Core invocation link topology generation method, device, equipment, medium and product

Abstract

The application relates to a core call link topology generation method, device, equipment, medium and product. The method comprises the following steps: aggregating call links of the same service into a weighted directed graph according to the number of calls; generating a random network of an original call link directed graph based on a zero model, wherein the random network contains the same nodes and edges as the original call link directed graph; comparing edges in the weighted directed graph with edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network; filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the entire random network to obtain a core call link topology. Therefore, the core call link topology can be focused on and mined from top to bottom, the dependence relationship of a business system can be understood from a whole perspective with the help of a backbone network, the call link topology can be effectively extracted and simplified, and key services and main call paths can be highlighted.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer network architecture, and particularly relates to a core call link topology generation method and device, computer equipment, a computer readable storage medium and a computer program product. BACKGROUND

[0002] With the development of micro-service technology of the backend system, a large number of business applications are also refined into more fine-grained micro-services. The micro-service architecture can improve the flexibility, scalability and universality of the system. In view of the characteristics of the micro-service architecture, the call chain tracking technology is introduced to assist in outlining the complete dependency path in the business system, so as to facilitate the viewing of the entire request call chain.

[0003] In the traditional technology, the call chain directed graph is usually generated according to the calling order of each service in the business interface. When a new transaction is triggered, the business interface generates a new call chain directed graph. Each node of the directed graph is a service, and each service can be described by multiple features. The cosine similarity of two directed graphs is compared based on these features, so as to determine the difference between the old transaction call link and the new transaction call link at the calling layer. If the difference is small, the calling layer of the new transaction call link and the calling layer of the old transaction call link can be fused to obtain the updated call link topology.

[0004] However, the above method does not consider the characteristics of the call chain on the network, nor does it consider the role played by each service node in the entire call chain network, resulting in an inaccurate call chain topology. SUMMARY

[0005] Therefore, it is necessary to provide a core call link topology generation method, device, computer equipment, computer readable storage medium and computer program product capable of generating a more effective and accurate core call chain in view of the above technical problems.

[0006] In a first aspect, the present application provides a core call link topology generation method, comprising:

[0007] aggregating the call links of the same business into a weighted directed graph according to the number of calls;

[0008] generating a random network of the original call link directed graph based on a zero model, the random network containing the same nodes and edges as the original call link directed graph;

[0009] comparing the edges in the weighted directed graph with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network;

[0010] According to the percentage of the significance probability of each edge in the whole random network, the edges in the weighted directed graph are filtered to obtain a core call link topology.

[0011] In one of the embodiments, the aggregating the call links of the same service into a weighted directed graph according to the number of calls comprises:

[0012] According to the call sequence of the services under the same service field, a directed graph of the call links is generated,

[0013] The number of times that the downstream nodes are called by the upstream nodes in each call link is counted respectively; wherein, the call link comprises the calling relationship between the upstream node and the downstream node;

[0014] According to the number of times that the downstream nodes are called, the weight of the corresponding call link is determined; wherein, the higher the number of times that the downstream nodes are called, the higher the weight of the corresponding call link;

[0015] The weight of the call link is fused in the generated directed graph to obtain a weighted directed graph.

[0016] In one of the embodiments, the comparing the edges in the weighted directed graph with the edges in the random network to obtain the percentage of the significance probability of each edge in the whole random network comprises:

[0017] Assuming that the weight of the edge in the weighted directed graph is uniformly distributed, the out-edge weight and the in-edge weight of each edge connected with the node are determined respectively for each node in the weighted directed graph;

[0018] The ratio of the out-edge weight of each edge to the total out-edge weight of the node and the ratio of the in-edge weight of each edge to the total in-edge weight of the node are determined;

[0019] According to the out-degree of the node, the in-degree of the node, the definite integral of the uniform distribution assumption, the ratio of the out-edge weight of each edge to the total out-edge weight of the node, and the ratio of the in-edge weight of each edge to the total in-edge weight of the node, the significance probability of each edge is determined;

[0020] According to the significance probability of each edge, the percentage of the significance probability of each edge in the whole random network is determined.

[0021] In one of the embodiments, the determining the percentage of the significance probability of each edge in the whole random network according to the significance probability of each edge comprises:

[0022] The significance probability of a given edge is set;

[0023] comparing the significance probability of all edges in the weighted directed graph with the significance probability of the given edge;

[0024] obtaining the number of all edges whose significance probability is not greater than the significance probability of the given edge;

[0025] taking the ratio of the number of all edges whose significance probability is not greater than the significance probability of the given edge to the total number of edges in the weighted directed graph as the percentile of the significance probability of each edge in the whole random network.

[0026] In one embodiment, before filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the whole random network, the method further comprises:

[0027] determining a percentile filtering value or a percentile filtering range for filtering the weighted directed graph according to the business requirement.

[0028] In one embodiment, filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the whole random network to obtain the core call link topology comprises:

[0029] filtering the edges in the weighted directed graph according to the percentile filtering value, and only keeping the edges whose percentile of the significance probability in the whole random network is equal to the percentile filtering value, or

[0030] according to the percentile filtering range, only keeping the edges whose percentile of the significance probability in the whole random network is within the percentile filtering range, to obtain the core call link topology.

[0031] In a second aspect, the application also provides a core call link topology generation device, which comprises:

[0032] an aggregation module configured to aggregate the call links of the same service into a weighted directed graph according to the number of calls;

[0033] a random network generation module configured to generate a random network of the original call link directed graph based on a zero model, wherein the random network comprises the same nodes and edges as the original call link directed graph;

[0034] a comparison module configured to compare the edges in the weighted directed graph with the edges in the random network to obtain the percentile of the significance probability of each edge in the whole random network;

[0035] a filtering module configured to filter the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the whole random network to obtain the core call link topology.

[0036] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0037] According to the number of calls, the call links of the same service are aggregated into a weighted directed graph;

[0038] A random network of the original call link directed graph is generated based on a zero model, wherein the random network comprises the same nodes and edges as the original call link directed graph;

[0039] The edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the whole random network;

[0040] According to the percentile of the significance probability of each edge in the whole random network, the edges in the weighted directed graph are filtered to obtain a core call link topology.

[0041] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the following steps:

[0042] According to the number of calls, the call links of the same service are aggregated into a weighted directed graph;

[0043] A random network of the original call link directed graph is generated based on a zero model, wherein the random network comprises the same nodes and edges as the original call link directed graph;

[0044] The edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the whole random network;

[0045] According to the percentile of the significance probability of each edge in the whole random network, the edges in the weighted directed graph are filtered to obtain a core call link topology.

[0046] In a fifth aspect, the present application provides a computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the following steps:

[0047] According to the number of calls, the call links of the same service are aggregated into a weighted directed graph;

[0048] A random network of the original call link directed graph is generated based on a zero model, wherein the random network comprises the same nodes and edges as the original call link directed graph;

[0049] comparing the edges in the weighted directed graph with the edges in the random network to obtain a percentile of a significance probability of each edge in the weighted directed graph in the entire random network;

[0050] filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the entire random network to obtain the core call link topology.

[0051] The core call link topology generation method, device, computer device, computer readable storage medium, and computer program product can aggregate call links of the same service into a weighted directed graph according to the number of calls, so that the call link network associated with the number of calls can be obtained according to the call order of the services under the same service, and the core call link can be conveniently extracted subsequently. The random network of the original call link directed graph is generated based on a zero model, and the random network includes the same nodes and edges as the original call link directed graph. The edges in the weighted directed graph are compared with the edges in the random network to obtain a percentile of a significance probability of each edge in the weighted directed graph in the entire random network, so that the influence of the randomly generated features on the call link can be excluded, and the core call link extracted subsequently is more effective and accurate. The edges in the weighted directed graph are filtered according to the percentile of the significance probability of each edge in the entire random network to obtain the core call link topology. Therefore, the core call link topology can be focused on and mined from top to bottom, the dependency relationship of the business system can be understood from a whole perspective with the help of the backbone network, the call link topology can be effectively extracted and simplified, and the key services and main call paths are highlighted. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0053] Figure 1 A schematic diagram of the generation principle of the weighted directed graph in one embodiment;

[0054] Figure 2 A schematic diagram of the flow of the core call link topology generation method in one embodiment;

[0055] Figure 3 A schematic diagram of the principle of the core call link topology generation method in one embodiment;

[0056] Figure 4 A schematic diagram of the flow of the core call link topology generation method in another embodiment;

[0057] Figure 5 Structure block diagram of the core call link topology generation device in one embodiment;

[0058] Figure 6 Structure block diagram of the core call link topology generation device in another embodiment;

[0059] Figure 7 Internal structure diagram of the computer device in one embodiment. DETAILED DESCRIPTION

[0060] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0061] In order to facilitate understanding of the technical solutions in the embodiments of the present application, the professional terms appearing in the embodiments of the present application are briefly described as follows:

[0062] Backbone Network: In network science, it refers to the most important and key connection part or subset in the network, usually composed of a small number of nodes and their connections. Backbone network is often used to highlight the core structure and key components of the network in network analysis and visualization.

[0063] Call Chain Tracing: It is a technology for analyzing and monitoring the execution flow and call relationship of distributed systems or applications. It records and traces the transmission path and call link of requests or transactions in the system, which can provide detailed visualization and analysis of system performance and behavior.

[0064] Call Chain Topology: It refers to the topology structure formed by the mutual call relationship between functions or services in computer systems or software applications. It describes the call order and dependency relationship between functions or services.

[0065] Weighted Directed Graph: It is a directed graph in which each edge is assigned a weight or cost.

[0066] Null Model: In network science and complex network analysis, it is a random model or benchmark model used to compare and evaluate the actual network structure. The purpose of the Null Model is to provide a reference benchmark so that researchers can judge whether the observed characteristics in the network are due to randomness or reflect the true structural characteristics.

[0067] Exemplary, Figure 1 A schematic diagram of the generation principle of the weighted directed graph in an embodiment is shown in Figure 1 To be able to analyze the dependency path of the business system from a global perspective, a batch of calling links under the same business field can be aggregated into a directed graph, and by analyzing the form and flow of the graph, the key nodes and dependency paths can be quickly sorted out, which is the calling chain link topology. For example, as shown in Figure 1 Application a has a complex dependency relationship with multiple downstream applications (b, c, d, e), and the path of each call is not the same. In order to sort out the complete calling dependency of application a, multiple calling chains are aggregated into a weighted directed graph according to the number of calls, and through further mining of the directed graph, the core applications and key paths in the calling chain can be obtained. It should be noted that in actual production environment, the link topology graph is often more complex than Figure 1 the one shown, but the principle of analyzing the calling dependency is the same.

[0068] In this embodiment, when the business system is very complex, the weighted directed graph of the complex calling link needs to be simplified to mine the core calling link, so that during the execution of the business, the core and non-core links can be protected through hierarchical strategies, thereby improving the stability and availability of the entire business system while reducing the labor cost.

[0069] In an exemplary embodiment, as shown in Figure 2 A core calling link topology generation method is provided, and the method in this embodiment can include the following steps 201 to 204.

[0070] Step 201, according to the number of calls, the calling links of the same business are aggregated into a weighted directed graph.

[0071] In this embodiment, as shown in Figure 1 First, all the calling links of application a are sorted out, and then the number of calls from the upstream node to the downstream node is counted. For example, the number of calls from a to b is 3, the number of calls from a to c is 7, the number of calls from c to d is 3, and the number of calls from c to e is 4, thereby obtaining the number of calls of each calling link in the directed graph. The more the number of calls, the higher the degree of dependence of the business system on the calling link. In order to reflect such a relationship in the graph, the weight of each calling link is set according to the number of calls.

[0072] According to the calling sequence of services under the same business scene, a directed graph of the calling link is generated, and the number of times that each downstream node is called by an upstream node in each calling link is counted respectively; the calling link includes the calling relationship between the upstream node and the downstream node; the weight of the corresponding calling link is determined according to the number of times that the downstream node is called; the higher the number of times that the downstream node is called, the higher the weight of the corresponding calling link; and the weight of the calling link is fused in the generated directed graph to obtain a weighted directed graph.

[0073] In step 202, a random network of the original calling link directed graph is generated based on a zero model.

[0074] The random network contains the same nodes and edges as the original calling link directed graph.

[0075] In this embodiment, the original calling link directed graph refers to a directed graph without weight, and the construction manner is generally as follows: according to the calling sequence of services under the same business scene, a directed graph of the calling link is generated. For example, the upstream node and the downstream node can be determined according to the service calling condition of the business, and then the connection relationship between all nodes is combined to form a directed graph. The random network is generated based on the original calling link by the zero model (which is actually a kind of node network similar to the directed graph). The features generated due to randomness in the weighted directed graph can be filtered out through the random network, so that the real features of the network result can be extracted.

[0076] In step 203, the edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network.

[0077] In this embodiment, assuming that the weights of the edges in the weighted directed graph are uniformly distributed, the out-edge weight and the in-edge weight of each edge connected to each node in the weighted directed graph are determined respectively; the ratio of the out-edge weight of each edge to the total out-edge weight of the node and the ratio of the in-edge weight of each edge to the total in-edge weight of the node are determined; the significance probability of each edge is determined according to the out-degree of the node, the in-degree of the node, the definite integral of the uniform distribution assumption, the ratio of the out-edge weight of each edge to the total out-edge weight of the node, and the ratio of the in-edge weight of each edge to the total in-edge weight of the node; and the percentile of the significance probability of each edge in the entire random network is determined according to the significance probability of each edge.

[0078] According to the local significance theory, it is determined whether the weight of a specific edge is significant for all connections of a node. By assuming that the weights of the edges are uniformly distributed, the significance probability of each edge under the distribution is calculated, so that those significant edges which have important contribution to the network structure are identified.

[0079] For example, for each node i, calculate its outgoing edge weight w. ij and the weight of the incoming edge w ji It accounts for a certain percentage of its total output weight and total input weight Ratio:

[0080]

[0081] Where j represents the j-th edge connecting node i. Indicate the edge weight w ij Weight of total output proportion, Indicates the weight of the incoming edge w ji Weight of total input The proportion.

[0082] Assuming the edge weights are uniformly distributed, calculate the significance of each edge. and Indicate the saliency of the edges. The values ​​indicate the significance of the incoming edges. The lower the significance of both values, the less likely the edge weight is to be a random value, and the stronger the significance.

[0083] The formula for calculating the significance of outgoing edges is as follows:

[0084]

[0085] in, It is the out-degree of node i, and * represents the multiplication sign.

[0086] The formula for calculating the significance of incoming edges is as follows:

[0087]

[0088] in, It is the in-degree of node i.

[0089] Where DI represents the definite integral of the probability density function (uniform distribution assumption), and its calculation formula is as follows:

[0090]

[0091] Where: x represents the weight variable, and k represents the angle variable.

[0092] The formula above is derived from the cumulative distribution function and is used to evaluate the saliency probability of a particular edge's weight under the assumption of a uniform distribution. This saliency reflects the relative importance of the edge in the network.

[0093] For example, a saliency probability of a given edge is set; the saliency probabilities of all edges in the weighted directed graph are compared with the saliency probability of the given edge; the number of all edges whose saliency probability is not greater than that of the given edge is obtained; and the ratio of the number of all edges whose saliency probability is not greater than that of the given edge to the total number of edges in the weighted directed graph is used as the percentile of the saliency probability of each edge in the entire random network.

[0094] In this embodiment, the formula for calculating the percentile of the salience probability of each edge in the entire random network is as follows:

[0095] Percentile of edge significance = (Number of edges whose significance is not greater than the given edge significance / Total number of edges) * 100%

[0096] Step 204: Filter the edges in the weighted directed graph based on the percentile of the salience probability of each edge in the entire random network to obtain the core call link topology.

[0097] For example, such as Figure 3 As shown, the entire principle of extracting the core call chain is illustrated. First, call chain data is obtained; then, a directed graph of service calls is constructed based on the call chain data; finally, the core call chain is extracted using the backbone network algorithm. Figure 3 In this context, the backbone network algorithm refers to the method of filtering the call links in steps 203 and 204.

[0098] In this embodiment, based on the percentile of edge saliency obtained in step 203, the edges in the weighted directed graph are filtered to remove some unimportant or low-contribution edges, thus obtaining the core call link topology.

[0099] In the core call link topology generation method, the call links of the same service are aggregated into a weighted directed graph according to the number of calls, so that the call link network associated with the number of calls can be obtained according to the call sequence of the service, and the subsequent extraction of the core call link is facilitated. A random network of the original call link directed graph is generated based on a zero model, and the random network includes the same nodes and edges as the original call link directed graph. The edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network. Therefore, the influence of the randomly generated features on the call link can be excluded, and the subsequently extracted core call link is more effective and accurate. According to the percentile of the significance probability of each edge in the entire random network, the edges in the weighted directed graph are filtered to obtain the core call link topology. Therefore, the core call link topology can be focused on and mined from top to bottom, the dependency relationship of the business system can be understood from a whole perspective with the help of the backbone network, the call link topology can be effectively extracted and simplified, and the key services and main call paths are highlighted.

[0100] In another exemplary embodiment, as shown in Figure 4 A core call link topology generation method is provided. The method in the embodiment can include the following steps 401 to 405.

[0101] Step 401: The call links of the same service are aggregated into a weighted directed graph according to the number of calls.

[0102] Step 402: A random network of the original call link directed graph is generated based on a zero model.

[0103] The random network includes the same nodes and edges as the original call link directed graph.

[0104] Step 403: The edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network.

[0105] For the specific implementation process and technical effects of steps 401 to 403 in the embodiment, please refer to the related content of steps 201 to 203 in the method embodiment shown in Figure 2

[0106] Step 404: According to the business requirements, a percentile filtering value or a percentile filtering range for filtering the weighted directed graph is determined.

[0107] In the embodiment, it is assumed that there are 100 service nodes and 540 weighted call edges, and the calculated significance and percentile of each edge are shown in Table 1. ​

[0108] Table 1

[0109] Percentile Significance of edges 0.00 0.0000 0.10 0.0189 0.20 0.0559 0.30 0.0948 0.40 0.1570 0.50 0.2125 0.60 0.3238 0.70 0.4777 0.80 0.6972 0.90 0.9783

[0110] For the above Table 1, the percentile filtering value for filtering the weighted directed graph can be set, or the percentile filtering range. For example, a difference filter is made based on the percentile filtering value, and the call links of the weighted directed graph are filtered through the filter.

[0111] Step 405, filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the whole random network, to obtain the core call link topology.

[0112] In this embodiment, the edges in the weighted directed graph are filtered according to the percentile filtering value, and only the edges with the percentile of the significance probability equal to the percentile filtering value in the whole random network are retained, or according to the percentile filtering range, only the edges with the percentile of the significance probability within the percentile filtering range in the whole random network are retained, to obtain the core call link topology.

[0113] For example, as shown in Table 2, the number of remaining edges and nodes after filtering with different percentile filtering values is given.

[0114] Table 2

[0115] Difference filter Number of edges Number of nodes Percentile of edge saliency <= 0.50 266 96 percentile of edge significance <= 0.30 146 79 percentile of edge significance <= 0.10 24 24

[0116] For the above Table 2, different core call link topologies can be obtained by setting different percentile filtering values or percentile filtering ranges.

[0117] In this embodiment, the core call link topology (backbone network) mined can help identify services that play a key role in the call chain. These services are usually the core components of the system and are crucial to the performance and reliability of the system. After identifying the key services, more resources can be allocated to these services to ensure their high availability and high performance. For example, in a complex call chain, the backbone network can help extract the main call path, simplify the call chain topology, and make the analysis more intuitive. For the main call path, performance optimization can be focused on, which helps to improve the overall performance of the system. Moreover, the simplified call chain topology is more suitable for visual display, helping developers and operation and maintenance personnel better understand the system structure.

[0118] In addition, in a distributed system, the complexity of the call chain increases the difficulty of fault location. The backbone network can help simplify this process. After identifying the key services and paths, fault recovery strategies can be more efficiently formulated, reducing system downtime.

[0119] It should be understood that although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0120] Based on the same inventive concept, the embodiments of the present application also provide a core call link topology generation device for implementing the core call link topology generation method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more core call link topology generation device embodiments provided below can refer to the limitations of the core call link topology generation method in the foregoing, which will not be repeated here.

[0121] In an exemplary embodiment, as shown in Figure 5 A core call link topology generation device is provided, comprising: an aggregation module 501, a random network generation module 502, a comparison module 503, and a filtering module 504, wherein:

[0122] The aggregation module 501 is configured to aggregate the call links of the same service into a weighted directed graph according to the number of calls;

[0123] The random network generation module 502 is configured to generate a random network of the original call link directed graph based on a zero model, the random network containing the same nodes and edges as the original call link directed graph;

[0124] The comparison module 503 is configured to compare the edges in the weighted directed graph with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network;

[0125] The filtering module 504 is configured to filter the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the entire random network to obtain the core call link topology.

[0126] According to the same service scene, the aggregation module 501 is specifically configured to generate a directed graph of the calling link according to the calling sequence of the services, and respectively count the number of times that each downstream node is called by an upstream node in each calling link; wherein the calling link includes the calling relationship between the upstream node and the downstream node; and determine the weight of the corresponding calling link according to the number of times that the downstream node is called; wherein the higher the number of times that the downstream node is called, the higher the weight of the corresponding calling link; and fuse the weight of the calling link in the generated directed graph to obtain a weighted directed graph.

[0127] According to the same service scene, the aggregation module 501 is specifically configured to generate a directed graph of the calling link according to the calling sequence of the services, and respectively count the number of times that each downstream node is called by an upstream node in each calling link; wherein the calling link includes the calling relationship between the upstream node and the downstream node; and determine the weight of the corresponding calling link according to the number of times that the downstream node is called; wherein the higher the number of times that the downstream node is called, the higher the weight of the corresponding calling link; and fuse the weight of the calling link in the generated directed graph to obtain a weighted directed graph.

[0128] According to the same service scene, the aggregation module 501 is specifically configured to generate a directed graph of the calling link according to the calling sequence of the services, and respectively count the number of times that each downstream node is called by an upstream node in each calling link; wherein the calling link includes the calling relationship between the upstream node and the downstream node; and determine the weight of the corresponding calling link according to the number of times that the downstream node is called; wherein the higher the number of times that the downstream node is called, the higher the weight of the corresponding calling link; and fuse the weight of the calling link in the generated directed graph to obtain a weighted directed graph.

[0129] In another exemplary embodiment, as shown in Figure 6 The core calling link topology generation device is provided, and on the basis of the device shown in Figure 5 The determination module 505 is further configured to determine a percentile filtering value or a percentile filtering range for filtering the weighted directed graph according to the service demand.

[0130] According to the same service scene, the aggregation module 501 is specifically configured to generate a directed graph of the calling link according to the calling sequence of the services, and respectively count the number of times that each downstream node is called by an upstream node in each calling link; wherein the calling link includes the calling relationship between the upstream node and the downstream node; and determine the weight of the corresponding calling link according to the number of times that the downstream node is called; wherein the higher the number of times that the downstream node is called, the higher the weight of the corresponding calling link; and fuse the weight of the calling link in the generated directed graph to obtain a weighted directed graph.

[0131] Each module in the core call link topology generation apparatus described above can be implemented by software, hardware, and combinations thereof, in whole or in part. The modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0132] In an example embodiment, a computer device, which can be a terminal, can have an internal structure diagram as shown in Figure 7 The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be achieved through WIFI, mobile cellular network, near field communication (NFC), or other technologies. The computer program is executed by the processor to implement a core call link topology generation method. The display unit of the computer device is configured to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball, or touchpad arranged on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0133] Those skilled in the art can understand that Figure 7 The structure shown in the above

[0134] In an example embodiment, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0135] In an embodiment, a computer readable storage medium is provided, and the computer readable storage medium has stored thereon a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0136] In an embodiment, a computer program product is provided, and the computer program product includes a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0137] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of the related data need to comply with relevant regulations.

[0138] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0139] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0140] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for generating a core call link topology, characterized by, The method comprises: According to the number of calls, the call links of the same service are aggregated into a weighted directed graph; the weight of each call link is set according to the number of calls; Based on the zero model, a random network of the original call link directed graph is generated, the random network contains the same nodes and edges as the original call link directed graph; the original call link directed graph refers to a directed graph without weighting; The edges in the weighted directed graph are compared with the edges in the random network to obtain the percentile of the significance probability of each edge in the weighted directed graph in the entire random network; assuming that the weights of the edges in the weighted directed graph are uniformly distributed, the out-edge weight and the in-edge weight of each edge connected to each node in the weighted directed graph are determined respectively; the ratio of the out-edge weight of each edge to the total out-edge weight of the node and the ratio of the in-edge weight of each edge to the total in-edge weight of the node are determined; the significance probability of each edge is determined according to the out-degree of the node, the in-degree of the node, the definite integral of the uniform distribution assumption, the ratio of the out-edge weight of each edge to the total out-edge weight of the node, and the ratio of the in-edge weight of each edge to the total in-edge weight of the node; the percentile of the significance probability of each edge in the entire random network is determined according to the significance probability of each edge; According to the percentile of the significance probability of each edge in the entire random network, the edges in the weighted directed graph are filtered to obtain the core call link topology.

2. The method of claim 1, wherein, The method comprises: According to the call sequence of the services under the same service field, a directed graph of the call link is generated, The number of times that the downstream nodes in each call link are called by the upstream nodes is counted respectively; wherein the call link includes the calling relationship between the upstream nodes and the downstream nodes; According to the number of times that the downstream nodes are called, the weight of the corresponding call link is determined; wherein the higher the number of times that the downstream nodes are called, the higher the weight of the corresponding call link; The weight of the call link is fused in the generated directed graph to obtain a weighted directed graph.

3. The method of claim 1, wherein, The method comprises: The significance probability of a given edge is set; The significance probability of all edges in the weighted directed graph is compared with the significance probability of the given edge; The number of all edges whose significance probability is not greater than the significance probability of the given edge is obtained; The ratio of the number of all edges whose significance probability is not greater than the significance probability of the given edge to the total number of edges in the weighted directed graph is taken as the percentile of the significance probability of each edge in the entire random network.

4. The method according to any one of claims 1 to 3, characterized in that, Before filtering the edges in the weighted directed graph according to the percentile of the significance probability of each edge in the entire random network, the method further comprises: According to the business requirements, a percentile screening value or a percentile screening range for filtering the weighted directed graph is determined.

5. The method of claim 4, wherein, The method comprises: The edges in the weighted directed graph are filtered according to the percentile screening value, retaining only edges whose significance probability in the entire random network is equal to the percentile screening value, or Based on the percentile filtering range, only edges whose salience probability in the entire random network falls within the percentile filtering range are retained to obtain the core call link topology.

6. A core call link topology generation apparatus characterized by comprising: The device includes: The aggregation module is used to aggregate call chains for the same business into a weighted directed graph based on the number of calls; and to set the weight of each call chain based on the number of calls. The random network generation module is used to generate a random network of the original call chain directed graph based on the zero model. The random network contains the same nodes and edges as the original call chain directed graph. The original call chain directed graph refers to an unweighted directed graph. The comparison module is used to compare the edges in the weighted directed graph with the edges in the random network to obtain the percentile of the saliency probability of each edge in the weighted directed graph in the entire random network. Assuming the weights of the edges in the weighted directed graph are uniformly distributed, for each node in the weighted directed graph, the outgoing and incoming weights of each edge connected to the node are determined. The ratio of the outgoing weight of each edge to the total outgoing weight of the node, and the ratio of the incoming weight of each edge to the total incoming weight of the node are determined. Based on the node's out-degree, in-degree, definite integral of the uniform distribution assumption, the ratio of the outgoing weight of each edge to the total outgoing weight of the node, and the ratio of the incoming weight of each edge to the total incoming weight of the node, the saliency probability of each edge is determined. Based on the saliency probability of each edge, the percentile of the saliency probability of each edge in the entire random network is determined. The filtering module is used to filter the edges in the weighted directed graph based on the percentile of the salience probability of each edge in the entire random network, so as to obtain the core call link topology.

7. The apparatus of claim 6, wherein, The aggregation module is specifically used for: generating a directed graph of call chains based on the call order of services under the same business field; counting the number of times each downstream node in each call chain is called by the upstream node; wherein, the call chain includes the call relationship between the upstream and downstream nodes; determining the weight of the corresponding call chain based on the number of times the downstream node is called; wherein, the higher the number of times the downstream node is called, the higher the weight of the corresponding call chain; and merging the weights of the call chains into the generated directed graph to obtain a weighted directed graph.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images