Disclosure of Invention
The invention provides a historical data analysis platform which is used for solving the problems that a method for processing data through experience by a person is difficult to process increasingly huge data volume, and is difficult to consider all relevant factors from the whole, so that more reasonable processing decisions are difficult to give in time.
A historical big data analysis platform, comprising:
The data access module is used for acquiring historical data acquired by the data monitoring equipment and transmitting the historical data to the historical database;
the data report generation module is used for carrying out structural arrangement on the historical data in the historical database to generate a data report and carrying out data analysis on the data;
The data display module is used for carrying out 3D visual conversion on the data visual report to generate a high-dimensional visual model;
and the data result processing module is used for establishing a tree distribution model of the abnormal event solving strategy based on the high-dimensional visualization according to the high-dimensional visualization model.
As an embodiment of the present invention, the data access module:
a data distribution unit for determining different monitoring targets according to the data monitoring equipment and generating a historical data distribution diagram based on the monitoring targets,
The historical data distribution map is used for determining the generation source of the historical data and collecting the historical data;
the acquisition unit is used for determining the service corresponding to the historical data according to the historical data distribution diagram, and establishing acquisition channels based on different historical data according to the service corresponding to the historical data;
the database unit is used for establishing a history database of a corresponding type according to the acquisition channel,
The history database is composed of four visual layers, wherein the four visual layers comprise a data attribute layer, a data structure layer, a data wrapping layer and a data rendering layer;
the data attribute layer is used for storing attribute data of historical data;
the data structure layer is used for storing structure data of historical data;
the data wrapping layer is used for storing logic data of historical data;
the rendering layer is used for storing preset 3D visual rendering data.
As one embodiment of the invention, the data report generation module comprises:
The statistical report unit is used for carrying out structural analysis on the historical data types according to preset standards to generate a data catalog of the corresponding types,
The preset factors comprise a service type judgment standard, a service cause and effect differentiation standard, a service flow division standard, a service walk range standard, a service relevance judgment mark and a work order type identification standard;
the data statistics report forms carry out particle analysis according to a preset period;
the preset period comprises a daily period, a monthly period and a annual period;
The preset period includes
The custom template report unit is used for creating report templates with various different purposes, merging the data statistics report and automatically generating a report;
the data report analysis unit is used for calling various reports in the statistical report unit to analyze data, wherein,
The data analysis comprises business type analysis, business flow analysis, business cause and effect analysis, business spread analysis, business association analysis and business work order analysis.
As one embodiment of the invention, the data report analysis unit comprises:
The data attribute analysis subunit is used for determining the calling modes of the historical data with different attributes according to the historical database and carrying out service type analysis and service flow analysis according to the data attributes;
The data structure analysis subunit is used for determining the composition architecture and execution logic of the historical data of different structures and different logics according to the historical database and carrying out service causal analysis and service association analysis;
the data range analysis subunit is used for determining service types and execution logics of historical data with different attributes and different logics according to the historical database and carrying out service dispersion analysis;
The business work order analysis subunit is used for determining the completion degree evaluation data of different businesses according to the historical database and analyzing the business work orders according to the completion degree evaluation data;
the service work order analysis subunit further comprises the following steps:
s1, acquiring a service work order in historical data, evaluating the completion degree of the work order, and determining a problem work order and a problem-free work order;
S2, carrying out safety marking on the problem-free work order;
S2, the problem work orders are investigated to determine the service information of the problem work orders, wherein,
The service information comprises configuration information, provider information of products, technical description of the products and work order abnormality information;
s3, determining the abnormal reason of the problem work order according to the service information, and determining a solution strategy;
and S4, analyzing the solution strategy of the problem work order according to the solution strategy.
As an embodiment of the present invention, the data display module includes:
the chart conversion unit is used for converting the visual report into a 3D visual image through image segmentation, feature extraction, two-dimensional gradient operation and image format recombination;
And the 3D display unit is used for carrying out high-latitude superposition on the historical data according to the 3D visual image to generate a high-dimensional visual model.
As an embodiment of the present invention, the 3D display unit includes:
The visual image space arrangement subunit is used for building an infinite three-dimensional space through the position relation and the space arrangement of visual elements;
And the stereoscopic simulation camera subunit is used for importing the 3D visual image into an infinite three-dimensional space, establishing a high-dimensional symmetrical model based on a simulation camera in the infinite three-dimensional space, and forming a high-dimensional visual model based on depth simulation.
As an embodiment of the present invention, the data result processing module includes:
the workbench unit is used for displaying different historical data analysis results through the high-dimensional visual model and generating service simulation operation scenes corresponding to different historical data;
The abnormal event judging unit is used for judging whether an abnormal event occurs according to the service simulation operation scene and the service operation result in the historical data;
The strategy generating unit is used for extracting the solution strategy of the corresponding event in the historical data and optimizing the strategy when the abnormal event can occur;
And the tree distribution unit is used for carrying out parallel tree arrangement on the service behaviors corresponding to the solving strategies and the abnormal events.
As an embodiment of the invention the system further comprises:
the data particle model construction module is used for dividing grid particles according to the historical data through a WRF mode and determining particle distribution information;
the attribute assignment module is used for carrying out attribute assignment on the data examples of each historical data according to the example distribution information;
the particle association module is used for determining the association between different examples according to the attribute assignment, and carrying out particle connection according to the association to generate a particle model;
And the verification module is used for comprehensively verifying the data of the high-dimensional visual model according to the particle model and outputting a verification result.
As an embodiment of the invention the system further comprises:
the data acquisition module is used for carrying out data analysis on the real-time data acquired by the data monitoring equipment and generating a visual report based on real-time service;
the data calling module is used for calling the same kind of service data as the real-time service from the tree-shaped distribution model according to the visual report;
the data comparison module is used for comparing the similar service data with the real-time data and judging whether data abnormality exists or not;
and the anomaly discovery module is used for calling the anomaly data when the data is anomalous, and determining a corresponding anomaly event solving strategy through the tree distribution model.
As an embodiment of the present invention, the abnormality discovery module includes:
the visual information judging unit is used for determining the data attribute and the data dimension of the abnormal data according to the abnormal data;
The data identification unit is used for determining the distribution position of the abnormal event in the tree-shaped distribution model according to the data dimension and the data tree shape;
And the rule comparison module is used for determining comparison rules of the historical data corresponding to the distribution positions according to the distribution positions, carrying out rule judgment on the abnormal data according to the comparison rules, and acquiring corresponding solution strategies after all the comparison rules are met.
The historical data analysis platform has the beneficial effects that the historical data analysis platform is a data value discovery and utilization platform, and provides a professional, quick and easy-to-use tool for large data analysis, mining and visual display for clients. The platform aims at data increment, provides specialized historical data processing and analyzing methods for clients, and meets the requirements of organizing data value mining and application of different roles. The method is oriented to personnel for data analysis and data value utilization of enterprises, and integrates data visual search, data depth analysis and model application development. The method can realize access and processing of multiple data sources, realize multiple capabilities of data access, data processing, data analysis, result application and abnormal event processing strategies, and enable a user to intuitively analyze in a data visualization mode and also to mine depth rules implicit in data through data mining. Can be used together for enterprise leadership, business personnel and technicians at all levels. The product not only saves cost for customers and improves efficiency, but also creates value for customers.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Example 1:
the embodiment of the invention provides a historical big data analysis platform, as shown in fig. 1, comprising:
The data access module is used for acquiring historical data acquired by the data monitoring equipment and transmitting the historical data to the historical database;
the data report generation module is used for carrying out structural arrangement on the historical data in the historical database to generate a data report and carrying out data analysis on the data;
The data display module is used for carrying out 3D visual conversion on the data visual report to generate a high-dimensional visual model;
and the data result processing module is used for establishing a tree distribution model of the abnormal event solving strategy based on the high-dimensional visualization according to the high-dimensional visualization model.
The working principle of the technical scheme is that in the prior art, on the monitoring data processing of a data center, the historical data is limited to be utilized or is not directly deleted, most of the historical data is simply summarized or stored as standby information, but with the development of enterprises, the same business and the same situation as the historical data are likely to be encountered, and the historical data are the best reference data at the moment. Therefore, the invention collects the historical data according to the data monitoring equipment, because all business data are collected through the data monitoring equipment, the historical data are collected through the data monitoring equipment, a special historical database is established for storage, the purpose of generating the data report is to comprehensively analyze the historical data, and when the existing events are the same as the corresponding events in the historical data, all the comparisons are based on the data report, and the comparison of statistical analysis contents of different data. The 3D visual conversion is a virtual state conversion, for example, a data report table shows fluctuation states of data flow, the data report table can be data flow details and flow states at all times, after the 3D visual conversion, the flow states and the fluctuation states are converted into the same data fluctuation as the water flow in a pipeline, the water flow fluctuates, and the data state is the water flow state. Finally, the method is converted into a high-dimensional visual model, which is a processing strategy based on a high-dimensional visual technology, namely, a high-dimensional visual graph is generated on the basis of 3D conversion, and based on the graph, abnormal events are more clearly analyzed and predicted.
The technical scheme has the beneficial effects that the historical data analysis platform is a data value discovery and utilization platform, and provides a professional, quick and easy-to-use tool for large data analysis, mining and visual display for clients. The platform aims at data increment, provides specialized historical data processing and analyzing methods for clients, and meets the requirements of organizing data value mining and application of different roles. The method is oriented to personnel for data analysis and data value utilization of enterprises, and integrates data visual search, data depth analysis and model application development. The method can realize access and processing of multiple data sources, realize multiple capabilities of data access, data processing, data analysis, result application and abnormal event processing strategies, and enable a user to intuitively analyze in a data visualization mode and also to mine depth rules implicit in data through data mining. Can be used together for enterprise leadership, business personnel and technicians at all levels. The product not only saves cost for customers and improves efficiency, but also creates value for customers.
Example 2:
in a specific embodiment, as shown in fig. 2, the data access module includes:
a data distribution unit for determining different monitoring targets according to the data monitoring equipment and generating a historical data distribution diagram based on the monitoring targets,
The historical data distribution map is used for determining the generation source of the historical data and collecting the historical data;
the acquisition unit is used for determining the service corresponding to the historical data according to the historical data distribution diagram, and establishing acquisition channels based on different historical data according to the service corresponding to the historical data;
the database unit is used for establishing a history database of a corresponding type according to the acquisition channel,
The history database is composed of four visual layers, wherein the four visual layers comprise a data attribute layer, a data structure layer, a data wrapping layer and a data rendering layer;
the data attribute layer is used for storing attribute data of historical data;
the data structure layer is used for storing structure data of historical data;
the data wrapping layer is used for storing logic data of historical data;
the rendering layer is used for storing preset 3D visual rendering data.
The technical scheme has the working principle that because the method is mainly used for processing the data collected by the monitoring equipment of the book data center, the method is used for determining the type of the monitored data based on the monitoring purpose and establishing a historical data distribution map based on the type of the data and the distribution area of the data, so that when the method is compared with the existing abnormal data, the abnormal event can be found out more quickly. Data acquisition is carried out through different acquisition channels during acquisition, so that the system is more convenient and more targeted. The data is more comprehensive and accurate. The database is built according to the acquired acquisition channels and the data types, so that different kinds of data can be stored conveniently. Finally, the invention needs to perform 3D visual conversion and high-dimensional visual conversion, so that different visual layers store different data corresponding to visual technology, and data retrieval is convenient, thereby more rapidly establishing 3D, high-dimensional and data charts.
The technical scheme has the beneficial effects that huge data can be arranged through the database and displayed in a report form, so that the data is easier to express and understand, and an intuitive and clear data source is provided for later data result analysis.
Example 3:
in a specific embodiment, the statistical reporting unit includes:
the statistical report unit is used for carrying out structural analysis on the historical data types according to preset standards to generate a data chart of a corresponding type,
The preset standards comprise data type judgment standards, data causal distinction standards, data flow division standards, data range standards, data relevance judgment labels and worksheet type recognition standards;
The custom template report unit is used for creating report templates with various different purposes, merging the report templates into the data report and automatically generating a report;
the data report analysis unit is used for calling various reports in the statistical report unit to analyze data, wherein,
The data analysis comprises business type analysis, business flow analysis, business cause and effect analysis, business spread analysis, business association analysis and business work order analysis.
The technical scheme has the working principle that when the data report is generated, structural analysis is carried out according to the type of the historical data, professional analysis can be carried out on different types of data, and a special analysis technology is adopted. The invention generates data report according to the data standard, the data type judging standard, the data causal distinguishing standard, the data flow dividing standard, the data range standard, the data relativity judging mark and the work order type identifying standard, which correspondingly generates the type chart, the causal chart of data abnormity or data development, the data flow dividing rule icon, the data monitoring range icon, the icons of data and equipment related to or related to the data, and the history function corresponding to each data, wherein each data monitoring target is a work order in technical real-time.
The technical scheme has the beneficial effects that the data report of different types is generated through the mode, the report is automatically generated, and the data can be specifically analyzed through the report.
Example 4:
in one particular embodiment, as shown in figure 3,
The data report analysis unit comprises:
The data attribute analysis subunit is used for determining the calling modes of the historical data with different attributes according to the historical database and carrying out service type analysis and service flow analysis according to the data attributes;
The data structure analysis subunit is used for determining the composition architecture and execution logic of the historical data of different structures and different logics according to the historical database and carrying out service causal analysis and service association analysis;
the data range analysis subunit is used for determining service types and execution logics of historical data with different attributes and different logics according to the historical database and carrying out service dispersion analysis;
The business work order analysis subunit is used for determining the completion degree evaluation data of different businesses according to the historical database and analyzing the business work orders according to the completion degree evaluation data;
as shown in fig. 3, the service work order analysis subunit further includes the following steps:
s1, acquiring a service work order in historical data, evaluating the completion degree of the work order, and determining a problem work order and a problem-free work order;
S2, carrying out safety marking on the problem-free work order;
S2, the problem work orders are investigated to determine the service information of the problem work orders, wherein,
The service information comprises configuration information, provider information of products, technical description of the products and work order abnormality information;
s3, determining the abnormal reason of the problem work order according to the service information, and determining a solution strategy;
and S4, analyzing the solution strategy of the problem work order according to the solution strategy.
The working principle of the technical scheme is that when data analysis is carried out, historical data is used as corresponding business to be analyzed according to the tree shape, the structure and the logic of the data, different worksheets are used for tracking and processing different types of matters, how to use the worksheets is determined according to specific requirements, and one worksheet can be a work task of a project.
The goal of worksheet analysis is to minimize business failures due to infrastructure failures and to prevent the reoccurrence of events associated with these errors.
Thus, the focus of the present invention is also to find the root cause of the event occurrence and then take action to improve or correct this situation. Events are potential causes of some or more accidents, and the problem of analysis of historical data is to minimize the influence of defects or mistakes in service infrastructure, human errors, external events, etc. on clients and prevent them from repeatedly occurring.
When in specific real time:
When the work order is evaluated, the submitted work order needs to be evaluated, one is that the problem really exists and needs to be solved, an operation and maintenance person accepts the work order and clicks a hyperlink of 'begin investigation', the corresponding work order state becomes 'in investigation', and the other is that the work order description which does not need to be solved or submitted is not clear, investigation is refused, and the corresponding work order state becomes 'refused'. The "start investigation" is an action, and the status of the work order after execution correspondingly changes into "in investigation", which means that the cause of the problem is currently being investigated and diagnosed, and the investigation and diagnosis may be a repeated process, which needs to be repeated multiple times, and the repeated times are all closer to the solution that we want. Executing the start survey does not require entering a remark, but rather an action to transition the work order status from "in-evaluation" to "in-survey" to indicate that the current operation and maintenance personnel has been working to deal with the problem. Where the system will be in linkage with the configuration management and infrastructure components. Configuration information, vendor information for the product, technical specifications and error information for the product, etc. may be associated, with node run status information, such as availability reports, performance reports, etc., that creates problems. Providing a basis for staff to analyze the reasons of the problems.
The investigation may be ended when the problem analyst finds the cause of the problem and finds a temporary or permanent solution to the problem, in the operation and maintenance specifications, the problem is then converted into a "known error" state, the investigated cause of the problem must be entered when the investigation is ended, and the work order state becomes a "known error" after the action is performed.
The technical scheme has the beneficial effects that the working efficiency of data center personnel is improved through accurate analysis of data, the data report analysis unit is realized to promote enterprises to reduce operation cost and effectively promote customer retention, and the enterprise management takes up important roles. Can play roles in front-and-back communication, cross-department collaboration and cross-enterprise collaboration.
Example 5:
The data display module includes:
the chart conversion unit is used for converting the visual report into a 3D visual image through image segmentation, feature extraction, two-dimensional gradient operation and image format recombination;
And the 3D display unit is used for carrying out high-latitude superposition on the historical data according to the 3D visual image to generate a high-dimensional visual model.
The technical scheme has the working principle that a chart conversion unit converts chart data into a 3D visual image through two-dimensional gradient operation and format recombination, and then a high-dimensional visual model is generated through high-latitude superposition, wherein the high-dimensional visual model is a technology for converting complex data superposed by the 3D visual image into a scattered point matrix and reducing the image complexity of space coordinates, and the technical model for blurring or directly filtering irrelevant data comprehensively recorded in some chart data and not displaying the irrelevant data as main characteristics of the data exists. The 3D visual image is displayed through a naked eye 3D display screen, the high-dimensional visual model is a space model, is a model for comprehensively analyzing historical data, and can also realize diagrammatizing.
In the process of converting chart data into a 3D visual image through two-dimensional gradient operation and format recombination, the method comprises the following steps:
establishing a data set of chart data:
X＝{x1,x2,…,xi};Y＝{y1,y2,…,yi};Z＝{z1,z2,…,zi}
X and Y are sample data of chart data respectively, for example, table data X represents row data, Y represents column data, Z represents result data, X represents abscissa if a graph is obtained, Y represents ordinate, Z represents vertical coordinate, i is a positive integer, and n data are obtained in total;
After the data set determination, we perform 3D conversion by:
Above-mentioned
F (X, Y, Z) represents the 3D conversion function;
According to the 3D conversion function, a space architecture is established, and a 3D visual image can be formed by adding logic data and 3D visual rendering data.
The technical scheme has the beneficial effects that the three-dimensional data is converted into the three-dimensional data through the chart data in the technology, the three-dimensional data is converted into the high-dimensional visual model, the management functions of the data center, the cabinet and various devices based on the three-dimensional environment are realized, and the visual platform of the data center environment, the devices and the management information is constructed.
Example 6:
As a specific embodiment of the present invention, the 3D display unit includes:
The visual image space arrangement subunit is used for building an infinite three-dimensional space through the position relation and the space arrangement of visual elements;
And the stereoscopic simulation camera subunit is used for importing the 3D visual image into an infinite three-dimensional space, establishing a high-dimensional symmetrical model based on a simulation camera in the infinite three-dimensional space, and forming a high-dimensional visual model based on depth simulation.
The working principle of the technical scheme is that the method comprises the steps of establishing an infinite three-dimensional space, establishing coordinate systems of different positions in the space, guiding a 3D visual image into the infinite three-dimensional space, and establishing a high-dimensional symmetrical space model, wherein the space model fits dimensions of the 3D visual image when the 3D visual image is converted by a simulation camera, fitting same data into one block in the fitting process, and then generating a space abstract image, such as a scattered point matrix space, because dot data are left after the same data are counteracted in the fitting process, and a scattered point state is presented.
The technical scheme has the beneficial effects that the chart data are converted into the 3D image for display by building an infinite three-dimensional space, so that the data are easier to manage after being converted into a high-dimensional visual model, the transparency of management is realized more obviously for the abstract state of the data, namely the abnormal state, and the efficiency of asset management and monitoring management is further effectively improved, so that a three-dimensional and visual new generation historical data analysis platform is truly realized.
Example 7:
the data result processing module comprises:
the workbench unit is used for displaying different historical data analysis results through the high-dimensional visual model and generating service simulation operation scenes corresponding to different historical data;
The abnormal event judging unit is used for judging whether an abnormal event occurs according to the service simulation operation scene and the service operation result in the historical data;
The strategy generating unit is used for extracting the solution strategy of the corresponding event in the historical data and optimizing the strategy when the abnormal event can occur;
And the tree distribution unit is used for carrying out parallel tree arrangement on the service behaviors corresponding to the solving strategies and the abnormal events.
The working principle of the technical scheme is that the workbench is used for simulating operation scenes of businesses corresponding to different historical data, for example, the flow monitoring equipment monitors detection indexes of the flow monitoring equipment and a real-time detected flow chart. When judging the abnormal event, according to the method, whether the abnormal event occurs in the historical data is judged firstly, so that whether the abnormality occurs or not can be more vividly displayed according to the simulation of a scene, and then the corresponding solution policy is called in the historical data, policy optimization is realized, and finally, the abnormal event of the service and the corresponding service behavior are arranged in a tree arrangement mode, namely, the abnormal condition of the monitored data center is monitored.
The technical scheme has the advantages that when 3D display can be performed, data from different systems can be displayed on one interface, the most commonly used functions and the most focused service data of a personal user are integrated uniformly, the running states of all subsystems can be seen clearly at a glance, and the running states are collected as shortcuts of daily work of the user. Through the workbench unit, a user can quickly locate the work required to be performed, the efficiency of system application is improved, the daily maintenance support work is facilitated, but the data can be in abnormal conditions and solving strategies, for the most required data, events and strategies are correspondingly arranged in an artistic arrangement mode, and the event can be conveniently and timely found out when the event is actually implemented, and the invention can also alarm based on the situation of absorption.
The device has the advantages that the device has the functions of single-body device alarming and in-cabinet device alarming, the alarming icon and the device color changing prompt are directly displayed on the device, the alarming icon is clicked or the detailed alarming information is displayed by the alarming device, the in-cabinet device alarming is used for displaying the detailed alarming information by clicking the alarming icon and the cabinet color changing prompt on the belonging cabinet, in the alarming state, the cabinet is entered by double clicking, the alarming device color changing prompt is displayed by the alarming device, and the detailed alarming information interface is displayed by clicking the device. The equipment alarm display machine room equipment alarm information display requires that all alarm information of the machine room be displayed on a machine room 3D display interface in real time. The type of the alarm equipment is mainly shown as two types, namely, the alarm of the single equipment is directly displayed on the equipment, an alarm icon and an equipment color change prompt are displayed, and the alarm icon is clicked or the detailed alarm information is shown by the alarm equipment. The equipment in the cabinet gives an alarm, an alarm icon and a cabinet color change prompt appear on the belonging cabinet, the alarm icon is clicked or the alarm equipment displays detailed alarm information, in the alarm state, the equipment color change prompt appears on the alarm equipment, the equipment is clicked, and a detailed alarm information interface appears. The device with the alarm function is characterized in that the alarm function is set for the single-body device and the alarm function is set for the device in the cabinet, so that a worker can find out the failed device faster, the worker can repair the device in time conveniently, and the platform can enter a normal running state faster.
Judging whether an abnormal event occurs or not according to service simulation operation scenes and service operation results in historical data in actual real time;
in the process, firstly, a real-time data model between operation scenes is simulated based on a service;
H represents a coefficient of data change in a service simulation running scene, wherein the coefficient changes along with an event, Qj,t represents a data state of a j-th device in monitoring equipment at a moment T, Rj,t represents a data characteristic of data monitored by the j-th device in the monitoring equipment at the moment T, Wj,t represents a coefficient of change of the j-th device in the monitoring equipment at an initial moment T, T0 represents the initial moment, T represents a final moment of monitoring, T represents real time when the moment is represented, j is a positive integer, j is epsilon m, and m represents the total number of the monitoring equipment;
establishing a history model of similar conditions in the history data based on the history data;
wherein Dg represents the event type characteristic corresponding to the G-th historical data, Ng represents the event content characteristic corresponding to the G-th historical data, represents the event result characteristic corresponding to the event type corresponding to the G-th historical data, V represents an abnormal event, G is a positive integer, G is E G, and G represents the total number of the historical data;
then, judging whether the abnormal event exists or not through mutual comparison of the two models;
Wherein, the data content of the historical event and the real-time simulation operation scene are the same when P=0, and therefore, the event is abnormal. In the process, a real-time data model diagram is built aiming at a scene of simulated operation, and then all abnormal events in historical data are built into a model, and whether the real-time abnormal event is an abnormal event or not is judged in a one-by-one fitting mode after the real-time abnormal event is determined to belong to the abnormal event in the historical data.
Example 8:
as a specific embodiment, the system further comprises:
the data particle model construction module is used for dividing grid particles according to the historical data through a WRF mode and determining particle distribution information;
the attribute assignment module is used for carrying out attribute assignment on the data examples of each historical data according to the example distribution information;
the particle association module is used for determining the association between different examples according to the attribute assignment, and carrying out particle connection according to the association to generate a particle model;
And the verification module is used for comprehensively verifying the data of the high-dimensional visual model according to the particle model and outputting a verification result.
The principle of the technical scheme is that the obtained model is verified through the gridding particle division, so that the obtained model is ensured to have high accuracy. In the early course, WRF mode represents a prediction mode, and for particles showing anomalies in data particles, prediction is to be carried out, but after the particles are connected into a particle model, the particles should have a similar spatial shape with a high-dimensional visual model. However, since the granulation is only used for clearly dividing the data, the monitoring of the abnormal data by the high-dimensional particle model is not more accurate.
The technical scheme has the beneficial effects that the data of the high-dimensional visual model can be verified comprehensively through the particle call of the historical data, so that the visual data in the technology is ensured to be comprehensive, and the method can meet various conditions.
Example 9:
In a specific embodiment, the system further comprises:
the data acquisition module is used for carrying out data analysis on the real-time data acquired by the data monitoring equipment and generating a visual report based on real-time service;
the data calling module is used for calling the same kind of service data as the real-time service from the tree-shaped distribution model according to the visual report;
the data comparison module is used for comparing the similar service data with the real-time data and judging whether data abnormality exists or not;
and the anomaly discovery module is used for calling the anomaly data when the data is anomalous, and determining a corresponding anomaly event solving strategy through the tree distribution model.
The technical scheme has the working principle and beneficial effects that the invention further comprises a data acquisition module, the acquired data are monitored in real time, the most important purpose of the invention is to take historical data as an abnormal event strategy of the existing data to be provided with a library, and a rapid solution for abnormality occurrence during the existing actual monitoring is realized. Therefore, in the process, the invention also generates the visualized report form of the monitoring data of the real-time service, and because the tree distribution model is established based on the high-dimensional visualized model and has the basis of the report form of the visualized data, the comparison of the existing data and the historical data is easier to carry out through the tree distribution model, and whether the data abnormality exists is judged, so that the corresponding solving strategy is found.
Example 10:
the anomaly discovery module includes:
the visual information judging unit is used for determining the data attribute and the data dimension of the abnormal data according to the abnormal data;
The data identification unit is used for determining the distribution position of the abnormal event in the tree-shaped distribution model according to the data dimension and the data tree shape;
And the rule comparison module is used for determining comparison rules of the historical data corresponding to the distribution positions according to the distribution positions, carrying out rule judgment on the abnormal data according to the comparison rules, and acquiring corresponding solution strategies after all the comparison rules are met.
The principle of the technical scheme is that in the processing of the abnormal data, according to the data attribute and the dimension of the data, namely, what type of data the data belongs to and what abnormality the abnormal data shows, the invention can determine the corresponding abnormal data in the tree distribution model and the corresponding solving strategy when the abnormal data belong to. However, before the corresponding strategy is extracted, the comparison rule is mainly compared according to the historical data, and when the comparison rule is compared, the comparison rule can be set according to the simulated scene because of the 3D simulated image scene converted by the historical data, and then the comparison rule is compared with the actual scene, so that the strategy is obtained.
The technical scheme has the beneficial effects that through the arrangement of the comparison rules, the actual situation can be more attached to the situation corresponding to the historical data, and the more accurate abnormal event solving strategy corresponding to the abnormal data can be obtained.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.