A topology arrangement and operation and maintenance data integrated system and method

Abstract

The present disclosure relates to the technical field of distributed orchestration, and particularly provides a topology orchestration and operation and maintenance data integrated system and method. The system comprises: a topology orchestration management module, which constructs a grid layered topology graph based on a layer column constraint framework, and layer columns cross to form grid cells; generates grid nodes according to dynamic Boolean query conditions and resource multi-dimensional attributes to orchestrate each grid cell; and converges multiple paths with the same source and target into topology edges. An operation and maintenance data unified access module acquires operation and maintenance alarm data within a dynamic time window and matches a corresponding grid node by using string similarity. A topology template dynamic generation module performs feature clustering on historical topology configuration data to generate a parameterized template associated with a business type; and dynamically generates a target topology graph according to the business type of a target business by calling a parameterized template of the corresponding type. The present disclosure can realize a full-process automatic closed loop of node screening, cross-layer topology automatic orchestration, alarm automatic mounting and topology template reuse.

Description

Technical Field

[0001] This disclosure relates to the field of distributed orchestration technology, and in particular to an integrated system and method for topology orchestration and operation and maintenance data. Background Technology

[0002] With the rapid development of mobile cloud, the number of microservices within the system is growing exponentially, the architecture is changing dynamically, and the dependencies between business modules are becoming increasingly complex, resulting in a dramatic increase in the amount of observation data. Topology views, as a means of intuitively presenting the system structure, are crucial for fault location and potential problem detection.

[0003] While the topology construction process has been simplified to some extent in related technologies, there are still significant shortcomings in node selection, cross-layer orchestration, data integration, and template expansion. Summary of the Invention

[0004] This disclosure is made in view of the above-mentioned problems. This disclosure provides an integrated system and method for topology orchestration and operation and maintenance data.

[0005] According to one aspect of this disclosure, an integrated system for topology orchestration and operation and maintenance data is provided, comprising: a topology orchestration management module configured to construct a gridded hierarchical topology graph with grid nodes and topology edges based on a resource-based hierarchical constraint framework, wherein the layers and columns of the hierarchical constraint framework form multiple grid cells at their intersections; to orchestrate the nodes of each grid cell according to dynamically configured Boolean query conditions and multidimensional attributes of resources, generating grid nodes; to converge multiple topology paths with the same source node and target node between multiple grid nodes according to a preset strategy, obtaining topology edges; a unified operation and maintenance data access module configured to acquire key operation and maintenance alarm data of each grid node within a dynamic time window, and to mount the key operation and maintenance alarm data to the corresponding grid node using a string similarity matching algorithm; the dynamic time window is adaptively adjusted based on the current alarm frequency and the baseline alarm frequency; and a topology template dynamic generation module configured to perform feature clustering on multiple historical topology configuration data of the gridded hierarchical topology graph, generating and storing multiple parameterized topology templates associated with business types; and to dynamically generate the target topology graph in response to the business type of the target business by calling the parameterized topology template of the corresponding business type.

[0006] According to another aspect of this disclosure, a method for integrating topology orchestration and operation and maintenance data is provided, applied to the aforementioned integrated topology orchestration and operation and maintenance data system. The method includes: a topology orchestration management module constructing a gridded hierarchical topology graph with grid nodes and topology edges based on a resource layer-column constraint framework, where layers and columns of the layer-column constraint framework form multiple grid units at their intersections; orchestrating nodes in each grid unit according to dynamically configured Boolean query conditions and multidimensional attributes of resources to generate grid nodes; converging multiple topology paths with the same source and target nodes between multiple grid nodes according to a preset strategy to obtain topology edges; a unified operation and maintenance data access module acquiring key operation and maintenance alarm data for each grid node within a dynamic time window, and using a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node; the dynamic time window adaptively adjusting based on the current alarm frequency and the baseline alarm frequency; a topology template dynamic generation module performing feature clustering on multiple historical topology configuration data of the gridded hierarchical topology graph to generate and store multiple parameterized topology templates associated with business types; and dynamically generating the target topology graph in response to the business type of the target business by calling the corresponding parameterized topology template.

[0007] As will be described in detail below, the integrated topology orchestration and operation and maintenance data system according to embodiments of this disclosure uses a topology orchestration management module that divides grid units based on a hierarchical constraint framework. This solves the problems of chaotic hierarchical topologies and disordered resource classification in traditional systems, enabling hierarchical and domain-based standardized management of resources and improving the readability of the gridded hierarchical topology graph. Node orchestration is completed based on multi-dimensional resource attributes and dynamic Boolean query conditions, eliminating the traditional manual drag-and-drop topology creation method, reducing the workload of node orchestration, and ensuring the matching of grid nodes and resources. Convergence optimization is performed on topology paths with the same source and target, eliminating a large number of redundant parallel paths, reducing the overall complexity of the gridded hierarchical topology graph, and preventing redundant paths from interfering with subsequent fault tracing and bottleneck analysis, making the topology link relationships more closely aligned with the actual business call logic. The unified operation and maintenance data access module mounts key operation and maintenance alarm data within a dynamic time window to the corresponding grid nodes, breaking down the data silos between traditional topology and operation and maintenance data. This eliminates the need for operation and maintenance personnel to manually associate alarms with device nodes, significantly reducing alarm binding time. The dynamic topology template generation module performs feature clustering on multiple historical topology configuration data to uncover common architectural rules for similar business topologies. This avoids subjective biases caused by manually summarizing topology experience, improves the accuracy of matching parameterized topology templates with business types, and allows for the direct calling of parameterized topology templates for the corresponding business type to dynamically generate the target topology map when adding new businesses of the same type. This solves the problem of repeatedly building a topology from scratch for each new business launch, enabling dynamic expansion of topology templates, shortening the new business topology building cycle, and improving topology delivery efficiency.

[0008] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0009] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0010] Figure 1 This diagram illustrates the framework structure of the integrated topology orchestration and operation and maintenance data system provided in an exemplary embodiment of this disclosure. Figure 2 A schematic diagram of the alarm data mounting mechanism provided in an exemplary embodiment of this disclosure is shown. Figure 3 A flowchart illustrating the integrated topology orchestration and operation and maintenance data method provided by an exemplary embodiment of this disclosure is shown. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this disclosure more apparent, exemplary embodiments according to this disclosure will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments of this disclosure. It should be understood that this disclosure is not limited to the exemplary embodiments described herein.

[0012] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0013] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0014] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0015] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0016] Among related technologies, a topology and orchestration system for cloud applications is proposed, which establishes the relationship between each target node through a topology building module and outputs an orchestration package for the cloud application.

[0017] A method for obtaining and drawing a topology graph is also proposed. This method obtains the first-level group nodes and the connections between them, where the first-level group nodes are nodes whose parent group nodes are empty. The topology graph is then drawn based on the first-level group nodes and their connections. User actions are those that trigger the expansion of the first-level group nodes and their child nodes at all levels. The data corresponding to the user actions includes the child nodes of the first-level group nodes and the connections between them. The data corresponding to the user actions is then added to the topology graph to obtain a new topology graph.

[0018] It also provides a method for processing network data, using custom groups to achieve hierarchical and domain-based topology arrangement. During the drawing process, individual network elements can be selected into layers or domains. Furthermore, it proposes a method to save the already arranged topology model as a template, which can be directly copied to the operation panel via drag and drop, making full use of existing resources and simplifying the arrangement process.

[0019] The main drawbacks of the above solution are as follows: First, it lacks an efficient node filtering method: the existing data access interface is not flexible enough and does not support complex query conditions or dynamic data filtering, limiting the ability of operations and maintenance personnel to obtain key data; Second, it lacks a cross-layer topology diagram construction method: the existing visualization tools do not support the display of cross-layer data, making it difficult to intuitively understand the system architecture; Third, it lacks an operations and maintenance data integration mechanism: the processing of operations and maintenance data is relatively fragmented, lacking an intelligent filtering mechanism for massive alarms (such as dynamic time window adjustment), causing key operations and maintenance information to be overwhelmed by noise; Fourth, the topology template has insufficient scalability: the existing topology template does not support parameter-based customization and dynamic adjustment, making it difficult to adapt to the dynamic changes in the system architecture and the development of future needs.

[0020] To address the shortcomings of the existing technologies, this disclosure proposes an integrated system for topology orchestration and operation and maintenance data. This system aims to achieve dynamic node selection under multiple conditions through nested Boolean operations, topology edge data orchestration through cross-layer path configuration, unified access to operation and maintenance data through dynamic adaptive algorithms, and dynamic generation and management of topology templates through clustering algorithms. This effectively solves the aforementioned problems in the existing technologies.

[0021] Figure 1 A schematic diagram of the framework structure of the integrated topology orchestration and operation and maintenance data system provided in an exemplary embodiment of this disclosure is shown. Figure 1 As shown, the integrated topology orchestration and operation and maintenance data system 100 includes: The topology orchestration management module 101 is configured to construct a gridded hierarchical topology graph with grid nodes and topological edges based on a resource-based hierarchical constraint framework. The layers and columns of the hierarchical constraint framework form multiple grid cells at their intersections. The module orchestrates the nodes of each grid cell according to dynamically configured Boolean query conditions and the multidimensional attributes of the resources to generate grid nodes. Multiple topological paths with the same source node and target node between multiple grid nodes are converged according to a preset strategy to obtain topological edges. The unified access module 102 for operation and maintenance data is configured to acquire key operation and maintenance alarm data of each grid node within a dynamic time window, and to attach the key operation and maintenance alarm data to the corresponding grid node using a string similarity matching algorithm; the dynamic time window is adaptively adjusted based on the current alarm frequency and the baseline alarm frequency. The topology template dynamic generation module 103 is configured to perform feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map, generate and store multiple parameterized topology templates associated with business types; in response to the business type of the target business, it calls the parameterized topology template of the corresponding business type to dynamically generate the target topology map.

[0022] Specifically, the topology orchestration management module 101 can pre-build a hierarchical constraint framework for resources. Here, resources can be understood as all observable and orchestratable operational entities in a distributed system. In the aforementioned hierarchical constraint framework, layers can be used to represent the vertical logical layering of resources, columns can be used to represent the horizontal logical partitioning within a layer, and layers and columns intersect to form multiple grid cells.

[0023] This disclosure allows for the pre-defined layer-column constraints of the layer-column constraint framework, including layering rules and cross-layer mapping rules. The layering rules can strictly define the vertical layered structure of the topology (e.g., front-end layer → middleware layer → database layer), and the number of columns in each layer can be determined according to actual business needs (e.g., column 1 for login services, column 2 for payment services), and can be dynamically expanded based on subsequent business requirements. Cross-layer mapping supports logical binding between non-contiguous layers (e.g., the front-end layer can be directly associated with the database layer, skipping the middleware layer), providing flexible underlying support for cross-layer path configuration.

[0024] After the hierarchical constraint framework is built, users (or maintenance personnel) can dynamically configure Boolean query conditions through the front-end interface of the maintenance platform. The topology orchestration management module 101 orchestrates nodes of each grid unit according to the dynamically configured Boolean query conditions and the multi-dimensional attributes of resources, generating grid nodes for each grid unit. This enables accurate multi-condition retrieval of nodes in complex scenarios and avoids interference from irrelevant nodes.

[0025] The topology orchestration management module 101 can also establish topology paths based on the calling relationships and deployment dependencies between multiple grid nodes. Multiple topology paths may exist between the same source and target nodes (end-to-end, where both the source and target nodes are grid nodes). To eliminate redundant paths (such as service→host and service→cached→host), the topology orchestration management module 101 can converge and merge these topology paths according to a preset strategy, ultimately establishing only one topology path—a single-sided path—between the same source and target nodes, serving as the topology edge connecting them. Utilizing the short-circuiting properties of Boolean logic and a gridded index structure, query response time is significantly reduced, retrieval efficiency under large-scale topologies is improved, and query performance is optimized.

[0026] The topology orchestration management module 101 can also, after determining the grid nodes and topology edges, combine the layer-column constraint framework to store the quadruple relationship of layer-column-node-edge, and establish a cross-layer path index table to accelerate complex queries.

[0027] Based on this, the topology orchestration management module 101 can divide grid cells by relying on the hierarchical constraint framework, solving the problems of chaotic hierarchical topology and disordered resource classification in traditional irregular topology, realizing hierarchical and domain-based standardized management of resources, and improving the readability of the gridded hierarchical topology map; it completes node orchestration based on the multi-dimensional attributes of resources and dynamic Boolean query conditions, getting rid of the traditional manual drag-and-drop topology building mode, reducing the workload of node orchestration, and ensuring the matching of grid nodes and resources; it optimizes the convergence of topology paths with the same source and the same goal, removes a large number of redundant parallel paths, reduces the overall complexity of the gridded hierarchical topology map, avoids redundant paths interfering with subsequent fault tracing and link bottleneck analysis, and makes the topology link relationship more in line with the actual business calling logic.

[0028] To enhance the readability and usability of the gridded hierarchical topology map, critical operational alarm data is seamlessly integrated and mounted onto each grid node of the gridded hierarchical topology map. Given that this critical operational alarm data was originally scattered across different tables in the data governance platform, with varying fields in each table, a unified access solution was designed and implemented. This solution enables cross-table and cross-field data integration and standardized processing, ensuring data consistency and efficient access, and providing operations personnel with a more intuitive and comprehensive topology view and data analysis support.

[0029] The unified access module 102 for operation and maintenance data can acquire key operation and maintenance alarm data from each grid node within a dynamic time window, and use a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node. This solves the problems of scattered and noisy multi-source operation and maintenance data, and promotes intelligent operation and maintenance decision-making. Here, the dynamic time window can be adaptively adjusted based on the current alarm frequency and the baseline alarm frequency. The specific adaptive adjustment formula is not specifically limited in this embodiment. It should be understood that the current alarm frequency is positively correlated with the length of the dynamic time window. That is, the higher the current alarm frequency, the longer the dynamic time window. Expanding the time window can capture more upstream and downstream alarms related to the root cause, avoiding the truncation of the causal chain due to the time window being too short. Conversely, the lower the current alarm frequency, the shorter the dynamic time window. Shortening the time window can eliminate invalid time intervals and retain only truly relevant data.

[0030] For example, the adaptive adjustment formula for the dynamic time window, which automatically expands / shrinks the time window based on the alarm frequency, can be expressed as: T = T0 × [1 + 0.2 × tanh((N - N0) / 10)]. Where T0 represents the base window, equal to 15 minutes; N represents the current alarm frequency (times / minute); N0 represents the baseline alarm frequency, i.e., the average historical alarm frequency; and tanh represents the hyperbolic tangent function, ensuring the adjustment range is within ±20%.

[0031] The unified access module 102 for operation and maintenance data mounts key operation and maintenance alarm data within a dynamic time window to the corresponding grid nodes, breaking down the data silos between traditional topology and operation and maintenance data. This eliminates the need for operation and maintenance personnel to manually associate alarms with device nodes, significantly reducing alarm binding time. After the operation and maintenance alarm data is accurately mounted to the grid nodes, the gridded hierarchical topology map and real-time operation and maintenance alarm data can be displayed simultaneously in the same view. Operation and maintenance personnel can directly locate alarm nodes based on the gridded hierarchical topology map, shortening alarm location time.

[0032] To improve the efficiency of configuration management and the accuracy of topology generation for business scenarios with similar structures, this disclosure implements a strategy for unified management of parameterized topology templates for different business types. This strategy identifies most shareable configuration conditions based on common characteristics among businesses, while also defining the set of parameterized topology templates (also known as key parameter sets) that require personalized adjustments.

[0033] The topology template dynamic generation module 103 can obtain multiple historical topology configuration data of the gridded hierarchical topology map. This historical topology configuration data may include the topology's business type, hierarchical architecture, redundancy design, node type, number of nodes, node associations, etc., and can be stored in XML or JSON format. Table 1 shows a schematic diagram of the storage table of historical topology configuration data provided by the exemplary embodiment of this disclosure.

[0034] Table 1. Schematic diagram of the storage table for historical topology configuration data.

[0035] Then, feature engineering and clustering are performed on multiple historical topology configuration data to generate multiple parameterized topology templates associated with business types. When a new target business is received subsequently, the parameterized topology template corresponding to the business type of the target business can be invoked to dynamically generate the target topology map, thus realizing the dynamic expansion of the topology templates.

[0036] Based on this, the dynamic topology template generation module 103 performs feature clustering on multiple historical topology configuration data, mines common architectural rules of similar business topologies, avoids subjective biases caused by manual summarization of topology experience, improves the accuracy of the adaptation of parameterized topology templates to business types, and solves the repetitive work of building a topology from scratch for each new business launch. When adding new businesses of the same type, the parameterized topology template of the corresponding business type can be directly called to dynamically generate the target topology map, realizing the dynamic expansion of topology templates. The new business topology building cycle is significantly shortened, the topology delivery efficiency is greatly improved, and the standardization of the topology architecture of similar businesses is ensured, which facilitates subsequent unified operation and maintenance management and compliance auditing.

[0037] In the technical solution of the exemplary embodiments disclosed herein, the three major modules—topology orchestration management module, unified access module for operation and maintenance data, and dynamic generation module for topology templates—work together to achieve a fully automated closed loop for node selection, automatic cross-layer topology orchestration, automatic alarm mounting, and topology template reuse. This completely solves the problems of cumbersome manual topology construction, redundant and messy topology links, disconnect between operation and maintenance data and topology, and inability to expand new business topology templates in traditional operation and maintenance systems. It reduces the overall labor cost of operation and maintenance, shortens the end-to-end location time of business alarms, and improves the scalability of new business topology templates.

[0038] Based on the above embodiments, in another embodiment provided in this disclosure, the Boolean query conditions may include at least one of the AND operator, OR operator and NOT operator, and the multidimensional attributes may include, but are not limited to, at least one of the following: service name, availability zone, service type, disaster recovery level and operating status.

[0039] Here, embodiments of this disclosure can perform multi-dimensional attribute modeling of resources. For example, dynamically associating resource pools, availability zones, and service names (also known as service tags) to construct a three-dimensional topological resource space (e.g., {East China Resource Pool, Suzhou AZ, E-commerce Business}).

[0040] Based on dynamically configured Boolean query conditions and multidimensional resource attributes, the nodes of each grid cell are arranged to generate grid nodes, which may include: Configure the corresponding node type template for each grid cell; It receives Boolean query conditions dynamically configured by the user, associates the Boolean query conditions with the node type template, and uses the Boolean query conditions to combine multidimensional attributes to generate a structured query statement. Execute structured query statements within each grid cell to pre-filter nodes and determine the initial node set for each grid cell; The initial node sets of each grid cell are aggregated across grid cells and deduplicated to generate grid nodes.

[0041] Specifically, firstly, a node type template is defined for each grid (Layer×Column) unit (e.g., first row (L1) first column (C1) = service node, abbreviated as L1C1 = service node). The node type of each grid unit is defined in advance, and resource partitioning and control are implemented from the architecture level to avoid node mismatch and cross-grid misclassification problems during subsequent query filtering, thereby improving the accuracy of node query.

[0042] The topology orchestration management module 101 provides a visual configuration interface, allowing maintenance personnel to dynamically configure custom Boolean query conditions based on maintenance and troubleshooting needs and resource filtering requirements, without writing code. It supports three basic logical operations: AND, OR, and NOT. The module automatically binds the user-configured Boolean query conditions to the corresponding grid cell's node type template (i.e., mandatory validation rules), matches the template's built-in node attribute fields, calls the resource multi-dimensional attribute library, logically combines the multi-dimensional attributes, and generates a structured query statement.

[0043] Logical combinations can include one of the following: exact matching, fuzzy coverage, condition exclusion, and complex queries. Exact matching can be: Business Name = BC-EC AND Availability Zone = Suzhou AZ; fuzzy coverage can be: Service Type IN (nginx, haproxy, keepalived) OR Disaster Recovery Level ≥ 3; condition exclusion can be: (Node Type = Database) AND (Status NOT IN Abnormal); complex queries can be: nested Boolean groups within the same grid cell (e.g., (Condition A OR Condition B) AND NOT Condition C), improving orchestration flexibility.

[0044] For example, the topology orchestration management module 101 receives Boolean query conditions, also known as compound query conditions, input by the user and generates a structured query statement that can be executed by the underlying database. For example, the query (region=East China AND type=database) OR (type=middleware AND load>80%) can be parsed into: (region = 'East China' AND node_type = 'database') OR (node_type = 'middleware' AND current_load>80).

[0045] The topology orchestration management module 101 performs preliminary node screening within each grid cell. It executes structured queries within each grid cell to pre-screen nodes, utilizing short-circuit logic of Boolean operators to optimize query performance and determine the initial node set for each grid cell, thereby avoiding a full scan and improving query efficiency. Here, the initial node set is also called the initial screening result for each grid cell, which can include multiple initial nodes. Then, the initial screening results for each grid cell are aggregated using Boolean logic, automatically performing node deduplication and priority determination (if a node simultaneously meets the node type conditions of multiple grid cells, only the highest priority instance is returned), obtaining the corresponding grid nodes for each grid cell.

[0046] For example, cross-mesh aggregation and deduplication can perform node deduplication and conflict detection, as well as cross-layer edge existence verification. Node deduplication and conflict detection can be performed as follows: when retrieving nodes based on Boolean conditions, automatic cross-mesh deduplication is performed (e.g., if a node satisfies both L1C1 and L2C3 conditions, only the highest priority mesh instance is returned). Cross-layer edge existence verification supports two query modes: exact match: source = service1, target = host1, maximum hop count = 1 → returns directly connected edges; path deduction: source = service3, target = host1, path template = service-*-database-host → automatically deduces intermediate nodes and returns converged edges.

[0047] A gridded hierarchical topology graph is constructed based on the grid nodes corresponding to multiple grid cells, forming a final node set, and the topological edges associated with these grid nodes. The final node set and related topological edge data are returned to the front end, which can then perform topology visualization rendering in conjunction with the grid layout.

[0048] By following the above configuration and retrieval steps, the following optimization effects can be achieved: Improved query accuracy: By combining multi-dimensional attributes with Boolean operators such as AND / OR / NOT, accurate node location can be achieved in complex scenarios, avoiding interference from irrelevant nodes; Enhanced retrieval flexibility: Nested Boolean groups are supported to adapt to diverse operation and maintenance query needs, such as fault diagnosis, resource auditing, and capacity planning; Optimized query performance: By utilizing the short-circuiting characteristics of Boolean logic and a gridded index structure, query response time is significantly reduced, and retrieval efficiency is improved under large-scale topologies.

[0049] Based on the above embodiments, in another embodiment provided in this disclosure, the preset strategy may include one of the shortest path, the highest weight path, and the most stable link. The topology orchestration management module is also configured to attach path metadata corresponding to multiple topology paths to topology edges; wherein, the path metadata includes at least the hop count and node type of intermediate nodes, and intermediate nodes are grid nodes located between the source node and the target node in each topology path; the weight of each topology path is calculated based on the path metadata.

[0050] Specifically, the preset strategy can be the shortest path, where the topology edge is the topology path with the fewest grid nodes between the source and target nodes among multiple topology paths with the same source and target nodes. Alternatively, the preset strategy can be the highest weight path, where the topology edge is the topology path with the highest weight among multiple topology paths with the same source and target nodes. Finally, the preset strategy can be the most stable link, where the topology edge is the most stable topology path among multiple topology paths with the same source and target nodes.

[0051] Each topological edge's source and target nodes can belong to different layers within the hierarchical constraint framework, allowing for cross-layer associations between source and target nodes. Breaking away from the traditional 1-hop limitation, it supports explicit definition of N-hop paths (N≤3) and automatically resolves intermediate nodes. For example, if the converged topological edge is switch→router, and the original topological path was switch→physical port→router, then this topological edge implements a 2-hop path (one "→" represents one hop); if the original topological path was switch→physical port 1→physical port 2→router, then this topological edge implements a 3-hop path.

[0052] The topology orchestration management module 101 can also attach path metadata corresponding to multiple topology paths to topology edges, so as to retain the original topology semantics while converging redundant paths into single edges, greatly simplifying the operation and maintenance complexity of complex systems. For each topology path on a topology edge, all grid nodes between the source node and the target node within the path are automatically stripped; the grid nodes between the source node and the target node are called intermediate nodes. Then, the two core path metadata of each topology path are counted: one is the hop count of the intermediate nodes, that is, the total number of intermediate grid nodes traversed by each topology path; the other is the node type corresponding to each intermediate node.

[0053] Its effect is to abandon the traditional topology edge's single information mode of only recording source and target nodes and link connectivity status, and supplement it with the topology information of intermediate nodes throughout the link transmission process, completely restoring the real link of each topology path. It solves the blind spot problem of traditional topology links only knowing the beginning and end, but not the intermediate forwarding details, so that topology edges are no longer just simple connections, but traceable and disassembled complete link data carriers, providing objective raw data support for subsequent path quality determination.

[0054] When the preset strategy is the highest weight path, the weights of multiple topological paths corresponding to each topological edge can be calculated based on path metadata. For example, the weight can be obtained by weighting the hop count and node type scores, and this disclosure does not specifically limit this.

[0055] Based on the above embodiments, in another embodiment provided in this disclosure, using a string similarity matching algorithm to attach key operation and maintenance alarm data to the corresponding grid node may include: Data cleaning is performed on the Internet Protocol address field in the critical alarm data of operation and maintenance to obtain the first string; data cleaning is performed on the Internet Protocol address field in the resource configuration information of multiple grid nodes to obtain the second string of multiple grid nodes; Calculate the similarity between the first string and the second string of multiple grid nodes; If the similarity is greater than or equal to the preset similarity threshold, the key operation and maintenance alarm data will be mounted to the corresponding grid node.

[0056] Specifically, Figure 2 A schematic diagram of the alarm data mounting mechanism provided in an exemplary embodiment of this disclosure is shown. Figure 2 As shown, the alarm management system continuously pushes new alarms. At this time, it is necessary to mount the relevant data of these alarms to the corresponding grid nodes so that the operation and maintenance personnel can investigate the alarms.

[0057] In production environments handling tens of millions of alarms daily, traditional manual configuration methods face the following problems: First, data overload: only 30% of alarms are directly related to topology nodes; second, attribute heterogeneity: IP fields from different monitoring systems have different formats; and third, critical alarm dilution: fixed time windows may cause critical alarms to be overwhelmed, leading to a distraction of operations and maintenance personnel. Therefore, the following methods are used to establish an alarm-topology node mounting mechanism.

[0058] The alarm mounting conditions employ an intelligent recommendation algorithm, specifically a string similarity matching algorithm. In the alarm management system, different monitoring tools may report Internet Protocol (IP) addresses in different formats (e.g., 192.168.1.1, 192-168-1-1, or 192168011), causing traditional exact matching to fail. A fault-tolerant matching algorithm is needed to calculate the similarity between two IP addresses and establish an association when the similarity exceeds a preset similarity threshold.

[0059] Data cleaning is performed on the Internet Protocol address field in the critical operation and maintenance alarm data and the Internet Protocol address field in the resource configuration information of multiple grid nodes to obtain the first string corresponding to the critical operation and maintenance alarm data and the second string corresponding to the resource configuration information of multiple grid nodes. Data cleaning can remove all non-numeric characters and retain only the numeric part.

[0060] Data cleaning can be expressed by the following formula: , .in, Indicates critical alarm data for operation and maintenance. The corresponding first string; Represents the resource configuration information of grid nodes The corresponding second string; This indicates data cleaning. Example: ; .

[0061] When calculating the similarity between the first string and the second string of multiple grid nodes, the longest common subsequence can be calculated first. The longest sequence of characters in the same order between two strings (not necessarily consecutive); then calculate the similarity. Divide the longest common subsequence by the length of the shorter IP address field. Ensure the result is within the range of 0 to 1.

[0062] The calculation formula for this step is: , Example: , .

[0063] If the similarity is greater than or equal to a preset similarity threshold, the critical operation and maintenance alarm data is mounted to the corresponding grid node. Here, the preset similarity threshold can be determined according to actual needs, and this embodiment does not specifically limit it. In the method of this embodiment, the preset similarity threshold can be set to 0.8. In this case, the critical operation and maintenance alarm data is successfully associated with the grid node, and the critical operation and maintenance alarm data is mounted to the corresponding grid node.

[0064] Based on this, in terms of data governance, the embodiments of this disclosure adopt an intelligent string similarity matching algorithm to solve the problem of automatic association of heterogeneous data, and adaptively adjust the alarm mounting time interval based on the dynamic time window formula to ensure accurate capture of key alarms.

[0065] Based on the above embodiments, in another embodiment provided in this disclosure, feature clustering is performed on multiple historical topology configuration data of the gridded hierarchical topology map to generate and store multiple parameterized topology templates associated with business types, including: From multiple historical topology configuration data of the gridded hierarchical topology map, extract the topology configuration feature vector of each historical topology configuration data; the topology configuration feature vector includes business type, hierarchical architecture, node type distribution and connection relationship distribution; Cluster the topology configuration feature vectors of multiple historical topology configuration data to obtain multiple cluster centers associated with business types; Multiple cluster centers are stored as multiple parameterized topology templates.

[0066] Specifically, the topology template dynamic generation module 103 can perform feature engineering on each historical topology configuration data from multiple historical topology configuration data in the gridded hierarchical topology map, extracting standardized and quantifiable topology configuration feature vectors. In the topology configuration feature vector, the first dimension can be the business type, statistically analyzing the business types corresponding to the historical topology configuration data; the second dimension can be the hierarchical architecture, statistically analyzing the number of vertical layers in the topology, the rules for dividing each grid cell, and the fixed constraints between layers, quantifying the output topology hierarchical structure parameters. For example, the hierarchical architecture can be represented as [1,2][2,2],[3,3]; the third dimension can be the node type distribution, statistically analyzing the node types of each unit in the hierarchical architecture, such as service nodes, database nodes, load balancing nodes, etc.; and the fourth dimension can be the connection relationship distribution, sorting out the connection order and connection method between nodes. Finally, the four-dimensional feature vectors are concatenated to generate a unique set of topology configuration feature vectors corresponding to the historical topology configuration data, eliminating data interference caused by differences in the original historical topology configuration data format. For example, the topology configuration feature vector can be represented as: Topology configuration feature vector = [Business type, Hierarchical architecture, Node type distribution, Connection relationship distribution].

[0067] By standardizing four-dimensional feature vectors, unstructured historical topology configuration data is transformed into structured data. This eliminates the need for subjective, manual summarization of topology architecture experience, and precisely characterizes the architectural attributes of each topology from four core essential dimensions: business type, level, nodes, and links. A unified feature dimension standard is established for all historical topology configuration data, resolving the issues of inconsistent topology configuration formats across different periods and the inability to directly compare and analyze them. This provides well-organized and unbiased raw input data for subsequent accurate clustering.

[0068] The dynamic topology template generation module 103 can employ the K-means unsupervised clustering algorithm to perform global clustering calculations on multiple topology configuration feature vectors. It automatically matches the corresponding business scenario with the business type, dividing multiple historical topology configuration data into multiple cluster centers associated with the business type. Each cluster center represents the optimal topology architecture among all topologies under the corresponding business type. Multiple cluster centers are stored as multiple parameterized topology templates, forming a fixed mapping relationship between business and topology architecture.

[0069] This disclosure embodiment can calculate the comprehensive distance between each topology configuration feature vector in multiple historical topology configuration data and each cluster center. The comprehensive distance between the topology configuration feature vector and the cluster center can be a weighted sum of the service type distance, hierarchical architecture distance, node type distribution distance, and connection relationship distribution distance.

[0070] The business type distance can be obtained by calculating the difference between business type codes:

[0071] in, This represents the service type distance between the topology configuration feature vector x and the cluster center y; This represents the business type encoding of the topology configuration feature vector x; This represents the business type code of cluster center y.

[0072] Hierarchical distance can be obtained by calculating the Euclidean distance between hierarchical arrays:

[0073] in, This represents the hierarchical architectural distance between the topological configuration feature vector x and the cluster center y; and A combined array of hierarchical architectures used to represent the topology configuration feature vector x; and A combined array of hierarchical structures used to represent cluster centers y.

[0074] The distance between node type distributions can be obtained by calculating the similarity between node type distribution dictionaries using cosine similarity:

[0075] in, This represents the distance between the topology configuration feature vector x and the cluster center y, representing the distribution distance of node types. This represents the node type of grid node i in the topology configuration feature vector x; This indicates the node type of grid node i in cluster center y.

[0076] The distance between connection distribution dictionaries can be obtained by calculating the similarity between connection distribution dictionaries using cosine similarity:

[0077] in, This represents the distribution distance of the connection relationship between the topological configuration feature vector x and the cluster center y; This represents the connectivity of topological edges j in the topological configuration feature vector x; This represents the connection relationship of the topological edge j in the cluster center y.

[0078] We obtain the combined distance by weighted summation:

[0079] in, This represents the combined distance between the topological configuration feature vector x and the cluster center y; α, β, γ, and δ are weighting coefficients that can be adjusted according to actual needs. The initial values ​​are α=0.5, β=0.2, γ=0.2, and δ=0.1.

[0080] Each historical topology configuration is assigned to the cluster corresponding to the nearest cluster center. Thus, each cluster contains a set of topology configurations that are relatively close in the feature space. Each cluster center corresponds to a parameterized topology template. Multiple cluster center correspondences are stored as multiple parameterized topology templates associated with different business types, forming a fixed mapping relationship between business and topology architecture. This ensures precise matching between business and topology templates, improving the efficiency of generating target topology maps for new businesses, reducing the cost of topology debugging and modification, and achieving template reusability and scalability.

[0081] Based on the above embodiments, in another embodiment provided in this disclosure, the topology template dynamic generation module is further configured to receive newly added topology configuration data and, in response to the newly added topology configuration data, update the parameterized topology template in a plurality of parameterized topology templates that corresponds to the service type of the newly added topology configuration data.

[0082] Specifically, the topology template dynamic generation module 103 can continuously monitor the topology configuration port in the background, receiving new topology configuration data generated by online business iterations and architecture expansions in real time. First, it extracts the business type corresponding to the new topology configuration data, accurately matching it to the business type of the parameterized topology template already in the database. Then, it synchronously extracts the business type, hierarchical architecture, node type distribution, and connection relationship distribution of the new topology configuration data, incorporating the new topology configuration feature vector into the topology configuration feature vector set of the corresponding business type cluster, completing the incremental supplementation of the sample set. Based on the updated topology configuration feature vector, the module recalculates the cluster center of the cluster, completing the adaptive iterative upgrade of the parameterized topology template. Topology templates for other unrelated business types retain their original configurations and are unaffected by this incremental update.

[0083] When adding a new topology for each business type, the cluster center needs to be updated, which means recalculating the average value of all topology configurations in the corresponding cluster for that business type.

[0084] The update formula for the cluster center can be: in, This indicates the cluster center of cluster k corresponding to the business type; Indicates the cluster k corresponding to the business type; x represents the topology configuration feature vector of the topology configuration data; This represents the total number of topology configuration feature vectors in cluster k corresponding to the business type.

[0085] The average service type can be the average of the service type codes of all topology configurations in the computation cluster.

[0086]

[0087] in, This represents the average value of the business types at the cluster center of cluster k corresponding to the business type.

[0088] The hierarchical mean can be calculated as the mean of the hierarchical array of all topology configurations in the cluster.

[0089]

[0090] in, This represents the average hierarchical architecture of the cluster center of cluster k corresponding to the business type.

[0091] The mean of the node type distribution can be calculated as the mean of the node type distribution dictionary for all topology configurations in the cluster using a weighted average.

[0092]

[0093] in, This represents the mean distribution of node types at the cluster center of cluster k corresponding to the business type.

[0094] The mean of the connectivity distribution can be:

[0095] in, This represents the mean distribution of connectivity relationships among cluster centers of cluster k corresponding to the business type.

[0096] After the cluster center is updated, the entire parameterized topology template generation system forms a complete and long-term operation and maintenance closed loop of initial clustering template building and online incremental iterative updates. It can quickly generate standardized initial templates based on historical data and continuously optimize template parameters as business evolves. The entire template update and iteration process is automated, eliminating the need for operation and maintenance personnel to manually modify the parameterized topology templates, further reducing the later maintenance costs of topology templates, and adapting to the normalized operation and maintenance scenarios of long-term business iteration and continuous resource expansion.

[0097] Based on the above embodiments, in another embodiment provided in this disclosure, the topology template dynamic generation module is further configured to receive the manual verification results of multiple parameterized topology templates by experts, and adjust the multiple parameterized topology templates based on the manual verification results.

[0098] Specifically, the topology template dynamic generation module 103 provides an expert manual verification portal, allowing operations and maintenance architecture experts to periodically conduct manual verification of all stored parameterized topology templates (i.e., multiple cluster centers and initial template classification results) within the platform. Based on the manual verification results, the cluster division is further optimized, supporting manual adjustment of the number of clusters K or the initial selection of cluster centers to improve the accuracy and usability of clustering results. Simultaneously, a tagging system is established, employing a combination of semantic tags and ontology tags to construct a multi-level, multi-dimensional tagging system, ensuring the accuracy and retrievalability of template classification.

[0099] While clustering algorithms ensure the efficiency of parameterized topology template generation and iteration, pure algorithms can only uncover objective patterns in historical data and cannot adapt to constraints such as industry standards, special deployment requirements, and security compliance. Human verification can fill the gaps in algorithms, eliminate unreasonable templates caused by data bias, and enable parameterized topology templates to have both high-efficiency automation capabilities and the backup capability of human experience, fully adapting to the complex topology orchestration and maintenance needs of government, enterprise, and financial multi-scenario applications.

[0100] Based on this, the integrated topology orchestration and operation and maintenance data system provided in the exemplary embodiments of this disclosure constructs a three-dimensional topology resource space through gridded hierarchical modeling and defines strict layer-column constraints. The resulting layer-column constraint framework supports logical binding between discontinuous layers. It also introduces nested Boolean operators to achieve dynamic node filtering and complex query condition combinations, significantly improving retrieval flexibility. Furthermore, it innovatively designs a cross-layer path configuration and edge convergence mechanism, breaking through the traditional one-hop limitation and optimizing topology display through multi-path aggregation. In terms of data governance, it employs an intelligent similarity matching algorithm to solve the problem of automatic association of heterogeneous data and adaptively adjusts the alarm mounting interval based on a dynamic time window formula to ensure accurate capture of critical alarms. In addition, it dynamically generates parameterized templates from historical topology configuration data through K-means clustering analysis, enabling one-click drawing of similar business scenarios, improving the reusability and customizability of topology templates, avoiding redundant construction, and adapting to dynamic changes in system architecture to meet the flexible configuration needs of different business scenarios. The entire solution effectively solves problems such as data redundancy, rigid queries, and fixed templates in traditional operations and maintenance through a closed-loop design of unified data access, intelligent topology orchestration, and dynamic template generation. While improving operational efficiency, it enhances the scalability and intelligence of the system, providing an innovative solution for operations and maintenance management in complex cloud environments.

[0101] The exemplary embodiments disclosed herein also provide a method for integrating topology orchestration and operation and maintenance data, applied to the aforementioned integrated topology orchestration and operation and maintenance data system. Figure 3 A flowchart illustrating the integrated topology orchestration and operation and maintenance data method provided in an exemplary embodiment of this disclosure is shown. Figure 3 As shown, this integrated topology orchestration and operation and maintenance data method includes: S301, the topology orchestration management module constructs a gridded hierarchical topology graph with grid nodes and topology edges based on the layer and column constraint framework of resources. The layers and columns of the layer and column constraint framework form multiple grid cells at their intersections. The module orchestrates the nodes of each grid cell according to dynamically configured Boolean query conditions and the multidimensional attributes of resources to generate grid nodes. Multiple topology paths with the same source node and target node between multiple grid nodes are converged according to a preset strategy to obtain topology edges. S302, the unified access module for operation and maintenance data obtains the key operation and maintenance alarm data of each grid node within the dynamic time window, and uses a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node; the dynamic time window is adaptively adjusted based on the current alarm frequency and the baseline alarm frequency; S303, the topology template dynamic generation module performs feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map, generates and stores multiple parameterized topology templates associated with business types; in response to the business type of the target business, it calls the parameterized topology template of the corresponding business type to dynamically generate the target topology map.

[0102] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0103] Based on the above embodiments, in another embodiment provided in this disclosure, the Boolean query conditions include at least one of the AND operator, OR operator and NOT operator, and the multidimensional attributes include, but are not limited to, at least one of the following: business name, availability zone, service type, disaster recovery level and operating status. Based on dynamically configured Boolean query conditions and multidimensional resource attributes, the nodes of each grid cell are arranged to generate grid nodes, including: Configure the corresponding node type template for each grid cell; It receives Boolean query conditions dynamically configured by the user, associates the Boolean query conditions with the node type template, and uses the Boolean query conditions to combine multidimensional attributes to generate a structured query statement. Execute structured query statements within each grid cell to pre-filter nodes and determine the initial node set for each grid cell; The initial node sets of each grid cell are aggregated across grid cells and deduplicated to generate grid nodes.

[0104] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0105] Based on the above embodiments, in another embodiment provided in this disclosure, the preset strategy includes one of the shortest path, the highest weight path, and the most stable link; the method further includes: The topology orchestration management module attaches path metadata corresponding to multiple topology paths to the topology edges; the path metadata includes at least the hop count and node type of intermediate nodes, and intermediate nodes are the grid nodes located between the source node and the target node in each topology path; the weight of each topology path is calculated based on the path metadata.

[0106] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0107] Based on the above embodiments, in another embodiment provided in this disclosure, a string similarity matching algorithm is used to attach key operation and maintenance alarm data to the corresponding grid node, including: Data cleaning is performed on the Internet Protocol address field in the critical alarm data of operation and maintenance to obtain the first string; data cleaning is performed on the Internet Protocol address field in the resource configuration information of multiple grid nodes to obtain the second string of multiple grid nodes; Calculate the similarity between the first string and the second string of multiple grid nodes; If the similarity is greater than or equal to the preset similarity threshold, the key operation and maintenance alarm data will be mounted to the corresponding grid node.

[0108] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0109] Based on the above embodiments, in another embodiment provided in this disclosure, feature clustering is performed on multiple historical topology configuration data of the gridded hierarchical topology map to generate and store multiple parameterized topology templates associated with business types, including: From multiple historical topology configuration data of the gridded hierarchical topology map, extract the topology configuration feature vector of each historical topology configuration data; the topology configuration feature vector includes business type, hierarchical architecture, node type distribution and connection relationship distribution; Cluster the topology configuration feature vectors of multiple historical topology configuration data to obtain multiple cluster centers associated with business types; Multiple cluster centers are stored as multiple parameterized topology templates.

[0110] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0111] Based on the above embodiments, in another embodiment provided in this disclosure, the method further includes: The topology template dynamic generation module receives new topology configuration data and, in response to the new topology configuration data, updates the parameterized topology templates in multiple parameterized topology templates that correspond to the service type of the new topology configuration data.

[0112] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0113] Based on the above embodiments, in another embodiment provided in this disclosure, the method further includes: The topology template dynamic generation module receives the manual verification results of multiple parameterized topology templates from experts, and adjusts the multiple parameterized topology templates based on the manual verification results.

[0114] For more information, please refer to the relevant content in the integrated system of topology orchestration and operation and maintenance data; it will not be repeated here.

[0115] The above description is merely an illustration of some embodiments of this disclosure and the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0116] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A topology orchestration and operation and maintenance data integrated system, characterized in that, include: The topology orchestration management module is configured to construct a gridded hierarchical topology graph with grid nodes and topological edges based on a resource-based layer-column constraint framework. Multiple grid cells are formed at the intersections of the layers and columns of the layer-column constraint framework. The module orchestrates the nodes of each grid cell according to dynamically configured Boolean query conditions and the multidimensional attributes of the resources, generating the grid nodes. Multiple topological paths with the same source and target nodes between the grid nodes are converged according to a preset strategy to obtain the topological edges. The unified access module for operation and maintenance data is configured to obtain the key operation and maintenance alarm data of each grid node within a dynamic time window, and to attach the key operation and maintenance alarm data to the corresponding grid node using a string similarity matching algorithm. The dynamic time window is adaptively adjusted based on the current alarm frequency and the baseline alarm frequency; The topology template dynamic generation module is configured to perform feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map, generate and store multiple parameterized topology templates associated with business types; in response to the business type of the target business, the module calls the parameterized topology template of the corresponding business type to dynamically generate the target topology map.

2. The system as described in claim 1, characterized in that, The Boolean query conditions include at least one of the AND, OR, and NOT operators, and the multidimensional attributes include, but are not limited to, at least one of the following: business name, availability zone, service type, disaster recovery level, and operating status. The step of arranging nodes for each grid cell based on dynamically configured Boolean query conditions and the multidimensional attributes of the resource to generate the grid nodes includes: Configure a corresponding node type template for each of the aforementioned mesh elements; Receive user-configured Boolean query conditions, associate the Boolean query conditions with the node type template, and use the Boolean query conditions to combine the multidimensional attributes to generate a structured query statement; Within each of the grid cells, the structured query statement is executed to pre-filter nodes and determine the initial node set for each grid cell; The initial node sets of each of the aforementioned grid cells are aggregated across grid cells and deduplicated to generate the grid nodes.

3. The system as described in claim 1, characterized in that, The preset strategy includes one of the following: shortest path, highest weight path, and most stable link; The topology orchestration management module is also configured to attach path metadata corresponding to the multiple topology paths to the topology edges; wherein, the path metadata includes at least the hop count and node type of intermediate nodes, and the intermediate nodes are grid nodes located between the source node and the target node in each topology path; the weight of each topology path is calculated based on the path metadata.

4. The system as described in claim 1, characterized in that, The step of using a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node includes: Data cleaning is performed on the Internet Protocol address field in the critical alarm data of the operation and maintenance to obtain a first string; data cleaning is performed on the Internet Protocol address field in the resource configuration information of multiple grid nodes to obtain a second string of multiple grid nodes; Calculate the similarity between the first string and the second strings of the plurality of grid nodes; If the similarity is greater than or equal to a preset similarity threshold, the key operation and maintenance alarm data will be mounted to the corresponding grid node.

5. The system as described in claim 1, characterized in that, The step of performing feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map to generate and store multiple parameterized topology templates associated with business types includes: From multiple historical topology configuration data of the gridded hierarchical topology map, extract the topology configuration feature vector of each historical topology configuration data; wherein, the topology configuration feature vector includes business type, hierarchical architecture, node type distribution and connection relationship distribution; Cluster the topology configuration feature vectors of the multiple historical topology configuration data to obtain multiple cluster centers associated with the business type; The multiple cluster centers are stored as multiple parameterized topology templates.

6. The system as described in any one of claims 1 to 5, characterized in that, The topology template dynamic generation module is also configured to receive new topology configuration data and, in response to the new topology configuration data, update the parameterized topology template among the plurality of parameterized topology templates that corresponds to the service type of the new topology configuration data.

7. The system as described in any one of claims 1 to 5, characterized in that, The topology template dynamic generation module is also configured to receive manual verification results from experts on the multiple parameterized topology templates, and adjust the multiple parameterized topology templates based on the manual verification results.

8. A method for integrating topology orchestration and operation and maintenance data, applied to the integrated topology orchestration and operation and maintenance data system according to any one of claims 1 to 7, characterized in that, The method includes: The topology orchestration management module constructs a gridded hierarchical topology graph with grid nodes and topology edges based on the layer and column constraint framework of resources. The layers and columns of the layer and column constraint framework form multiple grid cells at their intersections. The module orchestrates the nodes of each grid cell according to dynamically configured Boolean query conditions and the multidimensional attributes of the resources to generate the grid nodes. Multiple topology paths with the same source node and target node between multiple grid nodes are converged according to a preset strategy to obtain the topology edges. The unified access module for operation and maintenance data acquires the key operation and maintenance alarm data of each grid node within a dynamic time window, and uses a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node; the dynamic time window is adaptively adjusted based on the current alarm frequency and the baseline alarm frequency; The topology template dynamic generation module performs feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map, generates and stores multiple parameterized topology templates associated with business types; in response to the business type of the target business, it calls the parameterized topology template of the corresponding business type to dynamically generate the target topology map.

9. The method as described in claim 8, characterized in that, The Boolean query conditions include at least one of the AND, OR, and NOT operators, and the multidimensional attributes include, but are not limited to, at least one of the following: business name, availability zone, service type, disaster recovery level, and operating status. The step of arranging nodes for each grid cell based on dynamically configured Boolean query conditions and the multidimensional attributes of the resource to generate the grid nodes includes: Configure a corresponding node type template for each of the aforementioned mesh elements; Receive user-configured Boolean query conditions, associate the Boolean query conditions with the node type template, and use the Boolean query conditions to combine the multidimensional attributes to generate a structured query statement; Within each of the grid cells, the structured query statement is executed to pre-filter nodes and determine the initial node set for each grid cell; The initial node sets of each of the aforementioned grid cells are aggregated across grid cells and deduplicated to generate the grid nodes.

10. The method as described in claim 8, characterized in that, The preset strategy includes one of the following: shortest path, highest weight path, and most stable link; the method further includes: The topology orchestration management module attaches path metadata corresponding to the multiple topology paths to the topology edges; wherein, the path metadata includes at least the hop count and node type of intermediate nodes, and the intermediate nodes are the grid nodes located between the source node and the target node in each topology path; the weight of each topology path is calculated based on the path metadata.

11. The method as described in claim 8, characterized in that, The step of using a string similarity matching algorithm to attach the key operation and maintenance alarm data to the corresponding grid node includes: Data cleaning is performed on the Internet Protocol address field in the critical alarm data of the operation and maintenance to obtain a first string; data cleaning is performed on the Internet Protocol address field in the resource configuration information of multiple grid nodes to obtain a second string of multiple grid nodes; Calculate the similarity between the first string and the second strings of the plurality of grid nodes; If the similarity is greater than or equal to a preset similarity threshold, the key operation and maintenance alarm data will be mounted to the corresponding grid node.

12. The method as described in claim 8, characterized in that, The step of performing feature clustering on multiple historical topology configuration data of the gridded hierarchical topology map to generate and store multiple parameterized topology templates associated with business types includes: From multiple historical topology configuration data of the gridded hierarchical topology map, extract the topology configuration feature vector of each historical topology configuration data; wherein, the topology configuration feature vector includes business type, hierarchical architecture, node type distribution and connection relationship distribution; Cluster the topology configuration feature vectors of the multiple historical topology configuration data to obtain multiple cluster centers associated with the business type; The multiple cluster centers are stored as multiple parameterized topology templates.

13. The method according to any one of claims 8 to 12, characterized in that, The method further includes: The topology template dynamic generation module receives new topology configuration data and, in response to the new topology configuration data, updates the parameterized topology template among the multiple parameterized topology templates that corresponds to the service type of the new topology configuration data.

14. The method according to any one of claims 8 to 12, characterized in that, The method further includes: The topology template dynamic generation module receives the manual verification results of the multiple parameterized topology templates from experts, and adjusts the multiple parameterized topology templates based on the manual verification results.

Images