to characterize the reliability of the data source.

[0214] The third determining unit is used to determine the first attribute value corresponding to the target data source as the target attribute value when the dynamic weight value of the target data source is greater than a preset absolute advantage threshold among the plurality of data sources corresponding to the at least two first attribute values.

[0215] The fourth determining unit is configured to determine the second attribute value as the target attribute value when, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than a preset absolute dominance threshold, and the difference between the second attribute value and the third attribute value is greater than a preset difference threshold; wherein, the second attribute value is the attribute value with the largest dynamic weight value among the data sources corresponding to the at least two first attribute values; and the third attribute value is the attribute value with the smallest difference from the second attribute value among the at least two attribute values.

[0216] The fifth determining unit is configured to determine the target attribute value by a second confirmation instruction sent by a manual review terminal when there is no target data source whose dynamic weight value is greater than a preset absolute advantage threshold among the plurality of data sources corresponding to the at least two first attribute values, and the difference between the second attribute value and the third attribute value is less than or equal to a preset difference threshold; wherein the second confirmation instruction is obtained by sending the at least two first attribute values ​​and the dynamic weight value of the data source to the manual review terminal.

[0217] Optionally, in the aforementioned apparatus, the quantification index includes one or more of the following:

[0218] Historical accuracy of the data source, wherein the historical accuracy of the data source is the accuracy of the data source in reporting the attribute value corresponding to the target configuration item in a historical period;

[0219] Attribute value freshness, wherein the attribute value freshness is the freshness of the first attribute value corresponding to the data source;

[0220] Attribute value field completeness, wherein the attribute value field completeness is the completeness ratio of the field of the first attribute value corresponding to the data source;

[0221] The first acquisition unit includes:

[0222] The first acquisition component is used to acquire a health score for each data source based on the historical accuracy of the data source, the freshness of the attribute value, and the completeness of the attribute value field of the data source corresponding to each first attribute value.

[0223] The second acquisition component is used to normalize the health scores of multiple data sources and obtain the dynamic weight value of each data source.

[0224] Optionally, in the aforementioned apparatus, when the target attribute type is a dynamic attribute value, the second determining module 603 includes:

[0225] The second acquisition unit is used to acquire the dynamic weight value of each data source according to the quantization index of the data source corresponding to each first attribute value; wherein the quantization index is used to characterize the reliability of the data source.

[0226] The sixth determining unit is configured to: if, among the plurality of data sources corresponding to the at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset weight threshold, then determine the first attribute value corresponding to the target data source as the fourth attribute value; if, among the plurality of data sources corresponding to the at least two first attribute values, there does not exist a target data source whose dynamic weight value is greater than the preset weight threshold, then determine the fourth attribute value based on the compatibility sensitivity of the target configuration item.

[0227] The seventh determining unit is used to verify the fourth attribute value using a sandbox verification environment and determine the target attribute value.

[0228] Optionally, in the aforementioned apparatus, the sixth determining unit comprises:

[0229] A first determining component is configured to determine the fifth attribute value with the highest compatibility among the at least two first attribute values ​​as the fourth attribute value when the compatibility sensitivity of the target configuration item is greater than a preset sensitivity threshold.

[0230] The second determining component is used to determine the weighted aggregate value of the at least two first attribute values ​​as the fourth attribute value when the compatibility sensitivity of the target configuration item is less than or equal to the preset sensitivity threshold.

[0231] Optionally, in the aforementioned apparatus, the seventh determining unit comprises:

[0232] The third acquisition component is used to input the fourth attribute value into a sandbox verification environment isolated from the production environment for trial operation and to acquire the configuration health index; wherein, the configuration health index is used to characterize the adaptability of the fourth attribute value;

[0233] The third determining component is used to determine the fourth attribute value as the target attribute value when the configured health index exceeds a preset proportion threshold of the baseline value; and to determine the target attribute value through a third confirmation instruction sent by a manual review terminal when the configured health index does not exceed the preset proportion threshold of the baseline value; wherein the third confirmation instruction is obtained by sending the at least two first attribute values ​​and the configured health index to the manual review terminal.

[0234] It should be noted that the apparatus provided in this embodiment of the invention can implement all the method steps implemented in the above method embodiment and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.

[0235] To achieve the above objectives, embodiments of the present invention also provide a conflict resolution device, comprising: a processor, a memory, and a program or instructions stored in the memory and executable on the processor; wherein, when the processor executes the program or instructions, it implements the conflict resolution method as described above.

[0236] To achieve the above objectives, embodiments of the present invention also provide a readable storage medium having a program or instructions stored thereon, wherein the program or instructions, when executed by a processor, implement the steps in the conflict resolution method described above.

[0237] To achieve the above objectives, embodiments of the present invention also provide a computer program product, which includes computer instructions that, when executed by a processor, implement the steps of the conflict resolution method described above.

[0238] It should be further noted that the terminals described in this specification include, but are not limited to, smartphones, tablets, etc., and many of the functional components described are referred to as modules in order to emphasize the independence of their implementation.

[0239] In this embodiment of the invention, the module can be implemented in software so that it can be executed by various types of processors. For example, an identified executable code module may include one or more physical or logical blocks of computer instructions, which may be constructed as objects, procedures, or functions. Nevertheless, the executable code of the identified module does not need to be physically located together, but may include different instructions stored in different bits, which, when logically combined, constitute the module and achieve the module's intended purpose.

[0240] In practice, an executable code module can be a single instruction or many instructions, and can even be distributed across multiple different code segments, different programs, and across multiple memory devices. Similarly, operational data can be identified within the module and can be implemented in any suitable form and organized within any suitable data structure. This operational data can be collected as a single dataset or distributed across different locations (including different storage devices), and can exist, at least in part, solely as electronic signals within the system or network.

[0241] When a module can be implemented using software, considering the current level of hardware technology, modules that can be implemented in software can be implemented using hardware circuits by those skilled in the art to achieve the corresponding functions, without considering cost. These hardware circuits include conventional very-large-scale integrated circuits (VLSI) or gate arrays, as well as existing semiconductors such as logic chips and transistors, or other discrete components. Modules can also be implemented using programmable hardware devices, such as field-programmable gate arrays, programmable array logic, and programmable logic devices.

[0242] The exemplary embodiments described above are with reference to the accompanying drawings. Many different forms and embodiments are feasible without departing from the spirit and teachings of the invention. Therefore, the invention should not be construed as limiting the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided to make the invention complete and convey the scope of the invention to those skilled in the art. In these drawings, component dimensions and relative dimensions may be exaggerated for clarity. The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. As used herein, unless clearly indicated otherwise, the singular forms “a,” “an,” and “the” are intended to include all such forms. It will be further understood that the terms “comprising” and / or “including”, when used in this specification, indicate the presence of the stated features, integers, steps, operations, components, and / or elements, but do not exclude the presence or addition of one or more other features, integers, steps, operations, components, and / or groups thereof. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of the range and any subranges in between.

[0243] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A conflict resolution method, characterized in that, include: If it is determined that at least two first attribute values ​​of a target configuration item received by the configuration management database from different data sources conflict, the first attribute value is determined to belong to a target attribute type among multiple preset attribute types; wherein each first attribute value corresponds to one data source; Based on the target attribute type, determine the target conflict resolution strategy for the target configuration item from among multiple conflict resolution strategies; According to the target conflict resolution strategy, the target attribute value is determined from the at least two first attribute values; The target attribute value is stored in the configuration management database.

2. The method according to claim 1, characterized in that, The target attribute type includes one or more of the following: Unique attribute values; wherein, the unique attribute is an attribute value that is globally unique and cannot be easily changed; Soft attribute values; wherein, the soft attribute values ​​are attribute values ​​related to business processes and personnel information; Dynamic attribute values; wherein, the dynamic attribute values ​​are attribute values ​​that change over time and have time-sensitive characteristics.

3. The method according to claim 1 or 2, characterized in that, When the target attribute type is a unique attribute value, the target attribute value is determined from the at least two first attribute values ​​according to the target conflict resolution strategy, including: Among the at least two first attribute values, a second attribute value is determined; wherein the second attribute value is the attribute value with the smallest difference between the timestamp and the current time among the at least two first attribute values; Compare the hash value of the second attribute with the evidence storage hash value corresponding to the target configuration item in the blockchain; If the hash value of the second attribute value matches the evidence storage hash value, the second attribute value is determined as the target attribute value; if the hash value of the second attribute value does not match the evidence storage hash value, the target attribute value is determined by a first confirmation instruction sent by the manual review terminal; wherein, the first confirmation instruction is obtained by sending the at least two first attribute values ​​to the manual review terminal.

4. The method according to claim 3, characterized in that, The method further includes: The accuracy of the data source corresponding to the at least two first attribute values ​​is calculated based on the at least two first attribute values ​​and the target attribute value. The dynamic weight value of the data source is adjusted according to the accuracy corresponding to the data source; wherein, the dynamic weight value is the weight value corresponding to the data source in the target conflict resolution strategy when the target attribute type is a dynamic attribute value.

5. The method according to claim 1 or 2, characterized in that, When the target attribute type is a soft attribute value, the target attribute value is determined from the at least two first attribute values ​​according to the target conflict resolution strategy, including: Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source. If, among the plurality of data sources corresponding to at least two first attribute values, the dynamic weight value of the target data source is greater than a preset absolute dominance threshold, the first attribute value corresponding to the target data source is determined as the target attribute value. If, among the plurality of data sources corresponding to at least two first attribute values, the dynamic weight value of the target data source is not greater than the preset absolute dominance threshold, and the difference between the second attribute value and the third attribute value is greater than a preset difference threshold, the second attribute value is determined as the target attribute value. The second attribute value is the attribute value with the largest dynamic weight value among the data sources corresponding to the at least two first attribute values. The third attribute value is the attribute value with the smallest difference from the second attribute value among the at least two attribute values. If, among the plurality of data sources corresponding to at least two first attribute values, the dynamic weight value of the target data source is not greater than the preset absolute dominance threshold, and the difference between the second attribute value and the third attribute value is less than or equal to a preset difference threshold, the target attribute value is determined by a second confirmation instruction sent by a manual review terminal. The second confirmation instruction is obtained by sending the at least two first attribute values ​​and the dynamic weight value of the data source to the manual review terminal.

6. The method according to claim 5, characterized in that, The quantitative indicators include one or more of the following: Historical accuracy of the data source, wherein the historical accuracy of the data source is the accuracy of the data source in reporting the attribute value corresponding to the target configuration item in a historical period; Attribute value freshness, wherein the attribute value freshness is the freshness of the first attribute value corresponding to the data source; Attribute value field completeness, wherein the attribute value field completeness is the completeness ratio of the field of the first attribute value corresponding to the data source; Specifically, obtaining the dynamic weight value of each data source based on the quantification index of the data source corresponding to each of the first attribute values ​​includes: Based on the historical accuracy of the data source, the freshness of the attribute value, and the completeness of the attribute value field of the data source corresponding to each first attribute value, a health score is obtained for each data source. The health scores from multiple data sources are normalized to obtain the dynamic weight value for each data source.

7. The method according to claim 1 or 2, characterized in that, When the target attribute type is a dynamic attribute value, the target attribute value is determined from the at least two first attribute values ​​according to the target conflict resolution strategy, including: Based on the quantitative index of the data source corresponding to each of the first attribute values, a dynamic weight value for each data source is obtained; wherein, the quantitative index is used to characterize the reliability of the data source. If, among the plurality of data sources corresponding to the at least two first attribute values, there exists a target data source whose dynamic weight value is greater than a preset weight threshold, then the first attribute value corresponding to the target data source is determined as the fourth attribute value; if, among the plurality of data sources corresponding to the at least two first attribute values, there is no target data source whose dynamic weight value is greater than the preset weight threshold, then the fourth attribute value is determined based on the compatibility sensitivity of the target configuration item. The fourth attribute value is verified using a sandbox verification environment to determine the target attribute value.

8. The method according to claim 7, characterized in that, The fourth attribute value is determined based on the compatibility sensitivity of the target configuration item, including: If the compatibility sensitivity of the target configuration item is greater than a preset sensitivity threshold, the fifth attribute value with the highest compatibility among the at least two first attribute values ​​shall be determined as the fourth attribute value. If the compatibility sensitivity of the target configuration item is less than or equal to the preset sensitivity threshold, the weighted aggregate value of the at least two first attribute values ​​is determined as the fourth attribute value.

9. The method according to claim 7, characterized in that, The fourth attribute value is verified using a sandbox verification environment to determine the target attribute value, including: The fourth attribute value is input into a sandbox verification environment isolated from the production environment for trial operation to obtain the configuration health index; wherein, the configuration health index is used to characterize the degree of adaptation of the fourth attribute value; If the configured health index exceeds a preset percentage threshold of the baseline value, the fourth attribute value is determined as the target attribute value; if the configured health index does not exceed a preset percentage threshold of the baseline value, the target attribute value is determined by a third confirmation instruction sent by a manual review terminal; wherein, the third confirmation instruction is obtained by sending the at least two first attribute values ​​and the configured health index to the manual review terminal.

10. A conflict resolution device, characterized in that, include: The first processing module is configured to determine, when at least two first attribute values ​​of a target configuration item received from different data sources by the configuration management database conflict, that the first attribute value belongs to a target attribute type among a plurality of preset attribute types; wherein each first attribute value corresponds to one of the data sources; The first determining module is used to determine the target conflict resolution strategy of the target configuration item from multiple conflict resolution strategies based on the target attribute type. The second determining module is used to determine a target attribute value from the at least two first attribute values ​​according to the target conflict resolution strategy. The second processing module is used to store the target attribute value into the configuration management database.

11. A conflict resolution device, comprising: A processor, a memory, and a program or instructions stored in the memory and executable on the processor; characterized in that, when the processor executes the program or instructions, it implements the conflict resolution method as described in any one of claims 1-9.

12. A readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instructions are executed by the processor, they implement the steps of the conflict resolution method as described in any one of claims 1-9.

13. A computer program product, characterized in that, Includes computer instructions, which, when executed by a processor, implement the steps of the conflict resolution method as described in any one of claims 1-9.

Images