Network cutover service influence prediction method and device, equipment and storage medium

Abstract

The invention provides a business influence prediction method and device for network cutover, equipment and a storage medium, and relates to the technical field of big data, and the method comprises the steps: predicting an influenced business set when a to-be-cutover network element is unavailable according to a pre-constructed business path knowledge base; wherein the service path knowledge base is used for storing a mapping relationship between service identifiers and service path information; the service path information is used for indicating a plurality of network elements through which a service flow passes; the affected service set comprises a plurality of affected services, and service flows of the affected services pass through the network element to be cut over; and in multiple dimensions, the influence of the unavailable network element to be cut over on the affected service set is quantitatively analyzed, and a multi-dimensional influence quantitative result is obtained. According to the scheme of the invention, the accurate prediction and quantitative analysis of the influence of the network element cut over can be realized before the network element cut over, so that the security and reliability of the network cut over can be improved.

Description

Technical Field

[0001] This invention relates to the field of big data technology, specifically to a method, apparatus, device, and storage medium for predicting the service impact of network cutovers. Background Technology

[0002] Business impact forecasting refers to the forward-looking assessment of the potential impact on business operations by analyzing internal processes, changes in the external environment, and potential risk factors. This provides decision-makers with data support and response strategies, and is widely used in risk management, strategic planning, resource allocation, and crisis response. Its accuracy and comprehensiveness directly affect a company's operational efficiency and long-term development capabilities.

[0003] Currently, when conducting network cutover impact assessments, there is a lack of specialized impact prediction and simulation analysis capabilities for network cutover operations, making it difficult to accurately quantify the impact of network cutover on existing services; and the adaptability to dynamic changes in service paths is insufficient, making it unable to effectively respond to real-time adjustments and changes in service paths. Summary of the Invention

[0004] At least one embodiment of the present invention provides a method, apparatus, device and storage medium for predicting the service impact of network cutover, which solves the problems in the prior art of lacking the ability to predict and simulate the service impact of network cutover, making it difficult to accurately quantify the impact of network cutover on existing services, and being insufficiently adaptable to dynamic changes in service paths, thus failing to effectively cope with real-time adjustments and changes in service paths.

[0005] To solve the above-mentioned technical problems, the present invention is implemented as follows:

[0006] In a first aspect, embodiments of the present invention provide a method for predicting the service impact of network cutover, comprising:

[0007] Based on a pre-built service path knowledge base, the set of affected services when the network element to be cut over becomes unavailable is predicted; wherein, the service path knowledge base is used to store the mapping relationship between service identifiers and service path information; the service path information is used to indicate multiple network elements through which the service flow passes; the set of affected services includes multiple affected services, and the service flow of the affected services passes through the network element to be cut over;

[0008] The impact of the unavailability of the network element to be cut over on the set of affected services is quantitatively analyzed in multiple dimensions to obtain multi-dimensional impact quantification results.

[0009] Optionally, the network cutover service impact prediction method further includes:

[0010] The service tagging information recorded by multiple network elements during the transmission of historical service service flows is obtained. The service tagging information includes at least one of the following: service identifier, network element identifier, time information, and network element node sequence number.

[0011] Based on multiple service tagging information, service path information corresponding to the historical service is generated, and the service path information is used to indicate the time sequence information of the service flow of the historical service through multiple network elements;

[0012] The mapping relationship between the business identifier and the business path information is stored in the business path knowledge base.

[0013] Optionally, the network cutover service impact prediction method includes generating service path information corresponding to the historical service based on multiple service tagging information, including:

[0014] Based on the time information in the service tag information, sort the multiple service tag information to obtain the sorted multiple service tag information;

[0015] Based on the network element node sequence number in the service tagging information, the continuity and monotonically increasing nature of the network node sequence number of the sorted multiple service tagging information are verified to obtain the verification result.

[0016] If the verification result is successful, then the business path information corresponding to the historical business is generated based on the sorted business tag information.

[0017] Optionally, the network cutover service impact prediction method, wherein predicting the set of affected services when the network element to be cut over becomes unavailable, based on a pre-built service path knowledge base, includes:

[0018] In the service path knowledge base, obtain the affected service path information, including the network element to be cut over, and multiple affected service identifiers corresponding to the affected service path information;

[0019] Based on multiple affected service identifiers, predict the set of affected services when the network element to be cut over becomes unavailable.

[0020] Optionally, the network cutover service impact prediction method further includes at least one of the following:

[0021] Obtain the business type corresponding to the affected business set;

[0022] Based on the affected service path information, analyze the location information of the network element to be cut over in the affected service path;

[0023] Predict the propagation risk level of the affected services in the aforementioned set of affected services;

[0024] If there is a backup path that bypasses the network element to be cut over for the affected service path, the service quality index of the backup path is obtained, and the switching feasibility of the backup path is analyzed to obtain a feasibility report.

[0025] Optionally, the network cutover service impact prediction method includes, in multiple dimensions, quantifying the impact of the unavailability of the network element to be cut over on the affected service set to obtain multi-dimensional impact quantification results, including:

[0026] Based on the business type, geographical location, and key business ratio corresponding to the affected business set, obtain the quantitative results of the impact on the business dimension;

[0027] Predict the changes in performance metrics corresponding to the affected service set to obtain quantitative results of service dimension impact;

[0028] Based on the scale of affected users and the scope of influence on member users corresponding to the set of affected services, the quantitative results of the user dimension impact are obtained;

[0029] The impact results of the business dimension, the service dimension, and the user dimension are normalized to obtain multi-dimensional impact quantification results.

[0030] Optionally, the network cutover service impact prediction method further includes at least one of the following:

[0031] Based on the multi-dimensional impact quantification results, risk level classifications are obtained, and a business impact heatmap is generated based on the risk level classifications and the geographical locations corresponding to the affected business set.

[0032] Based on the cutover plan time of each of the multiple cutover network elements, and the multi-dimensional impact quantification results, cutover time series information of multiple cutover network elements is generated.

[0033] Secondly, embodiments of the present invention also provide a network cutover service impact prediction device, comprising:

[0034] The prediction module is used to predict the set of affected services when the network element to be cut over becomes unavailable, based on a pre-built service path knowledge base. The service path knowledge base is used to store the mapping relationship between service identifiers and service path information. The service path information is used to indicate the multiple network elements through which the service flow passes. The set of affected services includes multiple affected services, and the service flow of the affected services passes through the network element to be cut over.

[0035] The analysis module is used to quantitatively analyze the impact of the unavailability of the network element to be cut over on the set of affected services in multiple dimensions, and obtain multi-dimensional impact quantification results.

[0036] Thirdly, embodiments of the present invention also provide a service impact prediction device for network cutover, comprising: a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the processor executes the program or instructions to implement the service impact prediction method for network cutover as described in the first aspect.

[0037] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the service impact prediction method for network cutover as described in the first aspect.

[0038] Fifthly, embodiments of the present invention also provide a computer program product, including computer instructions, which, when executed by a processor, implement the service impact prediction method for network cutover as described in the first aspect.

[0039] Compared with existing technologies, embodiments of the present invention provide a method, apparatus, device, and storage medium for predicting the service impact of network cutover. Based on a pre-built service path knowledge base, it predicts the set of services affected when a network element to be cut over becomes unavailable. The service path knowledge base stores the mapping relationship between service identifiers and service path information. The service path information indicates multiple network elements through which the service flow passes. The set of affected services includes multiple affected services, and the service flow of the affected services passes through the network element to be cut over. The impact of the unavailability of the network element to be cut over on the set of affected services is quantitatively analyzed in multiple dimensions to obtain multi-dimensional impact quantification results. This solves the problems of existing technologies lacking the ability to predict and simulate the service impact of network cutover, making it difficult to accurately quantify the impact of network cutover on existing services, and lacking adaptability to dynamic changes in service paths, thus failing to effectively cope with real-time adjustments and changes in service paths. It enables accurate prediction and quantitative analysis of the impact of network element cutover before the cutover, thereby improving the security and reliability of network cutover. Attached Figure Description

[0040] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0041] Figure 1This is a flowchart illustrating the service impact prediction method for network cutover as described in an embodiment of the present invention.

[0042] Figure 2 This is a schematic diagram of the service tagging information in an embodiment of the present invention;

[0043] Figure 3 This is a flowchart illustrating one embodiment of the business path knowledge base construction process in this invention.

[0044] Figure 4 This is a flowchart illustrating another implementation of the business path knowledge base construction process in this invention.

[0045] Figure 5 This is a flowchart illustrating one embodiment of the process for predicting the impact of cutover in this invention.

[0046] Figure 6 This is a flowchart illustrating another embodiment of the process for predicting the impact of cutover in this invention.

[0047] Figure 7 This is a flowchart illustrating one embodiment of the multi-dimensional influence analysis process in this invention.

[0048] Figure 8 This is a flowchart illustrating another implementation of the multi-dimensional influence analysis process in this invention.

[0049] Figure 9 This is a flowchart illustrating the cutover decision-making process in an embodiment of the present invention;

[0050] Figure 10 This is a schematic diagram of the architecture of the application system for the network cutover service impact prediction method described in this embodiment of the invention;

[0051] Figure 11 This is a schematic diagram of the service impact prediction device for network cutover as described in an embodiment of the present invention;

[0052] Figure 12 This is a hardware block diagram of the network cutover service impact prediction device described in an embodiment of the present invention. Detailed Implementation

[0053] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, the "or" in this invention indicates at least one of the connected objects. For example, "A or B" covers three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0054] See Figure 1 This invention provides a method for predicting the service impact of network cutover, which is essentially a method for predicting the service impact before network cutover.

[0055] Furthermore, the method includes:

[0056] Step 101: Based on a pre-built service path knowledge base, predict the set of affected services when the network element to be cut over becomes unavailable; wherein, the service path knowledge base is used to store the mapping relationship between service identifiers and service path information; the service path information is used to indicate multiple network elements through which the service flow passes; the set of affected services includes multiple affected services, and the service flow of the affected services passes through the network element to be cut over.

[0057] It is understood that the unavailability of the network element to be cut over means that the network element to be cut over has failed, such as due to restarting, upgrading, or replacement.

[0058] Step 102: Quantitatively analyze the impact of the unavailability of the network element to be cut over on the set of affected services in multiple dimensions to obtain multi-dimensional impact quantification results.

[0059] In one embodiment, optionally, before step 101, the method further includes:

[0060] The service tagging information recorded by multiple network elements during the transmission of historical service service flows is obtained. The service tagging information includes at least one of the following: service identifier, network element identifier, time information, and network element node sequence number.

[0061] Based on multiple service tagging information, service path information corresponding to the historical service is generated, and the service path information is used to indicate the time sequence information of the service flow of the historical service through multiple network elements;

[0062] The mapping relationship between the business identifier and the business path information is stored in the business path knowledge base.

[0063] In this embodiment of the invention, before step 101, the above method further includes a data processing stage. This data processing stage mainly focuses on the collection, cleaning, and structured storage of business flow data marking and historical business flow trace data, providing a high-quality data foundation for subsequent path reconstruction and impact prediction.

[0064] Figure 2 A schematic diagram of service tagging information in an embodiment of the present invention. For example... Figure 2 As shown, when a user uses a mobile communication network service, the network service will mark and record the service information involved in the service on each network element it passes through when the service flow passes through different network elements, and obtain service marking information. The service marking information includes service identifier, network element identifier, time information (such as timestamp), network element node sequence number (currently the Nth node of the path), and area identifier (the originating area and target area of ​​the service, such as Nanchang calling Beijing).

[0065] Specifically, the data originates from various network elements, such as AMF (Authentication Management Function), SMF (Session Management Function), and UPF (User Plane Function), and consists of service tagging information dynamically recorded during service transmission. This service tagging information is typically stored as logs in the local database of each network element or in a centralized log management system. Each piece of service tagging information includes the following fields:

[0066] Business ID: A unique identifier for a business flow, such as a hash value generated based on the business name (e.g., voice, SMS, internet access);

[0067] Network Element ID (NE ID): Identifies the network element through which the service flow passes, such as device name, IP address, or device number;

[0068] Timestamp: Records the time when a service flow passes through this network element, accurate to the millisecond level;

[0069] Business Type: For example, VoIP, video streaming, data transmission, etc.

[0070] Geolocation: The user's geographical location, used for subsequent geographic coverage analysis;

[0071] Performance metrics include latency, packet loss rate, and bandwidth usage.

[0072] Network element node index (NE Index): Registered based on the network element node index through which the service flow passes.

[0073] Data acquisition employs a timed polling or event-triggered mechanism, with distributed acquisition agents deployed across various network elements. These agents retrieve raw service tagging information via standard interfaces such as SNMP, NetFlow, and sFlow, and then upload it to the central data warehouse. Table 1 below illustrates an example of the service tagging information reported by network elements:

[0074] Table 1: Examples of service tagging information reported by network elements

[0075]

[0076] To improve data quality and subsequent processing efficiency, the original business tagging information needs to be cleaned and normalized, including:

[0077] Data cleaning: Remove duplicate, invalid, or incorrectly formatted records, such as inconsistent timestamps or empty business IDs; use regular expression matching and outlier detection algorithms (such as the Z-score method) to identify and remove abnormal data;

[0078] Data normalization: Timestamps recorded by different network elements are uniformly converted to UTC time to ensure time consistency; geographic coordinates are standardized.

[0079] Data dimensionality reduction: Reduce redundant fields by using principal components analysis (PCA) or feature selection algorithms (such as information gain-based feature filtering) and retain fields that contribute significantly to path reconstruction and impact prediction.

[0080] The preprocessed raw business tagging information is stored in a structured format (such as Parquet or CSV) in a distributed file system (such as HDFS) for subsequent path reconstruction modules to use. Table 2 below illustrates an example of the stored business tagging information:

[0081] Table 2: Examples of Stored Business Tag Information

[0082]

[0083] In one implementation, optionally, business path information corresponding to the historical business is generated based on multiple business tagging information, including:

[0084] Based on the time information in the service tag information, sort the multiple service tag information to obtain the sorted multiple service tag information;

[0085] Based on the network element node sequence number in the service tagging information, the continuity and monotonically increasing nature of the network node sequence number of the sorted multiple service tagging information are verified to obtain the verification result.

[0086] If the verification result is successful, then the business path information corresponding to the historical business is generated based on the sorted business tag information.

[0087] In this embodiment of the invention, a service tag injection mechanism based on signaling messages can be adopted to perform real-time service tagging on key network elements in the 5G call process and dynamically inject service tagging information during signaling message transmission to ensure the integrity and accuracy of service path records.

[0088] Figure 3 This is a flowchart illustrating one embodiment of the business path knowledge base construction process in this invention. Figure 3 As shown, taking 5G messages as an example, the module processing flow begins with the collection of 5G signaling messages. The system captures signaling messages from interfaces such as N1, N2, N3, and N4, including key signaling processes such as registration requests, session establishment, and service requests. Each signaling message carries a specific service identifier to uniquely identify the service flow.

[0089] During the service tagging information injection phase, as the signaling message passes through each network element, that network element injects service tagging information during signaling processing. The service tagging information includes key fields such as service identifier, timestamp, network element identifier, and node role. Simultaneously, the network element node sequence number field in the signaling message automatically increments by 1 to ensure the sequential integrity of the path nodes.

[0090] During the service path record generation phase, service path records are generated based on the injected service tag information. Each service path record contains complete sequence information of network element nodes, reflecting the accurate order in which signaling messages pass through the network elements. The auto-incrementing mechanism of the network element node sequence numbers ensures the accuracy of path reconstruction and avoids confusion in node order.

[0091] In the business path information reconstruction phase, scattered business path records are aggregated and sorted according to business identifiers to restore the complete end-to-end business path. The change pattern of network element node numbers is analyzed to verify the completeness and rationality of the business path and identify abnormal business path situations.

[0092] Finally, the reconstructed business path information is stored in the business path knowledge base, forming a business path graph containing complete node sequences. The business path knowledge base records the complete transmission path of each service, the processing order of each network element, and the temporal characteristics of the path, laying the foundation for accurate impact analysis.

[0093] Figure 4 This is a flowchart illustrating another implementation of the business path knowledge base construction process in this invention. For example... Figure 4 As shown in the figure, this embodiment of the invention provides a service path reconstruction algorithm, which calculates based on the service tagging information of the service flow. The service path reconstruction algorithm is described in detail below:

[0094] The requirements for inputting raw signaling data include:

[0095] (1) The original business tagging information includes at least one of the following:

[0096] Service ID, Element ID, Timestamp, Node Index, Element Type, Interface Type;

[0097] (2) Data quality requirements:

[0098] Timestamps must be kept synchronized, and node numbers must be incremented continuously.

[0099] The calculation steps include:

[0100] (1) Data preprocessing stage:

[0101] Establish a set of data filtering rules to exclude non-business signaling and test data;

[0102] Perform field validation on the input data and mark abnormal data records;

[0103] Classify the data according to business identifiers and service types;

[0104] (2) Session association and sorting:

[0105] Create a business session group using the business identifier as the key;

[0106] Within each session group, sort by timestamp in ascending order;

[0107] Verify the continuity and monotonically increasing nature of the network element node sequence numbers;

[0108] (3) Path reconstruction processing:

[0109] Extract the sorted network element identifier sequence to form the initial service path;

[0110] Detect and handle abnormal situations (missing or duplicate) of network element node serial numbers;

[0111] Verify the logical rationality of the business path through time difference analysis;

[0112] (4) Construction of business path knowledge base:

[0113] Store the valid paths in a graph structure.

[0114] Establish a mapping relationship between business attributes and paths;

[0115] Calculate path characteristic indicators (hop count, latency distribution, etc.).

[0116] The output includes:

[0117] A structured business path knowledge base;

[0118] Business path integrity and quality assessment report;

[0119] Abnormal data recording and processing log;

[0120] Knowledge base version metadata and update timestamps.

[0121] In one implementation, optionally, based on a pre-built service path knowledge base, the set of affected services when the network element to be cut over becomes unavailable is predicted, including:

[0122] In the service path knowledge base, obtain the affected service path information, including the network element to be cut over, and multiple affected service identifiers corresponding to the affected service path information;

[0123] Based on multiple affected service identifiers, predict the set of affected services when the network element to be cut over becomes unavailable.

[0124] In one embodiment, optionally, the method further includes at least one of the following:

[0125] Obtain the business type corresponding to the affected business set;

[0126] Based on the affected service path information, analyze the location information of the network element to be cut over in the affected service path;

[0127] Predict the propagation risk level of the affected services in the aforementioned set of affected services;

[0128] If there is a backup path that bypasses the network element to be cut over for the affected service path, the service quality index of the backup path is obtained, and the switching feasibility of the backup path is analyzed to obtain a feasibility report.

[0129] Figure 5 This is a flowchart illustrating one embodiment of the process for predicting the impact of cutover in this invention. Figure 5 As shown, the cutover impact prediction engine utilizes the network element node sequence number in the service tag to perform accurate impact prediction analysis. Based on complete network element node sequence information, the cutover impact prediction engine can accurately assess the degree and scope of the impact of the cutover operation on the service path.

[0130] First, the system receives the cutover plan input, including parameters such as the list of network elements to be cut over, the operation type, and the planned cutover time. Based on the network element node sequence number in the service tagging information, it can accurately identify the position and role of each network element in the service path.

[0131] During the network element node sequence number analysis phase, the business path information stored in the business path knowledge base is queried to analyze the position of the network element to be cut over in each business path. By analyzing the network element node sequence number, the criticality of the network element in the business path can be determined, and core nodes and edge nodes can be distinguished.

[0132] The path integrity assessment step analyzes the impact of cutover operations on service path integrity based on the continuity check of network element node numbers. If the network element to be cut over occupies a critical position in the service path, its cutover may lead to path interruption; if the network element to be cut over occupies the end of the path, the impact is relatively small.

[0133] In the impact propagation analysis phase, the propagation path of the cutover operation's impact is simulated. Based on the sequential relationship of network element node numbers, the upstream and downstream propagation range of the impact can be predicted, and indirectly affected network elements and services can be identified.

[0134] The redundancy path detection step checks for the existence of backup paths that bypass the network element to be cut over. By analyzing the changing patterns of network element node numbers in historical path data, the multipath transmission characteristics of the service are identified, and the possibility of automatic service handover is assessed.

[0135] Finally, the engine generates a detailed impact prediction report, including node-level impact analysis, path disruption prediction, and business impact scope assessment. Based on accurate network element node sequence information, the report provides impact predictions accurate to specific business flows.

[0136] Figure 6 This is a flowchart illustrating another embodiment of the process for predicting the impact of cutover in this invention. Figure 6As shown in the figure, this embodiment of the invention provides a cutover impact propagation algorithm, which calculates the impact on network performance based on the set of affected services. The cutover impact propagation algorithm is described in detail below:

[0137] Input data requirements include:

[0138] (1) Cutover plan description file, including a list of network elements to be cut over, cutover plan time, and operation type (such as restart, upgrade or replacement).

[0139] (2) Business path knowledge base.

[0140] The calculation steps include:

[0141] (1) Initial impact analysis:

[0142] Search the business path knowledge base for all business paths that include network elements to be cut over;

[0143] Identify the set of services that are directly affected, i.e., the set of affected services;

[0144] Record the location information of the network element to be cut over in each service path;

[0145] (2) Calculation of impact propagation:

[0146] Analyze the topological role of the network element to be cut over in the service path;

[0147] For each affected service, simulate network element failure scenarios;

[0148] Calculate the scope and extent of the impact;

[0149] (3) Redundancy path assessment:

[0150] Check if there is an alternative path to bypass the network element to be cut over;

[0151] Evaluate the service quality indicators of alternative routes;

[0152] Assess the feasibility of automatic business switching;

[0153] (4) Factors affecting the synthesis of results:

[0154] Aggregate the analysis results of all affected businesses;

[0155] Classified by business type and customer level;

[0156] Calculate the overall impact index.

[0157] The output includes:

[0158] List of affected businesses and classification of impact types;

[0159] Business interruption risk level assessment;

[0160] Redundant path availability report;

[0161] A diagram illustrating the impact of the propagation path;

[0162] Prediction confidence index.

[0163] In one implementation, optionally, the impact of the unavailability of the network element to be cut over on the affected service set is quantitatively analyzed in multiple dimensions to obtain multi-dimensional impact quantification results, including:

[0164] Based on the business type, geographical location, and key business ratio corresponding to the affected business set, obtain the quantitative results of the impact on the business dimension;

[0165] Predict the changes in performance metrics corresponding to the affected service set to obtain quantitative results of service dimension impact;

[0166] Based on the scale of affected users and the scope of influence on member users corresponding to the set of affected services, the quantitative results of the user dimension impact are obtained;

[0167] The impact results of the business dimension, the service dimension, and the user dimension are normalized to obtain multi-dimensional impact quantification results.

[0168] Figure 7 This is a flowchart illustrating one embodiment of the multi-dimensional influence analysis process in this invention. Figure 7 As shown, this embodiment of the invention provides a multi-dimensional impact analysis algorithm. This algorithm evaluates the comprehensive impact of cutover operations from multiple dimensions based on accurate network element node sequence information and uses the network element node sequence number in the service tag information to perform refined impact analysis and prediction.

[0169] The business continuity analysis dimension focuses on the impact of network cutovers on service continuity. By analyzing the node sequence position of the target network element in the service path, its criticality to service continuity is assessed. The earlier the node sequence of a network element or the more critical its position, the greater its impact on service continuity.

[0170] The service quality prediction dimension is based on the performance data of network element node sequence information. It predicts the changes in service quality after the cutover, analyzes the performance characteristics of similar node locations in historical data, and predicts that the cutover may lead to changes in service quality such as increased latency and decreased throughput.

[0171] The user impact assessment dimension combines network element node serial numbers and user data to analyze the scope of the impact of network cutover on the user group. By analyzing the user group characteristics of the affected business path services, the degree of impact of network cutover on user experience is assessed.

[0172] The criticality analysis of node location assesses the criticality of each network element in the service path based on the node number of the network element. It considers factors such as the location of the network element in the service path, processing latency, and load conditions to calculate the criticality index of the network element.

[0173] The comprehensive risk rating stage integrates the analysis results from various dimensions to generate a comprehensive risk rating. Based on factors such as node criticality, business priority, and scope of impact, the risk score of each affected business is calculated and classified into levels.

[0174] Ultimately, a multi-dimensional impact analysis report is generated, including detailed content such as business interruption risk assessment, service quality prediction, and user impact analysis. The report provides accurate and reliable analysis results based on precise node sequence data.

[0175] Figure 8 This is a flowchart illustrating another implementation of the multi-dimensional influence analysis process in this invention. For example... Figure 8 As shown in the figure, this embodiment of the invention provides a multi-dimensional impact quantification algorithm. This multi-dimensional impact quantification algorithm adopts a weighted scoring method, which comprehensively considers the impact of three dimensions: business, performance, and user, and calculates the total impact score. The multi-dimensional impact quantification algorithm is described in detail below:

[0176] Input data requirements include:

[0177] (1) The set of affected services can be presented in list form;

[0178] (2) Business attribute data (SLA level, customer information, etc.);

[0179] (3) Historical performance benchmark data;

[0180] (4) User distribution and member user information;

[0181] (5) Weight configuration parameters (customizable).

[0182] The calculation steps include:

[0183] (1) Dimension indicator extraction:

[0184] Business-level statistics include the types of affected businesses, their geographical distribution, and the proportion of key businesses.

[0185] Predict changes in performance metrics (such as latency, packet loss rate, and availability) at the service level.

[0186] Calculate the scale of affected users and the scope of impact on member users from the user dimension;

[0187] (2) Data standardization processing:

[0188] Normalize all types of indicators;

[0189] Eliminate dimensional differences and unify them to the [0,1] interval;

[0190] Handling missing and outlier values;

[0191] (3) Weighted score calculation:

[0192] Apply weights to metrics across various dimensions;

[0193] The calculation involves three dimensions: business, service, and user.

[0194] Synthesizing multiple dimensions affects the quantification results;

[0195] (4) Risk level classification:

[0196] Risk levels are mapped based on the scoring results;

[0197] Generate a risk level distribution map;

[0198] High-risk areas of concern are marked.

[0199] The output includes:

[0200] Multi-dimensional impact analysis report;

[0201] Comprehensive risk score (e.g., 0 to 100 points);

[0202] Risk level classification (e.g., low, medium, high);

[0203] Detailed indicator data for each dimension;

[0204] Visual charts and heatmaps;

[0205] Recommendations for decision-making priorities.

[0206] In one embodiment, optionally, the method further includes at least one of the following:

[0207] Based on the multi-dimensional impact quantification results, risk level classifications are obtained, and a business impact heatmap is generated based on the risk level classifications and the geographical locations corresponding to the affected business set.

[0208] Based on the cutover plan time of each of the multiple cutover network elements, and the multi-dimensional impact quantification results, cutover time series information of multiple cutover network elements is generated.

[0209] Figure 9 This is a flowchart illustrating the cutover decision-making process in an embodiment of the present invention. Figure 9 As shown, the decision support and visualization interface, based on precise node sequence data, provides intuitive visualizations and intelligent decision support. The interface presents analysis results through various visualization methods, helping users understand complex influencing relationships.

[0210] The node topology visualization view graphically displays the node topology structure of the service path. The interface uses network element node numbers to accurately represent the positional relationship of network elements in the service path, distinguishes core nodes from edge nodes through different colors and shapes, and highlights network elements to be cut over.

[0211] The sequence number and timing display view presents the network element processing timing information in a timeline format. Based on the network element node sequence number and timestamp, the interface accurately displays the time sequence and processing time of signaling messages passing through each network element, helping users understand the timing characteristics of the service flow.

[0212] The impact heatmap view uses a heat map format to display the distribution of impact levels. Based on network element node numbers and impact assessment data, the interface uses color coding to display the impact intensity at different network element locations, intuitively identifying high-risk areas.

[0213] Interactive analytics features allow users to delve deeper into the analysis results. Node drill-down analysis supports viewing detailed information and impact data for individual nodes; path comparison supports comparative analysis of different business paths; and time-series simulation demonstrations dynamically showcase the propagation process of the cutover operation's impact.

[0214] The intelligent solution generation module generates optimized cutover solutions based on accurate network element node sequence information. The system considers the service flow timing characteristics reflected by the network element node sequence numbers, recommending the optimal cutover time sequence and operation order to minimize service impact.

[0215] The optimization recommendation section provides specific operational suggestions, including cutover timelines, business migration strategies, and contingency plans. These recommendations are based on accurate node sequence data analysis to ensure the feasibility and effectiveness of the proposed solutions.

[0216] Finally, the interface outputs a complete decision support report, including visual charts, detailed data analysis, and specific operational recommendations. Based on accurate network element node sequence information, the report provides a reliable basis for decision-making during network cutover operations.

[0217] Figure 10 This is a schematic diagram of the architecture of an application system for the network cutover service impact prediction method described in an embodiment of the present invention. Figure 10As shown, this invention provides a service impact prediction system for network cutovers, addressing issues such as potential service interruptions, performance degradation, and decreased user experience caused by a lack of comprehensive assessment of service impacts during network cutovers in existing technologies. By constructing a service path knowledge base, predicting cutover impacts, quantifying impacts from multiple dimensions, and generating optimization suggestions, the system achieves intelligent auxiliary decision-making for network cutover operations.

[0218] First, during the service information recording stage, when a user uses a mobile communication network service, the network service will mark and record the service information involved in the service on each network element it passes through as the data passes through, thus obtaining service marking information. This service marking information includes service identifier, area identifier (the originating area and target area of ​​the service, such as calling Beijing from Nanchang), and network element node sequence number (the Nth node in the path).

[0219] Secondly, during the business path knowledge base construction phase, trace data dynamically recorded by each network element during the end-to-end transmission of historical business flows is collected and stored, including key information such as business identifiers, network element identifiers, and timestamps. Based on this data, the historical actual paths of various services are reconstructed in reverse, and a business path knowledge base is established to record the static and dynamic relationships between services, paths, and network elements, providing a data foundation for subsequent impact analysis.

[0220] Secondly, in the cutover impact prediction phase, in response to the received cutover plan, the network element to be cut over and its cutover plan time are extracted. Based on the service path knowledge base, predictive simulation is performed to simulate the network state when the network element is unavailable. Service flows that pass through the network element in all historical service paths are selected to form a set of affected services.

[0221] Subsequently, in the multi-dimensional impact quantification analysis phase, a multi-dimensional impact quantification analysis was conducted on the affected business set, covering three levels: business impact scope, performance degradation prediction, and user impact assessment. Specifically, this included counting the number of affected business types, geographical coverage, and the proportion of key businesses; predicting changes in performance indicators of business paths after the cutover based on historical performance data; estimating the scale of affected users, the proportion of member users, and the degree of decline in user perceived quality, ultimately generating a multi-dimensional impact quantification report.

[0222] Finally, based on the multi-dimensional impact quantification report, optimization suggestions such as cutover operation recommendations, business migration plans, or execution priorities are provided to assist operations and maintenance personnel in formulating scientific and reasonable cutover strategies, thereby effectively reducing cutover risks and improving network operations and maintenance efficiency and service quality.

[0223] The specific implementation process of the embodiments of the present invention will be described below:

[0224] (1) Data collection and preprocessing: Deploy collection agents on each network element to collect business tag information on a regular basis and upload it to the central data warehouse. Clean, normalize and reduce the dimensionality of the data to form a structured dataset.

[0225] (2) Business path knowledge base construction: Call the path reconstruction algorithm to reverse reconstruct the historical path of the business based on the business tag information, and establish the business path mapping relationship and store it in the graph database.

[0226] (3) Cutover Impact Prediction: The operation and maintenance personnel input the cutover plan (e.g., the network element to be cut over is NE-001, and the cutover plan time is 2025-04-05 10:00:00). The prediction engine is called to simulate the network state when the network element to be cut over is unavailable, and the service flows corresponding to all historical service paths passing through the network element to be cut over are filtered out to form a set of affected services.

[0227] (4) Multi-dimensional impact quantitative analysis: The affected business set was analyzed. In terms of the scope of business impact, the number of affected business types was 12, and the geographical coverage involved 5 provinces, of which membership business accounted for 25%. In terms of performance degradation prediction, the average latency of the affected business was predicted to increase by about 15ms and the packet loss rate increased by 0.5%. In terms of user impact assessment, the scale of affected users was about 100,000, of which membership users accounted for 10%, and the user perception quality was expected to decrease by about 12%.

[0228] (5) Generate impact report and optimization suggestions: Generate a multi-dimensional impact quantification report and make optimization suggestions based on the multi-dimensional impact quantification report, such as suggesting that the cutover plan time be adjusted to 2:00-4:00 am to avoid affecting the user experience during peak hours; or suggesting that the membership business be temporarily migrated to reduce the risk of service quality decline.

[0229] Furthermore, this invention can be extended to the following implementation methods: replacing the open-source model, in the path reconstruction module, a path prediction model based on graph neural network (GNN) can be adopted to improve the accuracy of path reconstruction; replacing the communication method, in the data acquisition stage, an edge computing architecture can be adopted to realize local preprocessing and compression of data and reduce transmission overhead; in terms of project deployment, it can be deployed in a private cloud or hybrid cloud environment to support high-concurrency access and real-time analysis requirements.

[0230] Therefore, by constructing a business path knowledge base, predicting the impact of cutovers, quantifying the impact from multiple dimensions, and generating optimization suggestions, this embodiment of the invention achieves intelligent auxiliary decision-making for network cutovers, effectively reducing cutover risks and improving network operation and maintenance efficiency and service quality.

[0231] In summary, the service impact prediction method for network cutover described in this embodiment of the invention has the following beneficial effects:

[0232] (1) Accurately predict the impact of services and improve the scientific nature of cutover decisions: The embodiments of the present invention construct a service path knowledge base based on the service marking information of historical services, which can accurately identify which service flows will be affected by the network elements to be cut over, avoiding the problems of incomplete information and misjudgment caused by relying on manual verification and network topology diagrams in the prior art, and significantly improving the accuracy and comprehensiveness of the impact assessment before network cutover.

[0233] (2) Multi-dimensional marketing quantitative analysis to comprehensively assess cutover risks: The embodiments of this invention introduce three dimensions of quantitative analysis: business impact scope, performance degradation prediction and user impact degree. It not only focuses on the impact at the business level, but also considers the changes in user perceived quality, thereby achieving a comprehensive assessment of cutover risks and providing more comprehensive decision-making basis for operation and maintenance personnel.

[0234] (3) Supports dynamic path reconstruction to adapt to network changes: By reconstructing historical service paths in reverse, it can dynamically adapt to changes in network structure and service flow, avoiding the problem that static service path configuration cannot reflect the actual service flow, and improving the real-time performance and applicability of prediction results.

[0235] (4) Reduce manual intervention and improve operation and maintenance efficiency: In the existing technology, the network cutover impact assessment relies on manual verification and experience judgment, which is inefficient and prone to errors. However, the embodiments of the present invention realize the full-process automation from data collection to impact prediction and optimization suggestions by automating data collection, processing and analysis, which greatly improves the intelligence level and work efficiency of network operation and maintenance.

[0236] (5) Supports the generation of optimization suggestions to assist in the formulation of scientific cutover strategies: It can not only identify the set of affected services, but also generate specific optimization suggestions based on the results of multi-dimensional impact quantitative analysis, such as the selection of cutover plan time and business migration scheme, which effectively reduces the risks of business interruption, performance degradation and user complaints that may be caused during the cutover process.

[0237] (6) Enhance network stability and service quality: By predicting and avoiding high-risk cutover operations in advance, it helps to ensure the continuity and stability of network operation, reduce service interruption events caused by network cutover, and improve user service quality and satisfaction.

[0238] (7) Good scalability and compatibility: The application system of this invention can adopt a modular architecture design, support multiple data acquisition methods (such as SNMP, NetFlow, sFlow) and algorithm replacement (such as GNN path prediction model), which facilitates subsequent function expansion and technology upgrade, and adapts to different network environments and business needs.

[0239] See Figure 11 This invention also provides a network cutover service impact prediction device, comprising:

[0240] The prediction module 1101 is used to predict the set of affected services when the network element to be cut over becomes unavailable, based on a pre-built service path knowledge base; wherein, the service path knowledge base is used to store the mapping relationship between service identifiers and service path information; the service path information is used to indicate multiple network elements through which the service flow passes; the set of affected services includes multiple affected services, and the service flow of the affected services passes through the network element to be cut over;

[0241] The analysis module 1102 is used to quantitatively analyze the impact of the unavailability of the network element to be cut over on the set of affected services in multiple dimensions, and obtain multi-dimensional impact quantification results.

[0242] Optionally, the network cutover service impact prediction device further includes:

[0243] The first acquisition module is used to acquire service tag information recorded by multiple network elements during the transmission of historical service service flows. The service tag information includes at least one of the following: service identifier, network element identifier, time information, and network element node sequence number.

[0244] The first generation module is used to generate service path information corresponding to the historical service based on multiple service tag information, wherein the service path information is used to indicate the time sequence information of the service flow of the historical service through multiple network elements;

[0245] The storage module is used to store the mapping relationship between the business identifier and the business path information in the business path knowledge base.

[0246] Optionally, in the network cutover service impact prediction device, the generation module is specifically used for:

[0247] Based on the time information in the service tag information, sort the multiple service tag information to obtain the sorted multiple service tag information;

[0248] Based on the network element node sequence number in the service tagging information, the continuity and monotonically increasing nature of the network node sequence number of the sorted multiple service tagging information are verified to obtain the verification result.

[0249] If the verification result is successful, then the business path information corresponding to the historical business is generated based on the sorted business tag information.

[0250] Optionally, in the network cutover service impact prediction device, the prediction module 1101 is specifically used for:

[0251] In the service path knowledge base, obtain the affected service path information, including the network element to be cut over, and multiple affected service identifiers corresponding to the affected service path information;

[0252] Based on multiple affected service identifiers, predict the set of affected services when the network element to be cut over becomes unavailable.

[0253] Optionally, the network cutover service impact prediction device further includes at least one of the following:

[0254] The second acquisition module is used to acquire the business type corresponding to the affected business set;

[0255] The location analysis module is used to analyze the location information of the network element to be cut over in the affected service path based on the affected service path information;

[0256] The risk prediction module is used to predict the propagation risk level of the affected services in the affected service set;

[0257] The third acquisition module is used to acquire the service quality index of the backup path if there is a backup path that bypasses the network element to be cut over, and to analyze the switching feasibility of the backup path and acquire a feasibility report.

[0258] Optionally, in the network cutover service impact prediction device, the analysis module 1102 is specifically used for:

[0259] Based on the business type, geographical location, and key business ratio corresponding to the affected business set, obtain the quantitative results of the impact on the business dimension;

[0260] Predict the changes in performance metrics corresponding to the affected service set to obtain quantitative results of service dimension impact;

[0261] Based on the scale of affected users and the scope of influence on member users corresponding to the set of affected services, the quantitative results of the user dimension impact are obtained;

[0262] The impact results of the business dimension, the service dimension, and the user dimension are normalized to obtain multi-dimensional impact quantification results.

[0263] Optionally, the network cutover service impact prediction device further includes at least one of the following:

[0264] The fourth acquisition module is used to acquire risk level classification based on the multi-dimensional impact quantification results, and generate a business impact heatmap based on the risk level classification and the geographical location corresponding to the affected business set.

[0265] The second generation module is used to generate cutover time series information of multiple network elements to be cut over based on the cutover plan time of each of the multiple network elements to be cut over and the multi-dimensional influence quantification results.

[0266] It should be noted that the apparatus provided in the embodiments of the present invention can implement all the method steps implemented in the above-mentioned network cutover service impact prediction method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0267] This invention also provides a network cutover service impact prediction device, such as... Figure 12 As shown, it includes:

[0268] The processor 1201, memory 1202, transceiver 1203, and programs or instructions stored in the memory 1202 and executable on the processor 1201; when the processor 1201 executes the programs or instructions, it implements the various processes of the above-described network cutover service impact prediction method embodiment and achieves the same technical effect. To avoid repetition, these will not be described again here.

[0269] The transceiver 1203 is used to receive and send data under the control of the processor 1201.

[0270] Among them, Figure 12 In this context, the bus architecture can include any number of interconnected buses and bridges, specifically connecting various circuits of one or more processors represented by processor 1201 and memory represented by memory 1202. The bus architecture can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 1203 can be multiple elements, including transmitters and receivers, providing a unit for communicating with various other devices over a transmission medium. For different user equipment, the user interface 1204 can also be an interface capable of connecting external or internal devices, including but not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0271] The processor 1201 is responsible for managing the bus architecture and general processing, while the memory 1202 can store the data used by the processor 1201 when performing operations.

[0272] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described network cutover service impact prediction method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0273] This invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the various processes of the above-described network cutover service impact prediction method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0274] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0275] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0276] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method of predicting service impact of network disconnection, characterized by, The method comprises the following steps: According to the pre-constructed service path knowledge base, the affected service set when the to-be-interworked network element is unavailable is predicted; wherein, the service path knowledge base is used to store the mapping relationship between the service identifier and the service path information; the service path information is used to indicate the multiple network elements through which the service flow passes; the affected service set comprises multiple affected services, and the service flow of the affected service passes through the to-be-interworked network element; The influence of the unavailability of the to-be-interworked network element on the affected service set is quantitatively analyzed in multiple dimensions to obtain multi-dimensional influence quantization results.

2. The method of claim 1, wherein, The method further comprises the following steps: Obtain the service mark information recorded by multiple network elements in the transmission process of the service flow of the historical service, wherein the service mark information comprises at least one of the service identifier, the network element identifier, the time information and the network element node sequence number; According to the multiple service mark information, the service path information corresponding to the historical service is generated, and the service path information is used to indicate the time sequence information of the service flow of the historical service passing through the multiple network elements; The mapping relationship between the service identifier and the service path information is stored in the service path knowledge base.

3. The method of claim 2, wherein, According to the multiple service mark information, the service path information corresponding to the historical service is generated, comprising the following steps: According to the time information in the service mark information, the multiple service mark information is sorted to obtain the sorted multiple service mark information; According to the network node sequence number in the service mark information, the continuity and monotonicity of the network node sequence number of the sorted multiple service mark information are verified to obtain a verification result; If the verification result is passed, the service path information corresponding to the historical service is generated according to the sorted multiple service mark information.

4. The method of claim 1, wherein, According to the pre-constructed service path knowledge base, the affected service set when the to-be-interworked network element is unavailable is predicted, comprising the following steps: In the service path knowledge base, the affected service path information including the to-be-interworked network element and the multiple affected service identifiers corresponding to the affected service path information are obtained; According to the multiple affected service identifiers, the affected service set when the to-be-interworked network element is unavailable is predicted.

5. The method of claim 1, wherein, The method further comprises at least one of the following steps: Obtain the service type corresponding to the affected service set; According to the affected service path information, the position information of the to-be-interworked network element in the affected service path is analyzed; The propagation risk level of the affected service in the affected service set is predicted; If the affected service path has a backup path bypassing the to-be-interworked network element, the service quality index of the backup path is obtained, and the switching feasibility of the backup path is analyzed to obtain a feasibility report.

6. The method of claim 1, wherein, In multiple dimensions, the influence of the unavailability of the to-be-interworked network element on the affected service set is quantitatively analyzed to obtain multi-dimensional influence quantization results, comprising the following steps: According to the service type, the geographical position and the proportion of the key service corresponding to the affected service set, the service dimension influence quantization result is obtained; The performance index change corresponding to the affected service set is predicted to obtain a service dimension influence quantization result; According to the affected user scale corresponding to the affected business set and the member user influence range, a user dimension influence quantification result is obtained; The business dimension influence result, the service dimension influence result, and the user dimension influence quantification result are normalized to obtain a multi-dimension influence quantification result.

7. The method of claim 1, wherein, The method further includes at least one of the following: According to the multi-dimension influence quantification result, a risk level classification is obtained, and a business influence heat map is generated according to the risk level classification and the geographical location corresponding to the affected business set; According to the multi-dimension influence quantification result, a risk level classification is obtained, and a business influence heat map is generated according to the risk level classification and the geographical location corresponding to the affected business set; 8. A network outage service impact prediction apparatus, characterized by, According to the multi-dimension influence quantification result, a risk level classification is obtained, and a business influence heat map is generated according to the risk level classification and the geographical location corresponding to the affected business set. The method comprises: A prediction module is configured to predict an affected business set when a to-be-cut network element is unavailable according to a pre-constructed business path knowledge base; the business path knowledge base is used to store a mapping relationship between a business identifier and business path information; the business path information is used to indicate a plurality of network elements through which a business flow passes; the affected business set comprises a plurality of affected businesses, and the business flow of the affected business passes through the to-be-cut network element; 9. A network outage service impact prediction device, characterized by, An analysis module is configured to quantitatively analyze the influence of the unavailability of the to-be-cut network element on the affected business set in multiple dimensions to obtain a multi-dimension influence quantification result. The method comprises:

10. A computer-readable storage medium, characterized in that, A processor, a memory, and a program or instruction stored on the memory and executable on the processor, wherein the processor executes the program or instruction to implement the network cutting business influence prediction method according to any one of claims 1 to 7.

11. A computer program product, characterised in that, The computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the network cutting business influence prediction method according to any one of claims 1 to 7. The computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the network cutting business influence prediction method according to any one of claims 1 to 7.

Images