Smart park enterprise service integrated management method and system

Abstract

The invention provides a smart park enterprise service integrated management method and system, and the method comprises the steps: obtaining historical enterprise service information in advance, obtaining a call chain of the historical enterprise service information according to a distributed tracking algorithm, and generating a Gantt chart containing each call node according to the call chain of the enterprise service information; key word features of each calling node in the Gantt chart are obtained, calling of each calling node is classified according to the key word features, corresponding calling classes are obtained, and probability distribution between the classified calling classes is calculated according to a Monte Carlo algorithm; according to the probability distribution between the calling classes, calculating correlation degree values between all the calling classes of the corresponding enterprise service calling chain each time, and according to the correlation degree values, attributing the corresponding calling classes to the same class combination name; and verifying the enterprise service output result according to the association degree value of the calling class corresponding to the calling chain, and performing an operation including enterprise service name recombination according to the verification result.

Description

Technical Field

[0001] The present invention relates to enterprise service management technology, and in particular to a smart park enterprise service integrated management method and system Background Art

[0002] At present, there are many types of enterprise services in the park, including but not limited to policy services, financial services, talent services, supply chain services, emergency management services, security services, logistics services and environmental protection services. There may be multiple sub-services under the above different types of service categories, and there are certain similarities between the sub-service types. For example, taking the above-mentioned talent services as an example: specific talent services may involve talent certification, talent living subsidies, talent rental subsidies and talent housing subsidies, etc. Other related enterprise services may also contain talent service-related content, such as policy services will involve talent service content. The requirements of different sub-categories of services will have certain differences, and there will be certain differences in the corresponding service process templates and the calls between services. The existing enterprise service management method is generally still classified management. For example, all categories of sub-service types corresponding to talent services are classified into the talent services. However, the classification method of the above-mentioned service types in the existing technology still relies on manual processing, and the enterprise service classification is completed based on human experience, and the construction of enterprise service directory links is completed. The above-mentioned existing technologies will have major problems if they rely solely on manual experience to complete the classification of enterprise services. On the one hand, the management of a large and complex enterprise service directory requires a large amount of initial human cost investment and construction. On the other hand, the classification method and construction authentication method for new categories of enterprise services also need to be judged based on human experience. When the enterprise service directory is large and complex, relying on manual experience is often prone to errors, resulting in confusion in the enterprise service links displayed to the outside world, thereby reducing the enterprise service experience. Summary of the Invention

[0003] One of the inventive purposes of the present invention is to provide a method and system for integrated management of enterprise services in a smart park. The method and system use distributed tracing to obtain the historical enterprise service call links of each category or subcategory of enterprise services, and analyze the node characteristics of the historical call links. The node characteristics are used as characteristic parameters for the classification and construction of the enterprise services for statistical analysis to determine whether the classification and call of the enterprise services of the current category or subcategory meet the rationality of category attribution. If not, the enterprise services of the corresponding category will be reclassified to obtain a new enterprise service catalog, or the enterprise service call method will be regenerated.

[0004] Another object of the present invention is to provide a method and system for integrated management of enterprise services in a smart park. The method and system use a Monte Carlo algorithm to perform a probability analysis on the historical enterprise service call links of each category or subcategory of enterprise services to obtain the probability distribution of call target links of different categories or subcategories. The probability distribution is used to assign the corresponding category or subcategory services with the highest probability values ​​of the same or similar call links to the same type of enterprise services, thereby greatly improving the centralized configuration of the same type of enterprise services in the same or similar call links, which is conducive to the generation of a reasonable and centralized enterprise service catalog for external display.

[0005] Another object of the present invention is to provide a method and system for integrated management of enterprise services in a smart park, wherein the method and system utilize multi-layer probability distribution calculations including call field content to determine the affiliation level of corresponding service categories or subcategories; and in the present invention, enterprise service categories with the same call link but different affiliation enterprise service names are verified, reorganized or deleted according to their actual output results, thereby reducing the duplication of enterprise service categories in the directory and improving the simplicity of the enterprise service management.

[0006] Another object of the present invention is to provide a method and system for integrated management of enterprise services in a smart park. The method and system use a call chain Gantt chart to obtain the cross-service scheduling (span) of each enterprise service, wherein the cross-service scheduling (span) of the enterprise service records the call operation trajectory of each enterprise service, including the start time, duration, context relationship and key details of the call operation. The cross-service scheduling (span) of the enterprise service can be used to effectively analyze the core content of the call of the current enterprise service, and judge the similarity of the call chain based on the core content of the call, and classify the call chains with similarity higher than a certain threshold into the same type, thereby realizing integrated classification processing of the same enterprise service.

[0007] In order to achieve at least one of the above-mentioned objects, the present invention further provides a method for integrated management of enterprise services in a smart park, the method comprising: Pre-acquire historical enterprise service information, obtain the call chain of the historical enterprise service information according to a distributed tracing algorithm, and generate a Gantt chart containing each call node according to the call chain of the enterprise service information; Obtain keyword features of each call node in the Gantt chart, classify each call of the call node according to the keyword features to obtain a corresponding call class, and calculate the probability distribution between each call class according to the Monte Carlo algorithm; Calculating the correlation value between all call classes of each corresponding enterprise service call chain according to the probability distribution between each call class, and classifying the corresponding call classes into the same category combination name according to the correlation value; Each enterprise service class name is obtained, and the enterprise service output result is verified according to the call class corresponding to the call chain, and operations including enterprise service class name reorganization are performed according to the verification result.

[0008] According to one of the preferred embodiments of the present invention, the keyword feature classification method includes: obtaining the input keywords of the call, the call node name keywords and the corresponding call node output keywords according to the Gantt chart generated by the call chain to obtain keyword triplets, performing keyword feature classification according to the keyword triplets, calculating the similarity value between each keyword in the keyword triplets and the corresponding keywords of other keyword triplets, clustering all keyword triplets using a clustering algorithm to obtain n clustered call classes, and calculating the probability distribution of each call class and its adjacent call class, and selecting the call class with the highest probability distribution to group the enterprise service catalog.

[0009] According to another preferred embodiment of the present invention, the keyword triple clustering method includes: pre-generating n centroids according to the enterprise service call class, wherein the corresponding standard keyword triples are pre-assigned to the n centroids; after obtaining the Gantt chart of the corresponding call chain, performing similarity calculation on the keywords in each keyword triple in the Gantt chart and the keywords in the keyword triple of each centroid, wherein the similarity calculation adopts cosine similarity calculation, and selecting the centroid of the keyword triple with the highest total similarity value and satisfying the variance threshold rule as the target centroid for classification, after performing feature classification on the keyword triplets related to all call classes in the Gantt chart, n call classes are obtained, and the probability distribution of the n call classes and adjacent call classes is calculated according to the n call classes to calculate the correlation value between different call classes.

[0010] According to another preferred embodiment of the present invention, the probability distribution calculation method of n call classes and adjacent call classes includes: obtaining the globally unique identifier Trace ID of the corresponding enterprise service information according to the distributed tracing algorithm, defining a first call class and a second call class, wherein the first call class is the parent call of the second call class; obtaining the unique identifier Span ID / Parant Span ID of the first call class and the second call class according to the Gantt chart, and obtaining the operation description Opertion Name and key-value pair tags Tags corresponding to the first call class and the second call class according to the unique identifier, obtaining the keyword triples corresponding to the first call class and the second call class according to the operation description Opertion Name and the key-value pair tags Tags, and judging the category combination to which the two keyword triplets belong according to the clustering algorithm, wherein the category combination includes the same category combination and different category combinations, and counting the hierarchical relationship of each category combination.

[0011] According to another preferred embodiment of the present invention, the number of each category combination is obtained based on the historical enterprise service information, and the hierarchical relationship of each category combination corresponding to the calling class in the entire enterprise service information is recorded; the probability distribution of each category combination in all category combinations in the historical enterprise service information and the enterprise service information directory are centrally constructed according to the high and low probability distribution of each category combination; a first category combination probability threshold is set, and when the probability threshold of the corresponding category combination in all levels is greater than the first category combination probability threshold, the corresponding category combination is classified into the same parent category enterprise service information.

[0012] According to another preferred embodiment of the present invention, the number of each category combination is obtained based on the historical enterprise service information, and the hierarchical relationship of the call class corresponding to each category combination in the entire enterprise service information is recorded; the number of levels corresponding to each category combination is calculated, and after obtaining the number of all category combinations of the corresponding level, the probability distribution of each category combination at different levels is calculated, and the level with the highest probability distribution is selected as the construction related to the enterprise service information directory of the corresponding category combination, which is used to analyze the hierarchical distribution of the corresponding call class in different enterprise service information, and the call class includes the call of the parent level node to the child level node.

[0013] According to another preferred embodiment of the present invention, the node name of the first call class itself as the parent node is obtained, and the second call class with the same parent node as the first call class in all call classes is obtained based on the parent node name, and the clustering category to which all keyword triplets in the second call class belong is calculated, and the enterprise service information directory of the sub-node call class of which the first call class itself is the parent node is configured according to the clustering category, and the probability distribution of the category combination between the first call class and the second call class at the same level is judged through the first call class, and the second call class with a higher probability value is given priority as the enterprise service information to be displayed.

[0014] According to another preferred embodiment of the present invention, the method for verifying the output result includes: using the OpenTelemetry SDK to build an API for calling the service to analyze the enterprise service information call chain, using the Tracer component and Context Propagation component of the OpenTelemetry SDK to create Span and track different service calls, obtaining a call chain including the same Span ID / Parant Span ID and the same keyword triplet, and determining whether the enterprise service names corresponding to the call chains with the same Span ID / Parant Span ID and the same keyword triplet for each call class are the same; if different, reorganizing the enterprise service names and deleting the duplicate enterprise service names in the call chain.

[0015] In order to achieve at least one of the above-mentioned invention purposes, the present invention further provides a smart park enterprise service integrated management system, which executes the above-mentioned smart park enterprise service integrated management method.

[0016] The present invention further provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned smart park enterprise service integrated management method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Shown is a flow chart of a smart park enterprise service integrated management method of the present invention.

[0018] Figure 2 Shown is a structural diagram of the enterprise service call chain Gantt chart in the present invention. DETAILED DESCRIPTION

[0019] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are for illustrative purposes only, and those skilled in the art will readily appreciate other obvious variations. The basic principles of the present invention defined in the following description may be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the present invention.

[0020] It is understandable that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0021] Please combine Figure 1-Figure 2The present invention discloses a method and system for integrated management of enterprise services in a smart park, the method comprising the following steps: S01. Pre-acquire historical enterprise service information, obtain the call chain of the historical enterprise service information according to the distributed tracking algorithm, and generate a Gantt chart containing each call node according to the call chain of the enterprise service information; S02. Obtain keyword features for each call node in the Gantt chart, and classify each call of the call node according to the keyword features to obtain a corresponding call class, and calculate the probability distribution between each call class according to the Monte Carlo algorithm; S03. Calculate the correlation value between all call classes of each corresponding enterprise service call chain according to the probability distribution between each call class, and classify the corresponding call class into the same category combination name according to the correlation value; S04. Obtain each enterprise service name, and verify the enterprise service output result according to the call class corresponding to the call chain, and perform operations including enterprise service name reorganization according to the verification result.

[0022] Specifically, the enterprise service information described in the present invention includes but is not limited to enterprise service types and call service types, which specifically include but are not limited to talent services, supply chain services, emergency management services, security services, logistics services and environmental protection services, etc. The above-mentioned different enterprise services are integratedly configured in the smart park, and the API interface of each enterprise service can be integrated and configured. The API interface monitors the calls of different service types to achieve integrated analysis of enterprise service information. It should be noted that since enterprise service calls involve calls to different functional application layers, and different application layer services are generally distributed and deployed on different computing nodes, the call links may also be on different computing nodes. Therefore, the present invention needs to use a distributed tracing algorithm (Distributed Tracing) to perform analysis of the corresponding point call links to obtain the core call information in the enterprise service information.

[0023] Furthermore, the core technical solution of the present invention needs to solve the problem of centralized and highly adaptable management of different enterprise service types, including enterprise service types. Therefore, the present invention needs to use the historical enterprise service information of enterprises in the smart park, and search for corresponding call chain data from the historical enterprise service information, and perform feature analysis on the call chain data, and further rationally reconstruct and display the call chain data in the enterprise service information, including the enterprise service information directory. In this way, the same or similar types of enterprise service types or subtypes will be incorporated into the same parent node, or high-frequency call enterprise service types will be given priority in the enterprise service information directory, thereby greatly improving the rationality and efficiency of the enterprise services.

[0024] It is worth mentioning that the present invention uses the distributed tracing algorithm to effectively trace the complete business requests of the enterprise, and uses the distributed tracing algorithm to obtain a Gantt chart of the completed enterprise service, wherein the Gantt chart effectively traces the cross-service scheduling (Span) contained in the enterprise service through the request information in the historical enterprise service information. The specific method includes: pre-configuring the OpenTelemetry SDK in the API interface of the enterprise service, using the OpenTelemetry SDK to obtain the historical log data of the enterprise service request, and generating a globally unique identifier (TraceID) for each enterprise service request through the historical log data. It should be noted that the globally unique identifier (Trace ID) is used to record the call chain data of the enterprise service request throughout its life cycle. Furthermore, the present invention utilizes the OpenTelemetry SDK to obtain cross-service scheduling (Span) in the call chain data. It should be noted that the cross-service scheduling (Span) described in the present invention is an operation unit, which includes but is not limited to a function call, a data database query, and other operation units in the call chain data. Through the cross-service scheduling (Span), the call information details related to each call chain can be accurately tracked, where the relevant call information details include the call node, call method, and call content of the call chain. Because the same call node may be called differently by other call nodes, for example, the call protocol method may be an HTTP data header call, an RPC metadata call, or a message queue call. Therefore, the present invention utilizes the cross-service scheduling (Span) tracked by the OpenTelemetry SDK as a carrier for obtaining key information on the scheduling chain.

[0025] The content of the cross-service scheduling (Span) includes attribute information, wherein the attribute information includes but is not limited to the unique identifier Span ID of the target cross-service scheduling (Span), the parent unique identifier Parant Span ID of the target cross-service scheduling (Span), wherein the parent unique identifier Parant Span ID of the target cross-service scheduling (Span) has a sequential relationship with the unique identifier Span ID of the target cross-service scheduling (Span) in the scheduling method of the present invention. The attribute information of the cross-service scheduling (Span) also includes an operation description Operation Name, wherein the operation description Operation Name includes but is not limited to operations such as GET, PUT, ORDER and CREATE, and the attribute content of the target cross-service scheduling (Span) also includes the start time START TIME and the duration. The core attributes of the target cross-service scheduling (Span) required in the present invention include the operation description Operation Name and key-value pair tags (Tags), wherein the key-value pair tags record the input content and output content of the target cross-service scheduling (Span), and the operation description Operation Name also records the input content and output content of the target cross-service scheduling (Span). Therefore, the present invention can obtain the description details of the corresponding call action in the call chain based on the operation description Operation Name and key-value pair tags. The present invention performs feature conversion based on the description details of the call action, thereby effectively analyzing different call actions in the call chain.

[0026] Furthermore, the present invention utilizes the OpenTelemetry SDK to track the operation descriptions (Operation Names) and key-value tags (Tags) present in historical enterprise service requests. Keywords present in the operation descriptions (Operation Names) and key-value tags (Tags) are then extracted using a keyword extraction algorithm, including but not limited to the TF-IDF algorithm. Only one keyword is extracted for each cross-service scheduling (Span). It should be noted that the TF-IDF algorithm described in the present invention is a keyword extraction algorithm constructed based on an enterprise service database as its primary model, and therefore has good scenario adaptability. Furthermore, the TF-IDF algorithm is prior art and will not be further described in detail in the present invention. For example, the keywords obtained using the TF-IDF algorithm may include, but are not limited to, core keywords such as procurement, talent, rental, logistics, express delivery, and fire. These core keywords are then processed using a clustering algorithm to determine the corresponding cross-service scheduling (Span) keyword category.

[0027] To better reference and visualize the relationships between corresponding service calls, the present invention utilizes the OpenTelemetry SDK to track the operation names and key-value tags in historical enterprise service requests. Based on the timestamps of the corresponding keyword actions, a structured view of the spatiotemporal dual mapping of service call relationships is generated, with a timestamp scale and time relationship. This structured view facilitates the identification of interactions between call operations and temporal relationships, and can detect complex call patterns, including parallel calls. The Gantt chart stores the corresponding target cross-service scheduling (Span) content in blocks.

[0028] The method specifically includes: using, but not limited to, a K-Means clustering algorithm to classify the call classes, including the steps of: obtaining call node input keywords, call node name keywords, and corresponding call node output keywords based on a Gantt chart generated by the call chain to obtain keyword triplets; wherein each keyword in the keyword triplets is obtained based on the operation description Operation Name and key-value pair tags and a corresponding keyword extraction algorithm such as a TF-IDF algorithm; performing keyword feature classification based on the keyword triplets; and calculating a similarity value between each keyword in the keyword triplets and the corresponding keywords of other keyword triplets, wherein the similarity value can be obtained using, but not limited to, a cosine similarity algorithm in combination with a preset variance constraint rule, wherein the variance constraint rule can be less than a specific variance threshold and the total similarity must be greater than a certain similarity threshold. The present invention uses a clustering algorithm to cluster all keyword triplets to obtain n clustered call classes, and calculates the probability distribution of each call class with its adjacent call classes, and selects the call class with the highest probability distribution to group the enterprise service catalog. It should be noted that the n clusters in the present invention may be pre-configured or may be n centroids obtained by a clustering algorithm, wherein the K-Means clustering algorithm may be used to obtain the n centroids by minimizing the cluster sum of squares as an objective function.

[0029] In another preferred embodiment of the present invention, the keyword triple clustering method includes: pre-generating n centroids according to the enterprise service call class, wherein the n centroids are pre-assigned with corresponding standard keyword triples; after obtaining a Gantt chart of the corresponding call chain, performing similarity calculation on the keywords in each keyword triple in the Gantt chart and the keywords in the keyword triple of each centroid, wherein the similarity calculation adopts cosine similarity calculation, and selecting the centroid of the keyword triple with the highest total similarity value and satisfying the variance threshold rule as the target centroid for classification; performing feature classification on the keyword triples related to all call classes in the Gantt chart to obtain n call classes, and calculating the probability distribution of adjacent call classes based on the n call classes to calculate the correlation value between different call classes. It should be noted that the keyword triples in the present invention are used to describe the high similarity or identity between the input content, input nodes, and output content of the same target cross-service scheduling (Span). The present invention utilizes the three constraints of the keyword triples to effectively avoid the possibility of incorrect recognition of a single input node constraint simply because a scheduling node has multiple different types of cross-service scheduling (Span) functions. It also avoids the problem of different recognition failures caused by different input subjects and output target objects. In other words, the keyword triple constraints described in the present invention can significantly improve the accuracy of cross-service scheduling (Span) recognition.

[0030] Furthermore, because a call node can be called multiple times, and the types of different calls may differ and may not be related, the present invention defines the corresponding cross-service scheduling (Span) type during a call as a call class. The present invention requires obtaining the correlation between different call classes, so that the enterprise information types with the highest correlation between parent and child calls can be used as the same parent call name. Specifically, the present invention defines a first call class and a second call class, wherein the first call class is the parent call of the second call class; obtains the unique identifiers Span ID / Parant Span ID of the first and second call classes based on the Gantt chart, obtains the operation descriptions Opertion Name and key-value tags corresponding to the first and second call classes based on the unique identifiers, obtains keyword triplets corresponding to the first and second call classes based on the operation descriptions Opertion Name and key-value tags, and determines the category combinations to which the two keyword triplets belong using the clustering algorithm, wherein the category combinations include the same category combination and different category combinations, and calculates the hierarchical relationship of each category combination. This effectively integrates the enterprise service information directory between strongly related parent and child call classes. For example, the category combination is as follows: service A needs to call service B. At this time, the A-to-B service call combination is regarded as a category combination. When service A calls service A, or service B calls service A, it is a different category combination.

[0031] In one of the preferred embodiments of the present invention, the number of each category combination is obtained based on the historical enterprise service information, and the hierarchical relationship of the call class corresponding to each category combination in the entire enterprise service information is recorded; the probability distribution of each category combination in all category combinations in the historical enterprise service information and the high and low probability distribution of each category combination are used to centrally construct the enterprise service information directory; wherein a first category combination probability threshold is set, and when the probability threshold of the corresponding category combination in all levels is greater than the first category combination probability threshold, the corresponding category combination is classified into the same parent category enterprise service information. In this embodiment, the hierarchical relationship between different service calls is recorded. For example: the call of service A to service B can exist simultaneously in the call of the first parent node to the second child node, or the call of the second parent node to the third child node, or the call of the second child node to the second parent node. The parent-child call relationship in the present invention is represented as a successive call relationship in the order of timestamps. In another preferred embodiment of the present invention, the number of each category combination is obtained based on the historical enterprise service information, and the hierarchical relationship of the call class corresponding to each category combination in the entire enterprise service information is recorded; the number of levels corresponding to each category combination is calculated, and after obtaining the number of all category combinations of the corresponding level, the probability distribution of each category combination at different levels is calculated, and the level with the highest probability distribution is selected as the construction related to the enterprise service information directory of the corresponding category combination, which is used to analyze the hierarchical distribution of the corresponding call class in different enterprise service information, and the call class includes the call of the parent level node to the child level node.

[0032] Obtain the node name of the first call class itself as the parent node, and obtain the second call class with the same parent node as the first call class in all call classes according to the parent node name, and calculate the clustering category to which all keyword triplets in the second call class belong, and configure the enterprise service information directory of the child node call class of which the first call class itself is the parent node according to the clustering category, and judge the probability distribution of the category combination between the first call class and the second call class in the same level through the first call class, and give priority to displaying the second call class with a higher probability value as the enterprise service information. The method for verifying the output results in the present invention includes: using OpenTelemetrySDK to build an API for calling services to analyze the enterprise service information call chain, using the Tracer component and Context Propagation component of the OpenTelemetrySDK to create Span and track different service calls, obtaining a call chain including the same Span ID / Parant Span ID and the same keyword triplet, and determining whether the enterprise service names corresponding to the call chains with the same Span ID / Parant Span ID and the same keyword triplet for each call class are the same; if different, reorganizing the enterprise service names and deleting the duplicate enterprise service names in the call chain.

[0033] In the embodiments disclosed in the present invention, the processes described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above functions are not limited to those in the method of the present application. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wire segments, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, an electromagnetic signal, an optical signal, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical fiber cable, RF, etc., or any suitable combination thereof.

[0034] The flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0035] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be subject to any deformation or modification.

Claims

1. A smart park enterprise service integrated management method, characterized in that: The method comprises: Pre-acquire historical enterprise service information, obtain the call chain of the historical enterprise service information according to a distributed tracing algorithm, and generate a Gantt chart containing each call node according to the call chain of the enterprise service information; Obtain keyword features of each call node in the Gantt chart, classify each call of the call node according to the keyword features to obtain a corresponding call class, and calculate the probability distribution between each call class according to the Monte Carlo algorithm; Calculating the correlation value between all call classes of each corresponding enterprise service call chain according to the probability distribution between each call class, and classifying the corresponding call classes into the same category combination name according to the correlation value; Each enterprise service name is obtained, and the enterprise service output result is verified according to the call class corresponding to the call chain, and operations including enterprise service name reorganization are performed according to the verification result.

2. A smart park enterprise service integrated management method according to claim 1, characterized in that: The keyword feature classification method includes: obtaining the call node input keyword, the call node name keyword and the corresponding call node output keyword according to the Gantt chart generated by the call chain to obtain keyword triplets, performing keyword feature classification according to the keyword triplets, calculating the similarity value between each keyword in the keyword triplets and the corresponding keywords of other keyword triplets, clustering all keyword triplets using a clustering algorithm to obtain n clustered call classes, and calculating the probability distribution of each call class and its adjacent call class, and selecting the call class with the highest probability distribution to group the enterprise service catalog.

3. A smart park enterprise service integrated management method according to claim 1, characterized in that: The keyword triple clustering method includes: pre-generating n centroids according to enterprise service call classes, wherein corresponding standard keyword triples are pre-assigned to the n centroids; after obtaining a Gantt chart of the corresponding call chain, performing similarity calculation on the keywords in each keyword triple in the Gantt chart and the keywords in the keyword triple of each centroid, wherein the similarity calculation adopts cosine similarity calculation, and selecting the centroid of the keyword triple with the highest total similarity value and satisfying the variance threshold rule as the target centroid for classification; after performing feature classification on the keyword triples related to all call classes in the Gantt chart, n call classes are obtained, and the probability distribution of the n call classes and adjacent call classes is calculated based on the n call classes to calculate the correlation value between different call classes.

4. A smart park enterprise service integrated management method according to claim 1, characterized in that: The probability distribution calculation method of n call classes and adjacent call classes includes: obtaining the globally unique identifier Trace ID of the corresponding enterprise service information according to the distributed tracing algorithm, defining a first call class and a second call class, wherein the first call class is the parent call of the second call class; obtaining the unique identifier SpanID / Parant Span ID of the first call class and the second call class according to the Gantt chart, and obtaining the operation description Opertion Name and key-value pair tags Tags corresponding to the first call class and the second call class according to the unique identifier, obtaining the keyword triples corresponding to the first call class and the second call class according to the operation description Opertion Name and the key-value pair tags Tags, and determining the category combination to which the two keyword triplets belong according to the clustering algorithm, wherein the category combination includes the same category combination and different category combinations, and counting the hierarchical relationship of each category combination.

5. A smart park enterprise service integrated management method according to claim 1, characterized in that: According to the historical enterprise service information, the number of each category combination is obtained, and the hierarchical relationship of the call class corresponding to each category combination in the entire enterprise service information is recorded; according to the probability distribution of each category combination in all category combinations in the historical enterprise service information, and according to the high and low probability distribution of each category combination, the enterprise service information directory is centrally constructed; wherein a first category combination probability threshold will be set, and when the probability threshold of the corresponding category combination in all levels is greater than the first category combination probability threshold, the corresponding category combination will be classified into the same parent category enterprise service information.

6. A smart park enterprise service integrated management method according to claim 1, characterized in that: According to the historical enterprise service information, the number of each category combination is obtained, and the hierarchical relationship of the call class corresponding to each category combination in the entire enterprise service information is recorded; the number of levels corresponding to each category combination is calculated, and after obtaining the number of all category combinations of the corresponding level, the probability distribution of each category combination at different levels is calculated, and the level with the highest probability distribution is selected as the construction related to the enterprise service information directory of the corresponding category combination, which is used to analyze the hierarchical distribution of the corresponding call class in different enterprise service information, and the call class includes the call of the parent level node to the child level node.

7. A smart park enterprise service integrated management method according to claim 4, characterized in that: Obtain the node name of the first call class itself as the parent node, and obtain the second call class with the same parent node as the first call class in all call classes according to the parent node name, and calculate the clustering category to which all keyword triplets in the second call class belong, and configure the enterprise service information directory of the child node call class of which the first call class itself is the parent node according to the clustering category, and judge the probability distribution of the category combination between the first call class and the second call class in the same level through the first call class, and give priority to displaying the second call class with a higher probability value as the enterprise service information.

8. A smart park enterprise service integrated management method according to claim 4, characterized in that: The method for verifying the output result includes: using the OpenTelemetry SDK to build an API for calling the service to analyze the enterprise service information call chain, using the Tracer component and Context Propagation component of the OpenTelemetry SDK to create Span and track different service calls, obtaining call chains with the same Span ID / Parant Span ID and the same keyword triples, and determining whether the enterprise service names corresponding to the call chains with the same Span ID / Parant Span ID and the same keyword triples for each call class are the same; if different, reorganizing the enterprise service names and deleting duplicate enterprise service names in the call chains.

9. A smart park enterprise service integrated management system, characterized by: The system executes a smart park enterprise service integrated management method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a smart park enterprise service integrated management method as described in any one of claims 1-8.

Images