Information technology consultation visualization platform based on digital twinning

Abstract

The invention discloses an information technology consultation visualization platform based on digital twinning, and relates to the technical field of information technology consultation. According to the invention, the multi-dimensional operation parameters of the enterprise information technology architecture are obtained in real time through the data acquisition and preprocessing module, and the dynamic architecture view reflecting the real operation condition of the current system is constructed by using the dynamic state mapping module; the problem of suggestion lag and inaccuracy caused by dependence on historical data and expert experience in a traditional consultation mode is solved; the visual interaction module and the intelligent recommendation and decision support module are combined, so that the information display efficiency and the communication effect are greatly improved and the decision period is shortened through a visual three-dimensional visual interface and an intelligent recommendation function; the resource optimization suggestion module and the virtual verification and scheme confirmation module provide a scientific resource optimization scheme and verified implementation guidance for enterprises, the implementation risk is reduced, and the feasibility and effectiveness of the scheme are ensured.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of information technology consultation, and particularly relates to an information technology consultation visual platform based on digital twinning. BACKGROUND

[0002] As a core driving force of digital transformation, enterprise information technology consultation plays a vital role in the modern business environment. With the increasing complexity of enterprise information systems, the accuracy and timeliness of consultation services are increasingly demanding. Traditional consultation models have been unable to meet the rapidly changing business needs of enterprises, and current information technology consultation solutions generally have significant defects. Existing consultation tools mainly rely on historical data analysis and expert experience judgment, and lack dynamic perception ability of enterprise information technology architecture and business processes, resulting in time difference and accuracy deviation between consultation suggestions and actual running state. At the same time, the information display method in the traditional consultation process is single, and customers have difficulty in intuitively understanding complex technical solutions, resulting in low communication efficiency and long decision-making cycle. SUMMARY

[0003] The purpose of the present application is to provide an information technology consultation visual platform based on digital twinning, which significantly improves the accuracy, communication efficiency and decision-making scientificity of information technology consultation.

[0004] The purpose of the present application can be achieved by the following technical solutions: The present application provides an information technology consultation visual platform based on digital twinning, comprising: A data acquisition and preprocessing module acquires multi-dimensional running parameters in real time by deploying distributed data acquisition nodes at key node positions of enterprise information technology architecture; A dynamic state mapping module constructs a dynamic state mapping model of enterprise information technology architecture according to the acquired real-time running parameters, converts the physical architecture into a digital architecture topology graph through a state mapping algorithm, and obtains a dynamic architecture view reflecting the real running condition of the current system; A system behavior analysis module uses state data in the dynamic architecture view as an input source, identifies system performance change trends and resource usage patterns through time series analysis methods, and obtains system behavior analysis results containing trend characteristics and abnormal identifiers; A prediction analysis module uses a machine learning prediction model to perform prediction analysis on future system load, resource demand and performance bottlenecks according to trend characteristic data in the system behavior analysis results, calculates prediction confidence through historical pattern matching and regression analysis algorithms, and determines system running state prediction values in a future time window; A resource optimization suggestion module generates resource optimization suggestions by combining current architecture configuration information after obtaining the prediction analysis results, and obtains targeted architecture adjustment schemes and resource configuration suggestions; The visualization interaction module integrates the dynamic architecture view, the prediction analysis result and the optimization suggestion into an interactive visualization interface through a three-dimensional visualization rendering engine, adopts different colors and graphical symbols to identify the system state level and the risk degree, judges the user focus and provides detailed information display; The intelligent recommendation and decision support module intelligently recommends corresponding decision support information and solution options according to the user's interactive behavior and focus in the visualization interface, analyzes the cost-benefit ratio of different schemes through a decision tree algorithm, determines the optimal decision path and implementation priority.

[0005] Further, the data collection and preprocessing module specifically includes: The distributed collection nodes are deployed at key positions of the enterprise information technology architecture to obtain server performance indicators, network traffic data, application program running states and database access frequency multidimensional running parameters, a polling mechanism is adopted to periodically scan the system components, state changes are judged, and component state data sets are obtained; The component state data sets are processed through a time series analysis algorithm to determine abnormal parameter sets, traffic peak features are extracted from the network traffic data to obtain traffic anomaly data sets, and the application program running state and the database access frequency are obtained; A decision tree algorithm is used to analyze the traffic anomaly data set to determine the application program running abnormal reason to obtain an abnormal reason set, and a clustering algorithm is used to group the database access frequency to obtain an access mode set.

[0006] Further, the dynamic state mapping module specifically includes: The collected real-time multidimensional running parameters are stored in a data warehouse to obtain system component state data sets, and if the parameter values in the system component state data sets exceed the preset threshold range, a state update mechanism is activated through an event trigger to determine the component states that need to be updated; According to the trigger signal of the state update mechanism, a state mapping algorithm is used to process the system component state data sets to generate a state snapshot of the physical architecture data, and the state snapshot of the physical architecture data is converted into a digital architecture topology through the state mapping algorithm to obtain a digital topology description file; A visualization tool is used to render the digital topology description file to generate a dynamic architecture view reflecting the current system running state, and according to the topology structure in the dynamic architecture view, the abnormal mode of the system component state is detected to obtain potential system bottleneck points; By analyzing the potential system bottleneck points, the parameter weights of the state mapping algorithm are adjusted, the digital architecture topology is optimized, and a dynamic architecture view reflecting the real running state of the current system is obtained.

[0007] Further, the system behavior analysis module specifically includes: Obtain state data from a dynamic architecture view, remove noise data through preprocessing to obtain a cleaned state data set, decompose the cleaned state data set using a time series analysis method, extract performance change trends and resource usage patterns, and obtain a trend feature set; If there is fluctuation deviating from the preset threshold in the trend feature set, mark it as an abnormal fluctuation point through an anomaly detection algorithm to obtain an abnormal identifier set, construct a system behavior feature matrix according to the abnormal identifier set and the trend feature set, and obtain a system behavior analysis intermediate result; Classify the system behavior feature matrix through cluster analysis, identify potential risk points, obtain a risk point set, analyze the relationship between the risk point set and the resource usage pattern using an association rule mining method, obtain an association pattern of risk points and resource usage, optimize the system behavior analysis intermediate result according to the association pattern, and obtain a system behavior analysis result containing trend features and abnormal identifiers.

[0008] Further, the prediction analysis module specifically includes: According to the trend feature data in the system behavior analysis result, a linear regression model is trained to generate a system load prediction model, a load prediction value in a future time window is obtained, a support vector machine model is used to calculate a resource demand prediction value in combination with the resource allocation records in the historical data, and the future resource allocation demand is determined; If the resource demand prediction value exceeds the preset threshold, a decision tree model is used to identify potential performance bottlenecks through the performance bottleneck records in the historical data to obtain a bottleneck prediction result, a pattern matching method is used to analyze the running state pattern in the historical data, a prediction confidence is calculated, and a confidence evaluation value is obtained; The confidence evaluation value is combined with the load prediction value and the resource demand prediction value in the time window to determine a future system running state prediction value, the prediction model parameters are adjusted in combination with the dynamic changes in the trend data, and the system running state prediction value in the future time window is obtained.

[0009] Further, the resource optimization suggestion module specifically includes: Obtain the prediction analysis result, analyze the historical load data and the current running state through a machine learning model to obtain a load change trend, calculate the matching degree of resource utilization and capacity threshold in combination with the current architecture configuration information, and obtain a resource state evaluation; If the resource state evaluation shows that the predicted load exceeds the capacity threshold, generate an expansion suggestion through a load balancing strategy to determine a new resource configuration scheme, and if the resource state evaluation shows that the resource utilization rate is lower than the efficiency threshold, analyze the current architecture configuration information to generate a contraction suggestion and determine a resource reduction configuration scheme; The architecture adjustment scheme is generated through the capacity expansion suggestion and the capacity reduction suggestion, in combination with a load balancing strategy, to obtain a preliminary resource configuration optimization result, the feasibility of the load balancing strategy is verified in combination with current architecture configuration information, to obtain a final resource configuration suggestion, and a targeted architecture adjustment scheme is generated according to the final resource configuration suggestion, and an optimized resource allocation scheme is output.

[0010] Further, the visual interaction module specifically comprises: The dynamic architecture data, the prediction analysis result and the optimization suggestion are acquired through the three-dimensional rendering engine, interactive visual interface content is integrated and generated, the state level and the risk degree in the visual interface are identified by color coding and graphical symbols, and a visual identification mapping table is generated; If a user interaction operation is triggered, the interaction data is extracted according to the operation type, the state level and the risk degree corresponding to the interaction data are judged, the visual interface content is updated, the user focus is analyzed through the user interaction data, the dynamic architecture subset and the prediction analysis result corresponding to the focus are acquired, and a detailed information data set is generated; The detailed information data set is classified by using a support vector machine algorithm, the priority of the focus data is judged, the classified focus content is obtained, the three-dimensional rendering view after dynamic adjustment is generated in combination with the optimization suggestion, the interactive interface display is updated, the user focus is judged according to the updated interactive interface display, and detailed information display is provided.

[0011] Further, the intelligent recommendation and decision support module specifically comprises: The user interaction behavior data is acquired by recording the clicks, dwell time and browsing path of the user in the visual interface, the user focus is determined, the user preference model is obtained by using a clustering algorithm to analyze the user preference; The decision support information and the optimization scheme are generated in combination with preset business rules, the recommendation content is determined, if the user selects a certain optimization scheme, the cost-benefit ratio of the scheme is analyzed by using a decision tree algorithm, the evaluation result of each scheme is obtained, and the optimal decision path is determined; The implementation plan is generated according to the optimal decision path, the priority of the implementation plan is sorted by using a heuristic algorithm, and the resource allocation scheme is automatically adjusted according to the sorting result, to generate a final execution scheme.

[0012] Further, it further comprises a virtual verification and scheme confirmation module, the proposed architecture adjustment scheme is verified in a virtual environment by using the optimal decision path to drive the dynamic simulation module, the system running state and performance after implementation of the scheme are simulated by using a simulation engine, and the final consulting suggestion and implementation guidance scheme after verification are obtained.

[0013] Further, the virtual verification and scheme confirmation module specifically comprises: An architecture adjustment scheme is acquired from a business requirement database, key parameters in the scheme are extracted by using a structured parsing method to obtain a standardized architecture adjustment description, a dynamic simulation module is called to construct a system model corresponding to the scheme in a virtual environment to obtain a virtual system configuration; A simulation engine is used to simulate running of a preset workload scenario on the virtual system configuration to obtain system running state data and performance data, and a simulation result set is obtained, if the performance data in the simulation result set meets a preset performance improvement threshold, a logical judgment module is used to confirm the feasibility of the scheme to obtain a feasibility verification result; According to the feasibility verification result, a suggestion generation algorithm is called to extract key performance indicators from the simulation result set to generate a structured consultation suggestion to obtain a consultation suggestion document, and a template matching method is used to generate a detailed implementation guidance scheme in combination with the consultation suggestion document and the architecture adjustment description to obtain a final implementation guidance scheme.

[0014] The beneficial effects of the present application are: The present application realizes real-time dynamic perception of enterprise information technology architecture through a data acquisition and preprocessing module and a dynamic state mapping module, distributed data acquisition nodes can obtain multi-dimensional running parameters in real time, and the dynamic state mapping module converts these data into a dynamic architecture view reflecting the real running state of the current system, not only solving the problem of time difference and accuracy deviation between the consultation suggestion and the actual running state caused by relying on historical data and expert experience in the traditional consultation mode, but also making the consultation suggestion accurately reflect the current state of the system through a real-time data driven dynamic model, providing a more timely and accurate basis for the decision of the enterprise; The combination of the visual interactive module and the intelligent recommendation and decision support module greatly improves the problem of single information display mode and customer's difficulty in understanding complex technical schemes in the traditional consultation process, the dynamic architecture view, prediction analysis result and optimization suggestion are integrated into an interactive visual interface by a three-dimensional visual rendering engine, and color coding and graphical symbol identification system state level and risk degree are used to enable the customer to intuitively understand the system running state and potential problems, at the same time, the intelligent recommendation function provides targeted decision support information and solution options according to the user's interactive behavior, further improving the communication efficiency, helping the customer to quickly make decisions, effectively shortening the decision cycle, enabling the enterprise to more quickly respond to market changes and technical challenges; The resource optimization suggestion module and the virtual verification and scheme confirmation module provide comprehensive resource optimization and scheme verification support for enterprises. The resource optimization suggestion module generates targeted architecture adjustment schemes and resource configuration suggestions according to the prediction analysis results and current architecture configuration information, helping enterprises to reasonably plan resources and avoid resource waste or deficiency. The virtual verification and scheme confirmation module verifies the proposed architecture adjustment schemes in a virtual environment through the dynamic simulation module, simulates the system running state and performance after the implementation of the scheme, and ensures the feasibility and effectiveness of the scheme. Not only does it solve the problem of high implementation risk caused by the lack of effective verification means in traditional consulting, but also provides an verified final consulting suggestion and implementation guidance scheme for enterprises, reduces the implementation risk, improves the scientificity of enterprise decision-making and the success rate of implementation, and effectively promotes the digital transformation process of enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to better understand and implement, the technical solutions of the present application are described in detail below in combination with the drawings.

[0016] Fig. 1 A process schematic diagram of an information technology consulting visual platform based on digital twinning provided for Embodiment 1 of the present application; Fig. 2 A process schematic diagram of a dynamic state mapping module in an information technology consulting visual platform based on digital twinning provided for Embodiment 1 of the present application; Fig. 3 A process schematic diagram of an intelligent recommendation and decision support module in an information technology consulting visual platform based on digital twinning provided for Embodiment 1 of the present application. DETAILED DESCRIPTION

[0017] In order to better understand and implement, the technical solutions of the present application are described in detail below in combination with the drawings.

[0018] The terms used in the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "an" and "the" used in the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein means and includes any or all possible combinations of one or more associated listed items.

[0019] The specific implementation, features and effects according to the present application are described in detail below in combination with the drawings and preferred embodiments. Embodiments

[0020] Please refer to Figs. 1-3 The embodiment provides an information technology consultation visualization platform based on digital twinning, which comprises: The data acquisition and preprocessing module acquires multi-dimensional running parameters of server performance indexes, network traffic data, application program running states and database access frequencies by deploying distributed data acquisition nodes at key node positions of enterprise information technology architecture, and the data acquisition nodes periodically scan target system component state changes by using a polling mechanism. Further, the data acquisition and preprocessing module specifically comprises: The distributed acquisition nodes are deployed at key positions of enterprise information technology architecture to acquire multi-dimensional running parameters of server performance indexes, network traffic data, application program running states and database access frequencies, and the system components are periodically scanned by using a polling mechanism to determine state changes and obtain component state data sets. The component state data sets are processed by a time series analysis algorithm to determine abnormal parameter sets, traffic peak features are extracted from the network traffic data to obtain traffic abnormal data sets, and the application program running states and database access frequencies are determined. The traffic abnormal data sets are analyzed by using a decision tree algorithm to determine application program running abnormal reasons and obtain abnormal reason sets, and the database access frequencies are grouped by using a clustering algorithm to obtain access mode sets.

[0021] Specifically, the distributed data acquisition nodes are deployed at key nodes of enterprise information technology architecture to acquire multi-dimensional running parameters in real time, and the system component state changes are periodically scanned by using a polling mechanism, which solves the problem of lack of dynamic perception ability of enterprise information technology architecture in traditional consultation tools, can not only perceive system state changes in real time, but also accurately identify abnormal conditions and analyze their reasons, thereby providing accurate and timely data support for subsequent consultation analysis and improving the accuracy and timeliness of consultation suggestions.

[0022] The dynamic state mapping module constructs a dynamic state mapping model of enterprise information technology architecture according to the acquired real-time running parameters, converts the physical architecture into a digital architecture topology graph by using a state mapping algorithm, and obtains a dynamic architecture view reflecting the real running conditions of the current system.

[0023] Further, the dynamic state mapping module specifically comprises: S11, store the collected real-time multi-dimensional running parameters in the data warehouse to obtain a system component state dataset, if the parameter value in the system component state dataset exceeds the preset threshold range, activate the state update mechanism through an event trigger to determine the component state that needs to be updated; S12, according to the trigger signal of the state update mechanism, process the system component state dataset using a state mapping algorithm to generate a state snapshot of the physical architecture data, convert the state snapshot of the physical architecture data into a digital architecture topology through the state mapping algorithm, and obtain a digital topology description file; S13, render the digital topology description file using a visualization tool to generate a dynamic architecture view reflecting the current system running status, detect abnormal patterns of system component states according to the topology structure in the dynamic architecture view, and obtain potential system bottleneck points; S14, adjust the parameter weight of the state mapping algorithm by analyzing the potential system bottleneck points, optimize the digital architecture topology, and obtain a dynamic architecture view reflecting the real running status of the current system.

[0024] In the dynamic state mapping module, the digital topology description file is rendered by a visualization tool to generate a dynamic architecture view, which can intuitively reflect the current system running status. In the view, the system will detect abnormal patterns of system component states according to preset rules and algorithms, especially potential bottleneck points affecting system performance. Once potential bottleneck points are detected, the system will further analyze the characteristics and influencing factors of these bottleneck points. Based on these analysis results, the system will automatically adjust the parameter weight in the state mapping algorithm. For example, if the load of a server node is too high causing a performance bottleneck, the system may increase the weight of this node in the state mapping algorithm to more accurately reflect its importance and influence range in subsequent topology graph updates. Through this dynamic adjustment, the optimized digital architecture topology can more accurately reflect the actual running status of the system, providing more accurate basis for subsequent resource optimization and performance improvement, thereby improving the running efficiency and reliability of the entire system.

[0025] Specifically, by converting the collected real-time running parameters into a digital architecture topology graph, a dynamic model reflecting the real running status of enterprise information technology architecture is constructed, solving the technical problem that complex system architecture and running status cannot be intuitively presented in traditional consulting. By analyzing the bottleneck points and adjusting the algorithm parameter weight, the topology structure is optimized to ensure that the dynamic architecture view accurately reflects the real running status of the current system. This not only improves the visualization level of system status, but also provides accurate basis for subsequent analysis and optimization, enhancing the scientificity and practicality of consulting suggestions.

[0026] The system behavior analysis module takes state data in the dynamic architecture view as the input source, identifies system performance change trends and resource usage patterns through time series analysis, and marks potential risk points if abnormal fluctuation patterns are detected, obtaining system behavior analysis results containing trend characteristics and abnormal identifiers.

[0027] Further, the system behavior analysis module specifically includes: The state data is obtained from the dynamic architecture view, noise data is removed through preprocessing to obtain a cleaned state data set, and the cleaned state data set is decomposed using a time series analysis method to extract performance change trends and resource usage patterns, obtaining a trend feature set. The time series analysis method is applied to the cleaned state data set, which is first decomposed in chronological order to identify periodic, trend, and random components. Through this decomposition, the change trends of system performance, such as server load rise or fall, and resource usage patterns, such as memory and CPU usage peaks and troughs, can be clearly extracted. These extracted trends and patterns are integrated into the trend feature set, providing a key data foundation for subsequent anomaly detection and risk assessment, thus helping the system accurately identify potential problems and optimization directions.

[0028] If there are fluctuations in the trend feature set that deviate from the preset threshold, the abnormal fluctuation points are marked through an anomaly detection algorithm to obtain an abnormal identifier set. Based on the abnormal identifier set and the trend feature set, a system behavior feature matrix is constructed to obtain the system behavior analysis intermediate result. The system behavior feature matrix is classified through cluster analysis to identify potential risk points, obtaining a risk point set. The relationship between the risk point set and the resource usage pattern is analyzed using association rule mining methods to obtain the association pattern of risk points and resource usage. Based on the association pattern, the system behavior analysis intermediate result is optimized to obtain the system behavior analysis result containing trend characteristics and abnormal identifiers.

[0029] Specifically, by taking state data in the dynamic architecture view as the input source and using time series analysis and other methods, the technical problem of accurately identifying system performance change trends and resource usage patterns in traditional consulting is solved. Not only can the system accurately identify abnormal fluctuations and potential risks in system performance, but also can reveal the relationship between these risks and resource usage, providing a scientific basis for resource optimization and risk prevention for enterprises, significantly improving the accuracy and reliability of system behavior analysis.

[0030] The prediction analysis module uses a machine learning prediction model to perform prediction analysis on future system load, resource demand, and performance bottleneck based on trend feature data in the system behavior analysis result, calculates a prediction confidence through historical pattern matching and regression analysis algorithms, and determines a system running state prediction value in a future time window.

[0031] Further, the prediction analysis module specifically includes: According to trend feature data in the system behavior analysis result, a linear regression model is trained to generate a system load prediction model, a load prediction value in a future time window is obtained, a support vector machine model is used to calculate a resource demand prediction value in combination with resource allocation records in historical data, and future resource allocation demand is determined. Wherein, trend feature data in the system behavior analysis result is collected first, these data reflect the historical change trend of system load, and the linear regression model is trained using these data; during the training process, the model learns the relationship between system load and various features by minimizing the error between the prediction value and the actual value, and the trained linear regression model can predict the system load in a future time window according to the current feature data.

[0032] When calculating the resource demand prediction value, the system combines resource allocation records in historical data, the historical records contain resource usage and allocation decisions at different time points in the past; based on these data, a support vector machine (SVM) model is trained, the SVM model identifies key features and patterns in the data to establish a model for predicting future resource demand, and in the training process, the model finds the optimal hyperplane to maximize the separation between different categories of data, thereby improving the accuracy of prediction, the SVM model can predict the resource demand in a specific future time window according to the current system state and historical resource allocation patterns, providing a scientific basis for enterprise resource planning and ensuring the rationality and forward-looking nature of resource allocation.

[0033] If the resource demand prediction value exceeds the preset threshold, a decision tree model is used to identify potential performance bottlenecks through performance bottleneck records in historical data, obtain bottleneck prediction results, analyze running state patterns in historical data using pattern matching methods, calculate a prediction confidence, and obtain a confidence evaluation value. When the resource demand prediction value exceeds the preset threshold, the system triggers the performance bottleneck prediction process. First, the performance bottleneck records in the historical data are analyzed using a decision tree model. The decision tree model learns various features and conditions when performance bottlenecks occur in the historical data, constructs decision rules, and inputs the current running state features into the decision tree model. The model will determine whether there is a potential performance bottleneck according to the preset decision rules and output the bottleneck prediction result. At the same time, the system will analyze the running state patterns in the historical data using pattern matching methods. By identifying similar patterns in the current state, the system can evaluate the reliability of the prediction result. Specifically, the system calculates the similarity between the current state and the historical patterns and calculates the prediction confidence according to the similarity. If the current state is highly similar to the state in which a performance bottleneck occurs in the history, the prediction confidence will be correspondingly improved; conversely, if the similarity is low, the prediction confidence will be reduced. Finally, the system combines the bottleneck prediction result of the decision tree model and the confidence evaluation of the pattern matching method to obtain a comprehensive confidence evaluation value, which provides a scientific basis for subsequent resource optimization and performance improvement.

[0034] Through the confidence evaluation value, the load prediction value and the resource demand prediction value in the time window are determined to predict the future system running state, and the prediction model parameters are adjusted according to the dynamic changes in the trend data to obtain the system running state prediction value in the future time window.

[0035] Specifically, by using a machine learning prediction model, the technical problem of inaccurate prediction of future system load, resource demand and performance bottleneck in traditional consulting is solved. By dynamically adjusting the prediction model parameters to adapt to the changes in trend data, accurate future system running state prediction values can be provided to help enterprises plan resource allocation in advance, prevent performance problems, and thus optimize system performance and reduce operational risks.

[0036] The resource optimization suggestion module generates resource optimization suggestions based on the prediction analysis results and current architecture configuration information. If the predicted load exceeds the current capacity threshold, an expansion suggestion is generated. If the resource utilization rate is lower than the efficiency threshold, an optimization suggestion is generated to obtain targeted architecture adjustment schemes and resource configuration suggestions.

[0037] Further, the resource optimization suggestion module specifically includes: The prediction analysis result is obtained by analyzing the historical load data and the current running state through a machine learning model to obtain the load change trend. The matching degree of resource utilization rate and capacity threshold is calculated based on the current architecture configuration information to obtain the resource state evaluation. If the resource state evaluation shows that the predicted load exceeds the capacity threshold, a capacity expansion suggestion is generated through the load balancing strategy, and a new resource configuration scheme is determined. If the resource state evaluation shows that the resource utilization is lower than the efficiency threshold, the current architecture configuration information is analyzed to generate a capacity reduction suggestion, and a resource reduction configuration scheme is determined. Through the capacity expansion suggestion and the capacity reduction suggestion, combined with the load balancing strategy, an architecture adjustment scheme is generated to obtain a preliminary resource configuration optimization result. The feasibility of the load balancing strategy is verified in combination with the current architecture configuration information to obtain a final resource configuration suggestion. A targeted architecture adjustment scheme is generated according to the final resource configuration suggestion, and an optimized resource allocation scheme is output.

[0038] The specific process of generating the architecture adjustment scheme includes: according to the prediction analysis result and the resource state evaluation, the system generates a capacity expansion suggestion (when the predicted load exceeds the current capacity threshold) or a capacity reduction suggestion (when the resource utilization is lower than the efficiency threshold). The capacity expansion suggestion clearly indicates the type and quantity of resources that need to be added, such as increasing the number of servers or improving the network bandwidth. The capacity reduction suggestion indicates the resources that can be reduced to optimize costs. Then, the system combines the load balancing strategy to analyze how to reasonably allocate these added or reduced resources in the existing architecture to ensure the overall performance and stability of the system. The load balancing strategy considers multiple factors such as server load, network traffic distribution, application access mode, etc. to determine the optimal resource allocation scheme. Based on this information, the system generates an architecture adjustment scheme that describes in detail how to adjust the current architecture configuration, including resource addition, reduction or redistribution, to adapt to future load changes and optimize resource utilization. Finally, by verifying the feasibility of the load balancing strategy, it is ensured that the implementation of the architecture adjustment scheme can effectively improve the performance and efficiency of the system.

[0039] Specifically, by combining the prediction analysis result and the current architecture configuration information, the problem of lack of foresight and pertinence in traditional information technology consulting resource optimization suggestions is solved, not only improving resource utilization efficiency, but also ensuring the stability and reliability of the system when facing future load changes. It provides scientific and forward-looking resource optimization suggestions for enterprises, helping enterprises to efficiently use resources and reduce operating costs.

[0040] The visual interaction module integrates the dynamic architecture view, prediction analysis result, and optimization suggestion into an interactive visual interface through a three-dimensional visual rendering engine, uses different colors and graphical symbols to identify system state levels and risk levels, updates and displays content in real time when triggered by user operations, determines user focus points, and provides detailed information display.

[0041] Further, the visual interaction module specifically includes: The dynamic architecture data, the prediction analysis result and the optimization suggestion are obtained through a three-dimensional rendering engine, interactive visualization interface content is generated by integration, state levels and risk degrees in the visualization interface are identified by color coding and graphical symbols, and a visualization identification mapping table is generated; If a user interaction operation is triggered, interaction data is extracted according to an operation type, state levels and risk degrees corresponding to the interaction data are determined, the visualization interface content is updated, a focus point of the user is analyzed through user interaction data, a dynamic architecture subset and a prediction analysis result corresponding to the focus point are obtained, and a detailed information data set is generated; The detailed information data set is classified by using a support vector machine algorithm, the priority of the focus point data is determined, the classified focus point content is obtained, a three-dimensional rendering view after dynamic adjustment is generated by combining the optimization suggestion, the interactive interface display is updated, the focus point of the user is determined according to the updated interactive interface display, and detailed information display is provided.

[0042] If the focus point content changes, the dynamic architecture data is reobtained according to a change trend, the classification and rendering steps are cyclically executed, and a real-time updated visualization interface is obtained.

[0043] Specifically, the dynamic architecture view, the prediction analysis result and the optimization suggestion are integrated into an interactive visualization interface by a three-dimensional visualization rendering engine, the problems of single information display mode, difficulty for a customer to intuitively understand a complex technical scheme and low communication efficiency in traditional consultation are solved, the intuitiveness and interactivity of information display are significantly improved, the understanding of a complex system by the user is enhanced, and therefore the communication efficiency and decision-making speed are improved, and more efficient support is provided for digital transformation of an enterprise.

[0044] The intelligent recommendation and decision support module intelligently recommends corresponding decision support information and solution options according to the interaction behavior and focus point of the user in the visualization interface, generates an implementation plan and a risk assessment report automatically if the user selects a specific optimization suggestion, analyzes the cost-benefit ratio of different schemes by using a decision tree algorithm, and determines an optimal decision path and an implementation priority order.

[0045] Further, the intelligent recommendation and decision support module specifically includes: S21, user interaction behavior data is obtained by recording clicks, dwell time and browsing paths of the user in the visualization interface, a focus point of the user is determined, a user preference model is obtained by analyzing user preferences by using a clustering algorithm; S22, decision support information and optimization schemes are generated by combining a preset business rule through the user preference model, recommended content is determined, and if the user selects a certain optimization scheme, the evaluation results of the schemes are obtained by analyzing the cost-benefit ratio of the schemes by using a decision tree algorithm, and an optimal decision path is determined; S23, generating an implementation plan according to the optimal decision path, using a heuristic algorithm to prioritize the implementation plan, and automatically adjusting the resource allocation scheme according to the sorting result to generate a final execution scheme.

[0046] Specifically, by analyzing the interaction behavior and focus of the user in the visualization interface, the problems of inaccurate decision support information, difficult scheme selection and lack of scientific implementation plan in traditional consulting are solved, not only improving the scientificity and accuracy of decision-making, but also optimizing the efficiency and feasibility of the implementation plan, significantly improving the decision-making efficiency and resource utilization efficiency of the enterprise, and providing strong support for the digital transformation of the enterprise.

[0047] The virtual verification and scheme confirmation module uses the optimal decision path to drive the dynamic simulation module to verify the proposed architecture adjustment scheme in a virtual environment, simulates the system running state and performance after the implementation of the scheme through the simulation engine, and if the simulation result shows that the performance improvement reaches the expected target, the scheme is confirmed to be feasible, and the final consulting suggestion and implementation guidance scheme are obtained.

[0048] Further, the virtual verification and scheme confirmation module specifically includes: The architecture adjustment scheme is obtained from the business requirement database, the key parameters in the scheme are extracted using a structured analysis method to obtain a standardized architecture adjustment description, the dynamic simulation module is called to build a system model corresponding to the scheme in a virtual environment to obtain a virtual system configuration; The simulation engine is used to simulate the running of a preset workload scenario for the virtual system configuration, system running state data and performance data are obtained, a simulation result set is obtained, and if the performance data in the simulation result set meets a preset performance improvement threshold, the scheme is confirmed to be feasible by a logical judgment module, and a feasibility verification result is obtained; According to the feasibility verification result, a suggestion generation algorithm is called to extract key performance indicators from the simulation result set, generate a structured consulting suggestion, obtain a consulting suggestion document, and combine the consulting suggestion document and the architecture adjustment description to generate a detailed implementation guidance scheme using a template matching method to obtain a final implementation guidance scheme.

[0049] Specifically, by simulating and verifying the architecture adjustment scheme in a virtual environment, the problems of lack of effective scheme verification means and high implementation risk in traditional consulting are solved, not only ensuring the feasibility and effectiveness of the scheme, but also reducing the implementation risk and improving the scientificity and reliability of the consulting suggestion, providing a verified implementation path for the digital transformation of the enterprise.

[0050] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in any form. Although the present application has been disclosed with the preferred embodiments as above, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, as long as the changes or modifications do not deviate from the technical solution of the present application. Any modification, change, equivalent change and modification of the above embodiments made according to the technical essence of the present application still belong to the scope of the technical solution of the present application.

Claims

1. A visualization platform for information technology consulting based on digital twins, characterized by: include: The data acquisition and preprocessing module acquires multi-dimensional operating parameters in real time by deploying distributed data acquisition nodes at key nodes in the enterprise's information technology architecture. The dynamic state mapping module constructs a dynamic state mapping model of the enterprise's information technology architecture based on the collected real-time operating parameters. It converts the physical architecture into a digital architecture topology diagram through a state mapping algorithm, thereby obtaining a dynamic architecture view that reflects the current real operating status of the system. The system behavior analysis module uses the state data in the dynamic architecture view as the input source, identifies the system performance change trend and resource usage pattern through time series analysis, and obtains system behavior analysis results containing trend characteristics and anomaly indicators. The predictive analysis module uses machine learning prediction models to predict and analyze future system load, resource requirements, and performance bottlenecks based on trend feature data in the system behavior analysis results. It calculates prediction confidence through historical pattern matching and regression analysis algorithms to determine the predicted value of system operating status within the future time window. The resource optimization suggestion module generates resource optimization suggestions by combining the predictive analysis results with the current architecture configuration information, resulting in targeted architecture adjustment plans and resource configuration suggestions. The visualization and interaction module integrates dynamic architecture views, predictive analysis results, and optimization suggestions into an interactive visualization interface through a 3D visualization rendering engine. It uses different colors and graphic symbols to identify the system status level and risk level, determine the user's focus, and provide detailed information. The intelligent recommendation and decision support module intelligently recommends relevant decision support information and solution options based on the user's interaction behavior and focus in the visual interface. It analyzes the cost-benefit ratio of different solutions through decision tree algorithm to determine the optimal decision path and implementation priority ranking.

2. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The data acquisition and preprocessing module specifically includes: By deploying distributed collection nodes at key locations in the enterprise's IT architecture, multi-dimensional operating parameters such as server performance indicators, network traffic data, application running status, and database access frequency are obtained. A polling mechanism is used to periodically scan system components, determine status changes, and obtain component status datasets. The component status dataset is processed by time series analysis algorithm to determine the abnormal parameter set. Traffic peak characteristics are extracted from network traffic data to obtain the abnormal traffic dataset. The abnormal traffic dataset is obtained by analyzing the application running status and database access frequency. The decision tree algorithm is used to analyze the abnormal traffic dataset, determine the cause of application malfunction, obtain the abnormal cause set, and then the database access frequency is grouped by clustering algorithm to obtain the access pattern set.

3. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The dynamic state mapping module specifically includes: The collected real-time multi-dimensional operating parameters are stored in the data warehouse to obtain the system component status dataset. If the parameter values ​​in the system component status dataset exceed the preset threshold range, the status update mechanism is activated through an event trigger to determine the component status that needs to be updated. Based on the trigger signal of the state update mechanism, the state mapping algorithm is used to process the system component state dataset, generate a state snapshot of the physical architecture data, and convert the state snapshot of the physical architecture data into a digital architecture topology through the state mapping algorithm to obtain a digital topology description file. Visualization tools are used to render the digital topology description file, generating a dynamic architecture view that reflects the current system operation status. Based on the topology structure in the dynamic architecture view, abnormal patterns of system component states are detected, and potential system bottlenecks are identified. By analyzing potential system bottlenecks, adjusting the parameter weights of the state mapping algorithm, and optimizing the digital architecture topology, a dynamic architecture view reflecting the current actual operating status of the system is obtained.

4. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The system behavior analysis module specifically includes: State data is obtained from the dynamic architecture view. Noise data is removed through preprocessing to obtain a cleaned state dataset. Time series analysis is used to decompose the cleaned state dataset and extract performance change trends and resource usage patterns to obtain a trend feature set. If there are fluctuations in the trend feature set that deviate from the preset threshold, they are marked as abnormal fluctuation points by the anomaly detection algorithm to obtain an anomaly identifier set. Based on the anomaly identifier set and the trend feature set, a system behavior feature matrix is ​​constructed to obtain intermediate results of system behavior analysis. Cluster analysis is used to classify the system behavior feature matrix, identify potential risk points, and obtain a set of risk points. Association rule mining is used to analyze the relationship between the set of risk points and resource usage patterns, and obtain the association patterns between risk points and resource usage. Based on the association patterns, the intermediate results of system behavior analysis are optimized to obtain system behavior analysis results containing trend features and anomaly indicators.

5. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The predictive analysis module specifically includes: Based on the trend characteristic data in the system behavior analysis results, a linear regression model is used for training to generate a system load prediction model, which obtains the load prediction value within the future time window. Combined with the resource allocation records in historical data, a support vector machine model is used to calculate the resource demand prediction value and determine the future resource allocation demand. If the predicted resource demand exceeds the preset threshold, the potential performance bottlenecks are identified by using a decision tree model based on the performance bottleneck records in historical data, and the bottleneck prediction results are obtained. Combined with the pattern matching method, the operating status patterns in historical data are analyzed, the prediction confidence is calculated, and the confidence evaluation value is obtained. By combining the confidence level assessment value with the load forecast value and resource demand forecast value within the time window, the predicted value of the future system operating status is determined. By combining the dynamic changes in trend data, the parameters of the prediction model are adjusted to obtain the predicted value of the system operating status within the future time window.

6. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The resource optimization suggestion module specifically includes: Obtain predictive analysis results, analyze historical load data and current operating status through machine learning models to obtain load change trends, combine current architecture configuration information to calculate the matching degree between resource utilization and capacity threshold, and obtain resource status assessment; If the resource status assessment shows that the predicted load exceeds the capacity threshold, then the load balancing strategy generates expansion suggestions and determines the new resource configuration scheme. If the resource status assessment shows that the resource utilization is lower than the efficiency threshold, then the current architecture configuration information is analyzed, and the scaling-down suggestions are generated and the resource reduction scheme is determined. By combining expansion and reduction suggestions with load balancing strategies, an architecture adjustment plan is generated to obtain preliminary resource configuration optimization results. Based on the current architecture configuration information, the feasibility of the load balancing strategy is verified to obtain the final resource configuration suggestion. Then, a targeted architecture adjustment plan is generated based on the final resource configuration suggestion, and the optimized resource allocation plan is output.

7. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The visual interaction module specifically includes: The dynamic architecture data, predictive analysis results and optimization suggestions are obtained through a 3D rendering engine, and integrated to generate interactive visualization interface content. Color coding and graphic symbols are used to identify the status level and risk level in the visualization interface, and a visualization symbol mapping table is generated. If a user interaction is triggered, the interaction data is extracted according to the operation type, the status level and risk level corresponding to the interaction data are determined, the content of the visualization interface is updated, the user's focus is analyzed through the user interaction data, the dynamic architecture subset and predictive analysis results corresponding to the focus are obtained, and a detailed information dataset is generated. The support vector machine algorithm is used to classify the detailed information dataset, determine the priority of the focus data, obtain the classified focus content, and combine optimization suggestions to generate a dynamically adjusted 3D rendering view, update the interactive interface display, and determine the user's focus based on the updated interactive interface display and provide detailed information display.

8. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: The intelligent recommendation and decision support module specifically includes: By recording users' clicks, dwell time, and browsing paths in the visual interface, user interaction behavior data is obtained, user focus is determined, and user preferences are analyzed using clustering algorithms to obtain a user preference model. Based on preset business rules, decision support information and optimization solutions are generated, and recommended content is determined. If a user selects a certain optimization solution, the cost-benefit ratio of the solution is analyzed through a decision tree algorithm to obtain the evaluation results of each solution and determine the optimal decision path. An implementation plan is generated based on the optimal decision path. A heuristic algorithm is used to prioritize the implementation plans, and the resource allocation scheme is automatically adjusted based on the ranking results to generate the final execution plan.

9. The information technology consulting visualization platform based on digital twins according to claim 1, characterized in that: Also includes: The virtual verification and solution confirmation module uses the optimal decision path-driven dynamic simulation module to verify the proposed architecture adjustment solution in a virtual environment. The simulation engine simulates the system operation status and performance after the solution is implemented, and obtains the verified final consultation advice and implementation guidance solution.

10. The information technology consulting visualization platform based on digital twins according to claim 9, characterized in that: The virtual verification and scheme confirmation module specifically includes: The architecture adjustment plan is obtained from the business requirements database. The key parameters in the plan are extracted using a structured parsing method to obtain a standardized architecture adjustment description. The dynamic simulation module is then called to build the system model corresponding to the plan in a virtual environment to obtain the virtual system configuration. A simulation engine is used to simulate the virtual system configuration and run a preset workload scenario, obtain system running status data and performance data, and obtain a simulation result set. If the performance data in the simulation result set meets the preset performance improvement threshold, the feasibility of the solution is confirmed through the logic judgment module, and the feasibility verification result is obtained. Based on the feasibility verification results, the suggestion generation algorithm is invoked to extract key performance indicators from the simulation result set, generate structured consulting suggestions, obtain consulting suggestion documents, combine consulting suggestion documents and architecture adjustment descriptions, and use template matching methods to generate detailed implementation guidance schemes, thus obtaining the final implementation guidance scheme.

Images