Abnormal node positioning method and device, equipment, storage medium and computer program product

Abstract

The invention discloses an abnormal node positioning method and device, equipment, a storage medium and a computer program product, and relates to the technical field of big data platform operation and maintaining.The method comprises the steps that target data is packaged, and packaged data is obtained; determining a critical path based on the encapsulation data; and performing node analysis on the key path, and positioning abnormal nodes of the cross-domain calculation task according to an analysis result. According to the method, after the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform is packaged in a unified mode, the key path of the cross-domain computing task is determined based on the packaged data, and the key path is the path with the longest time for completing the sub-task in the cross-domain computing task; in other words, the critical path contains the most critical task for determining the circulation period of the cross-domain calculation task, so that the abnormal node which has the greatest influence on the circulation time of the whole cross-domain calculation task can be positioned more clearly by analyzing each node in the critical path.

Description

Technical Field

[0001] The present application relates to the field of big data platform operation and maintenance technology, and in particular to an abnormal node positioning method, device, equipment, storage medium and computer program product. Background Art

[0002] With the development and application of cloud computing, big data, and artificial intelligence technologies, computing tasks on big data cloud platforms are becoming increasingly complex and diverse. Currently, computing tasks on big data cloud platforms are often cross-domain, consisting of numerous subtasks that must be transferred between different computing domains. During this transfer process, an anomaly in any node processing a subtask can cause delays, failures, or other anomalies in the entire task. However, relevant technologies struggle to pinpoint the nodes that may be causing these anomalies, impacting task execution efficiency and reliability. Summary of the Invention

[0003] The main purpose of this application is to provide a method, device, equipment, storage medium and computer program product for locating abnormal nodes, aiming to solve the technical problem that related technologies are difficult to locate abnormal nodes in cross-domain computing tasks.

[0004] To achieve the above objectives, the present application provides a method for locating abnormal nodes, which includes the following steps:

[0005] Encapsulating target data to obtain encapsulated data, wherein the target data is operation and maintenance data included in a cross-domain computing task in a big data cloud platform;

[0006] Determine a critical path based on the encapsulated data, where the critical path is the path that takes the longest time to complete a subtask in the cross-domain computing task;

[0007] Perform node analysis on the critical path, and locate abnormal nodes of the cross-domain computing task based on the analysis results.

[0008] In one embodiment, the step of determining the critical path based on the package data includes:

[0009] Constructing a knowledge graph corresponding to the cross-domain computing task according to the encapsulated data;

[0010] Based on the knowledge graph, the execution completion time of the task execution path included in the cross-domain computing task is predicted, and the critical path is determined according to the prediction result.

[0011] In one embodiment, the step of predicting the execution completion time of the task execution path included in the cross-domain computing task based on the knowledge graph and determining the critical path according to the prediction result includes:

[0012] Obtaining a task dependency topology graph of subtasks in the cross-domain computing task from the knowledge graph, and determining a predicted start time and a predicted end time of the subtasks;

[0013] Predicting the execution completion time of the task execution path included in the cross-domain computing task based on the task dependency topology graph, the predicted start time, and the predicted end time;

[0014] According to the prediction result, the path in the task execution path that takes the longest time to complete the subtask is determined as the critical path.

[0015] In one embodiment, the step of determining the predicted start time and predicted end time of the subtask includes:

[0016] If the subtask belongs to the first node, the time of the subtask is predicted using the first prediction principle to obtain the predicted start time and the predicted end time;

[0017] If the subtask belongs to a non-first node, a time prediction is performed on the subtask using a second prediction principle to obtain a predicted start time and a predicted end time.

[0018] In one embodiment, the step of encapsulating the target data to obtain encapsulated data includes:

[0019] Determine the subtask attributes and subtask dependencies corresponding to the cross-domain computing task based on the target data;

[0020] The target data is encapsulated based on the subtask attributes and the subtask dependencies to obtain encapsulated data, where the encapsulated data is in a triple form.

[0021] In one embodiment, the step of performing node analysis on the critical path and locating abnormal nodes of the cross-domain computing task according to the analysis results includes:

[0022] Performing node analysis on the critical path according to the time parameters of each node in the critical path, and calculating the criticality score of each node according to the analysis results;

[0023] A node corresponding to a highest score among the criticality scores is determined as an abnormal node of the cross-domain computing task, wherein the criticality score is used to indicate the degree of time influence of the node on the critical path.

[0024] In addition, to achieve the above-mentioned purpose, the present application also proposes an abnormal node locating device, which includes:

[0025] A data encapsulation module is used to encapsulate target data to obtain encapsulated data, where the target data is the operation and maintenance data included in the cross-domain computing task in the big data cloud platform;

[0026] A critical path determination module, configured to determine a critical path based on the encapsulated data, wherein the critical path is a path that takes the longest time to complete a subtask in the cross-domain computing task;

[0027] The abnormal node locating module is used to perform node analysis on the critical path and locate abnormal nodes of the cross-domain computing task according to the analysis results.

[0028] In addition, to achieve the above-mentioned purpose, the present application also proposes an abnormal node locating device, which includes: a memory, a processor, and an abnormal node locating program stored in the memory and executable on the processor, wherein the abnormal node locating program is configured to implement the steps of the abnormal node locating method described above.

[0029] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium and stores an abnormal node locating program. When the abnormal node locating program is executed by a processor, the steps of the abnormal node locating method described above are implemented.

[0030] In addition, to achieve the above-mentioned object, the present invention further provides a computer program product, which includes an abnormal node locating program. When the abnormal node locating program is executed by a processor, the steps of the abnormal node locating method described above are implemented.

[0031] This application encapsulates the target data to obtain encapsulated data, where the target data is the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform; determines the critical path based on the encapsulated data, where the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task; performs node analysis on the critical path, and locates the abnormal nodes of the cross-domain computing task based on the analysis results. The above method of this application uniformly encapsulates the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform, and then determines the critical path of the cross-domain computing task based on the encapsulated data. Since the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task, that is, the critical path contains the most critical task that determines the flow cycle of the cross-domain computing task, this application analyzes each node in the critical path, so as to more clearly locate the abnormal node that has the greatest impact on the flow time of the entire cross-domain computing task. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0034] Figure 1 This is a flow chart of the first embodiment of the abnormal node location method of the present application;

[0035] Figure 2 This is a system collaboration diagram of the abnormal node positioning method of this application;

[0036] Figure 3 This is a flow chart of the third embodiment of the abnormal node locating method of the present application;

[0037] Figure 4 This is the first node prediction flow chart of the abnormal node positioning method of this application;

[0038] Figure 5 This is a non-first node prediction flow chart of the abnormal node positioning method of this application;

[0039] Figure 6 This is a flow chart of the third embodiment of the abnormal node locating method of the present application;

[0040] Figure 7 This is a diagram of the end-to-end flow visualization module of the abnormal node positioning method in this application;

[0041] Figure 8 This is an example diagram of the key task data flow topology of the abnormal node positioning method of this application;

[0042] Figure 9 This is the task tracking diagram of the abnormal node positioning method of this application;

[0043] Figure 10 This is a structural block diagram of the first embodiment of the abnormal node locating device of the present application;

[0044] Figure 11 This is a structural diagram of an abnormal node locating device in a hardware operating environment involved in an embodiment of the present application.

[0045] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not intended to limit the present application.

[0047] It should be noted that the execution entity of the embodiments of the present application can be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of performing the above functions, such as the above-mentioned abnormal node locating device. The following embodiments are described below using the abnormal node locating device as an example.

[0048] This application embodiment provides a method for locating abnormal nodes. Figure 1 , Figure 1 This is a flow chart of the first embodiment of the abnormal node location method of the present application.

[0049] In this embodiment, the abnormal node location method includes the following steps:

[0050] Step S1: Encapsulate target data to obtain encapsulated data, where the target data is operation and maintenance data included in a cross-domain computing task in a big data cloud platform.

[0051] It should be noted that the above-mentioned big data cloud platform is a comprehensive platform that can help users achieve full life cycle management of massive data, including collection, storage, processing, analysis and visualization, by integrating the elastic resources of cloud computing (such as computing, storage, and network) with big data processing technologies (such as distributed computing and data mining).

[0052] It is understood that the above-mentioned cross-domain computing tasks can refer to tasks that need to flow between different computing domains (such as infrastructure, cloud platforms, data centers, and software services in different regions). For example, a task that includes subtasks such as virtual machine instance startup, virtual machine initialization, container orchestration, microservice construction, microservice registration, data conversion operations, transaction request monitoring, and data reporting can be considered a cross-domain computing task.

[0053] It should be understood that the above-mentioned target data can be collected in real time from different business layers, including but not limited to the Infrastructure as a Service (IaaS) layer, the Platform as a Service (PaaS) layer, the Data as a Service (DaaS) layer, and the Software as a Service (SaaS) layer. Among them, the IaaS layer mainly focuses on data collection of infrastructure-related subtasks such as physical machines, virtual machines, containers, and clusters, including but not limited to the status of physical servers, the configuration and resource utilization of virtual machines, the deployment and operation status of containers, and the health and performance indicators of clusters; the PaaS layer involves data collection of subtasks related to platform components and services, mainly including platform-level application services, management of microservice architecture, and the operation status of various middleware; the DaaS layer mainly involves providing data on subtasks related to online transaction processing and online analytical task processing, including business request processing, business indicator analysis and calculation of each tenant; the SaaS layer focuses on data collection of subtasks related to system services and components, covering the operation and status monitoring of various software systems such as application software, operating system services, and database services.

[0054] In a specific implementation, the above-mentioned target data can be encapsulated and aggregated in a preset form (such as triple form, binary form, etc.), thereby forming a unified task operation and maintenance data storage structure (including node-based business data encapsulation and node dependency encapsulation) to obtain the above-mentioned encapsulated data.

[0055] Step S2: determining a critical path based on the encapsulated data, where the critical path is the path that takes the longest time to complete a subtask in the cross-domain computing task.

[0056] It should be understood that the subtasks on the critical path can be called critical subtasks (nodes), and their delays will directly affect the total duration of the project corresponding to the cross-domain computing task. Since the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task, the critical path is crucial to ensuring that the project is completed on time. In this embodiment, the critical path method (CPM) can be used to analyze the critical path for task execution. Based on the critical path method, the late start time (LST) and late finish time (LFT) of the subtask can be obtained: specifically, the start and completion time of each node contained in the task can be predicted in real time based on the real-time status and historical data of the task, for example, by calculating the median, minimum, maximum and quartile values of the task execution time in several historical cycles, the reasonable execution time of each subnode in the current task can be inferred. Then, a rule-based analysis method can be used to obtain the subgraph related to the task to form a task dependency topology diagram. Based on the LST and LFT analysis results, the completion time of the entire task flow process is predicted. If certain subtasks are delayed or their predicted execution times exceed the normal range, the overall estimated completion time for the task will be recalculated, and the estimated completion times of the dependent subtasks will be adjusted to provide early warning of potential delays. Therefore, the critical path method can be used to predict the completion time of each path in a cross-domain computing task, thereby identifying the path with the longest subtask completion time as the critical path.

[0057] Step S3: performing node analysis on the critical path, and locating abnormal nodes of the cross-domain computing task according to the analysis results.

[0058] It should be noted that the above-mentioned abnormal node may refer to a critical task node (also known as a bottleneck point / blocking point) that has the greatest impact on the flow time of the entire cross-domain computing task.

[0059] In specific implementation, the node analysis of the critical path can be achieved by determining the degree of influence of each node in the critical path on the entire critical path. For example, by comparing the difference between the actual end time and the expected end time of each node, the impact of the upstream task end time on the start time of the task, etc., the abnormal nodes in the critical path can be located according to the analysis results.

[0060] In addition, in order to more systematically and comprehensively describe the specific implementation of the technical solution in this embodiment, you can refer to Figure 2 , Figure 2This is a system collaboration diagram for the abnormal node location method of this application. It should be noted that the system can be built into the abnormal node location device described above and includes, but is not limited to, a data acquisition module, a data encapsulation module, a data aggregation and storage module, a task process knowledge graph storage module, a critical path and task discovery module, a task anomaly analysis and prediction module, and a task end-to-end flow visualization module. Specifically, the data collection module can collect task operation and maintenance data from different business layers in real time; the data encapsulation module encapsulates and aggregates the collected operation and maintenance data, and the data processed by the data encapsulation module can be stored in the cross-domain computing task flow data warehouse (including the data aggregation storage module and the task process knowledge graph storage module) to achieve data persistence and support detailed traceability of cloud computing task flow; the data aggregation storage module stores the attribute information of specific subtask nodes in the cross-domain computing task flow process, as well as the statistical and predicted data of subtask operation and maintenance indicators. The former provides a basis for task flow anomaly analysis, and the latter can help operation and maintenance personnel discover possible weaknesses in the task flow process in advance. Relevant indicators can be calculated and derived by the upper-level module based on real-time historical data inference; the task process knowledge graph storage module can construct a knowledge graph of the flow of cross-domain computing tasks between different domain levels and store it in the graph database, while providing management and query of task execution paths and task subgraphs. Function: The critical path and task discovery module confirms the bottleneck paths and bottleneck points that affect task execution by reading subtask history and real-time flow data, providing a basis for task anomaly analysis and indicator prediction; the task anomaly analysis and prediction module can read critical paths and task attributes, and can not only instantly infer the anomalies and bottleneck points that appear in the task flow process and trace their origins, but also predict future task flow processes and critical paths by summarizing historical data, and provide optimization suggestions for improving task operation efficiency; the task end-to-end flow visualization module can read task-related operation and maintenance data, anomalies, and prediction data from the data warehouse according to the cross-domain computing tasks specified by the user, and provide the ability to visually track the task delivery process. It can graphically display the propagation path and execution process of cloud computing tasks between various business layers, and provide a detailed analysis and tracing interface for subtasks, which can effectively help operation and maintenance personnel discover and understand the causes of performance bottlenecks or anomalies in cloud computing tasks.

[0061] Through the interaction between the above modules, this embodiment can achieve comprehensive observation and analysis of cross-domain computing tasks, provide the ability to accurately locate and predict the flow status of computing tasks, and effectively improve the execution efficiency and reliability of tasks. This embodiment effectively assists in the end-to-end operation and maintenance troubleshooting of cross-domain cloud computing tasks through the collaborative work of the data acquisition module, data encapsulation module, data aggregation and storage module, task process knowledge graph storage module, task anomaly analysis and prediction module, critical path and task discovery module, and task end-to-end flow visualization module. In addition, this embodiment can be mainly applied to, but not limited to, the following two scenarios: First, when an anomaly is detected by an existing monitoring and alarm device, the above method of this embodiment can combine the analysis data generated by the task anomaly analysis and prediction module to assist operation and maintenance personnel in locating and resolving the alarm problem; Second, during daily operation and maintenance inspections, the above method of this embodiment can use the knowledge graph combined with anomaly detection data and historical data to evaluate the health status of devices, platforms, data, and application systems in the cross-domain system, provide optimization suggestions, and ensure the efficient and stable operation of the system.

[0062] This embodiment encapsulates the target data to obtain encapsulated data, wherein the target data is the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform; determines the critical path based on the encapsulated data, wherein the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task; performs node analysis on the critical path, and locates the abnormal nodes of the cross-domain computing task based on the analysis results. After the above method of this embodiment uniformly encapsulates the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform, it determines the critical path of the cross-domain computing task based on the encapsulated data. Since the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task, that is, the critical path contains the most critical task that determines the flow cycle of the cross-domain computing task, this embodiment analyzes each node in the critical path, thereby being able to more clearly locate the abnormal node that has the greatest impact on the flow time of the entire cross-domain computing task.

[0063] refer to Figure 3 , Figure 3 This is a flow chart of the second embodiment of the abnormal node locating method of the present application.

[0064] In a feasible implementation, step S2 may include:

[0065] Step S21: Construct a knowledge graph corresponding to the cross-domain computing task based on the encapsulated data.

[0066] It should be noted that the aforementioned knowledge graph may include, but is not limited to, information such as task attributes and dependencies corresponding to each subtask in a cross-domain computing task. In this embodiment, a graph database can be used to store the aforementioned encapsulated data. By continuously loading and updating the encapsulated data, the knowledge graph corresponding to the cross-domain computing task can be dynamically constructed and improved, thereby providing operations and maintenance personnel with real-time task flow status monitoring and anomaly detection capabilities.

[0067] In a specific implementation, the above-mentioned knowledge graph can provide functions not limited to the following: schema management (including adding, modifying, and deleting vertex types, and adding, modifying, and deleting edge types), knowledge graph node management (inserting, deleting, and modifying vertex types), knowledge graph edge management (inserting, deleting, and modifying edges), and knowledge graph query functions (including vertex and edge queries, dependency queries, and qualified subgraph queries). In particular, these functions can rely on industry-wide graph database operation specifications for data definition and operations, such as Cypher (a declarative graph query language) and GQL (Graph Query Language), which are not limited in this embodiment.

[0068] Step S22: predicting the execution completion time of the task execution path included in the cross-domain computing task based on the knowledge graph, and determining the critical path according to the prediction result.

[0069] It should be understood that since a cross-domain computing task contains multiple task execution paths, and the knowledge graph can quickly and comprehensively obtain all the task execution paths specifically included in the cross-domain computing task, in this embodiment, the execution completion time of the task execution paths included in the cross-domain computing task can be predicted based on the knowledge graph, so that the critical path can be quickly determined based on the prediction results.

[0070] In a feasible implementation, step S22 may include:

[0071] Step S221: Obtain a task dependency topology graph of the subtasks in the cross-domain computing task from the knowledge graph, and determine the predicted start time and predicted end time of the subtasks.

[0072] It should be noted that the above-mentioned task dependency topology diagram refers to a schematic diagram of all dependency relationships between subtasks, the above-mentioned predicted start time indicates the time when the subtask starts to execute, and the above-mentioned predicted end time indicates the time when the subtask ends to execute.

[0073] Step S222: predicting the execution completion time of the task execution path included in the cross-domain computing task based on the task dependency topology graph, the predicted start time, and the predicted end time.

[0074] In a specific implementation, a graph database query mechanism can be used to reversely traverse all task dependencies starting from the specified end node in the above task dependency topology graph, so as to find all upstream tasks involved and construct all possible paths extending forward from the end node (that is, the task execution path included in the above cross-domain computing task), and then predict the execution completion time of each possible path based on the predicted start time and predicted end time of the above subtask.

[0075] Step S223: According to the prediction result, the path in the task execution path that takes the longest time to complete the subtask is determined as the critical path.

[0076] It should be understood that the time required to complete the subtasks of all the possible paths mentioned above can be arranged in descending order of the latest completion time, and then the path with the longest completion time can be selected as the critical path.

[0077] In a feasible implementation, step S221 may include:

[0078] Step S2211: If the subtask belongs to the first node, time prediction is performed on the subtask using a first prediction principle to obtain a predicted start time and a predicted end time.

[0079] Step S2212: If the subtask belongs to a non-first node, time prediction is performed on the subtask using the second prediction principle to obtain a predicted start time and a predicted end time.

[0080] It should be understood that the predicted start and end times of subtasks depend on real-time task execution analysis and historical data analysis. This process is completed regularly through batch processing tasks, which can be divided into primary and non-primary nodes for analysis. The first prediction principle mentioned above applies to primary nodes, and the second prediction principle applies to non-primary nodes.

[0081] In the specific implementation, you can refer to Figure 4 , Figure 4 This is the first node prediction flow chart of the abnormal node positioning method of this application. Figure 4As shown, for the first node, if the batch task has not yet started, you can determine whether the scheduled time has been reached: if it has, the predicted start time is the current time; otherwise, the predicted start time is the scheduled time of the node, and the predicted end time is the predicted start time plus the historical execution period. If the batch task is running, you can check the difference between the current time and the task start time: if the current time exceeds the historical execution period, the predicted start time is the actual start time of the node, and the predicted end time is the current time; otherwise, the predicted start time is the actual start time of the node, and the predicted end time is the start time plus the historical execution period. If the batch task has been completed, the actual start time can be used as the predicted start time, and the actual end time can be used as the predicted end time.

[0082] Accordingly, you can refer to Figure 5 , Figure 5 This is the non-first node prediction flow chart of the abnormal node positioning method of this application. For non-first nodes, Figure 5 Can be built on Figure 4 Based on the first node prediction flow chart shown, non-first node tasks are analyzed according to the task dependency tree structure, and the prediction time is traversed one by one along the dependency relationship. For each node, the current prediction end time can be compared with the prediction end time of other nodes on the dependency tree (corresponding to Figure 5 The other prediction end times in , the longer one is taken as the final prediction value.

[0083] This embodiment constructs a knowledge graph corresponding to the cross-domain computing task based on the encapsulated data; obtains a task dependency topology graph of the subtasks in the cross-domain computing task from the knowledge graph; if the subtask belongs to the first node, the subtask is time-predicted according to a first prediction principle to obtain a predicted start time and a predicted end time; if the subtask belongs to a non-first node, the subtask is time-predicted according to a second prediction principle to obtain a predicted start time and a predicted end time; the execution completion time of the task execution path included in the cross-domain computing task is predicted based on the task dependency topology graph, the predicted start time and the predicted end time; and the path in the task execution path that takes the longest time to complete the subtask is determined as the critical path according to the prediction result. The above method of this embodiment can accurately and comprehensively find all task execution paths in the cross-domain computing task by constructing a knowledge graph corresponding to the cross-domain computing task; at the same time, it predicts the execution completion time of the task execution path based on the task dependency relationship of each subtask in the cross-domain computing task as well as the predicted start time and predicted end time, thereby reducing the risk of false positives and missed positives, thereby accurately finding the critical path in the task execution path that takes the longest time to complete the subtask, and improving the accuracy of subsequent node analysis of the critical path.

[0084] refer to Figure 6 , Figure 6 This is a flowchart of the third embodiment of the abnormal node locating method of the present application.

[0085] In a feasible implementation, step S1 may include:

[0086] Step S11: Determine subtask attributes and subtask dependencies corresponding to the cross-domain computing task based on the target data.

[0087] It should be noted that the above-mentioned subtask attributes may include but are not limited to the business level at which the subtask is located, the type of operation and maintenance data, data characteristic attributes (such as statistical caliber, operation frequency), statistical attributes (such as start time, execution duration, resource utilization rate), etc. The above-mentioned subtask dependencies are stored on the relationship edge and may include but are not limited to sequential dependencies, conditional dependencies, parallel dependencies, etc.

[0088] In a specific implementation, since the target data is raw, standardized data collected from multiple domain sources, it may contain unnecessary data redundancy. Therefore, the target data can first be cleaned, deduplicated, and formatted, and then the subtask attributes and subtask dependencies can be determined from the processed data.

[0089] In particular, the above-mentioned target data can be collected through the data acquisition module. The data acquisition module provides the ability to obtain cloud computing task operation and maintenance monitoring data from different business layers in real time. The module divides the operation and maintenance objects and data involved in the flow of cross-domain computing tasks into multiple subject domains to standardize data collection for different operation and maintenance scenarios and data types, thereby improving the monitoring and operation and maintenance efficiency of cross-domain computing tasks. For example, the operation and maintenance object domain covers data oriented to the basic information of operation and maintenance objects, such as infrastructure objects, platform objects, business data objects, and software system objects; the operation and maintenance data domain covers data oriented to collection devices, such as indicators, logs, events, and tracking. The data acquisition module can unify the data transmission protocol based on the streaming message queue, and automatically identify and collect real-time data of key indicators and status information sent by the operation and maintenance objects according to the message source and data format, such as the running time of the task, resource utilization, execution status, etc.

[0090] Step S12: Encapsulating the target data based on the subtask attributes and the subtask dependencies to obtain encapsulated data, wherein the encapsulated data is in the form of triples.

[0091] It should be understood that heterogeneous data from different sources can be uniformly converted into the structured data format required by the business based on subtask attributes and subtask dependencies, and represented in the form of knowledge triples consisting of a head node, a relationship, and a tail node. Specifically, the head node stores higher-level subtask operation and maintenance data, indicating the task process to which the subtask belongs and the attributes of the subtask (such as running time and resource utilization); the tail node stores lower-level subtask operation and maintenance data; there is a dependency relationship between the head and tail nodes, and the relationship is used to indicate the connection between the head and tail nodes. For example, the head node task can only be executed after the tail node subtask has completed its operation.

[0092] For example, suppose in a cross-domain computing task, a virtual machine (the infrastructure domain in the operation and maintenance object domain) may report the log data of a data processing subtask. This subtask acts as the head node and stores the task's start time, execution duration, resource utilization and other attributes; then, the subtask running on another log data analysis platform acts as the tail node, which is triggered after the data processing is completed and outputs the indicator data of the previous task. The encapsulated data involved in this cross-domain computing task can be expressed as: (data processing task, {"business level":"infrastructure domain","data type":"log","running frequency":"1 day / time","start time":"2024-08-13 10:00","execution duration":"30 minutes","resource utilization rate":"75%"})-(sequential dependency)->("log analysis task", {"business level":"platform domain","data type":"indicator","running frequency":"1 hour / time","start time":"2024-08-13 10:30","execution duration":"15 minutes","resource utilization rate":"60%"}).

[0093] In addition, after obtaining the above-mentioned encapsulated data, it can be stored through the data aggregation and storage module. Specifically, the data aggregation and storage module uses the MPP (Massively Parallel Processing) architecture column storage database to provide the ability to efficiently store and read cross-domain computing task operation and maintenance task attributes, calculation and prediction indicator storage. While saving the attributes of the subtask nodes, this module also records various indicator data during the task operation process, such as message flow rate, traffic flow rate, key process status, back pressure, etc. for monitoring task status and resource usage; records the start time, completion time, predicted earliest and latest start time and end time of the task timeliness; records relevant information such as the availability of task scheduling tools and resource cluster status, etc.

[0094] In a feasible implementation, step S3 may include:

[0095] Step S31: performing node analysis on the critical path according to the time parameters of each node in the critical path, and calculating the criticality score of each node according to the analysis results.

[0096] It should be noted that the above time parameters may include the actual end time of the node, the predicted end time of the node, the end time of the upstream task, the time when the task should start, etc.

[0097] In a specific implementation, the criticality score of each node in the critical path can be determined based on the difference between the actual end time and the predicted end time of each node, as well as the impact of the upstream task end time on the task start time. For example, the criticality score of each node can be calculated based on the following formula:

[0098] C=|(τ actEnd -τ expectEnd )-|τ prevEnd -τ expectStart ||+Mp;

[0099] Among them, C represents the criticality score, τ actEnd is the actual end time of the node (if the actual end time of the node is not obtained during batch processing, the predicted end time is used as the actual end time), τ expectEnd is the predicted end time of the node, τ prevEnd is the actual end time of the latest node in the upstream sequential dependency (if the current node is the first node, the actual end time of the current node is taken), τ expectStart is the time when the task should start (determined based on the average start time of historical task run records). M is a maximum value, and p is a Boolean value (1 when the task runs abnormally or resources exceed the limit, and 0 when the task runs normally and resource utilization is normal).

[0100] Step S32: determining the node corresponding to the highest score among the criticality scores as the abnormal node of the cross-domain computing task, wherein the criticality score is used to indicate the degree of time influence of the node on the critical path.

[0101] In a specific implementation, all nodes in the critical path can be sorted in descending order according to their criticality scores. The node with the highest score has the greatest impact on the time of the entire cross-domain computing task flow, and can therefore be determined as an abnormal node of the cross-domain computing task.

[0102] In particular, the operation and maintenance data in the process of cross-domain computing task flow can be analyzed for anomalies and predicted for weaknesses based on the task anomaly analysis and prediction module. When the abnormal node is identified, the module can combine the real-time and historical data of the bottleneck (i.e., the abnormal node) to conduct root cause analysis, and use rule-based analysis methods to locate the specific cause of the bottleneck. For example, the resource bottleneck problem in the IaaS layer may be caused by insufficient computing resources, such as excessive use of CPU and memory. The module will confirm the problem by monitoring parameters such as CPU utilization, memory utilization, and network bandwidth. Hardware failures such as server downtime or disk failure will also be detected through server status and hardware health monitoring parameters. At the PaaS layer, service performance problems may be caused by too long middleware response time or insufficient service instances. At this time, the problem can be identified by monitoring the service response time and the number of service requests. If it is a service dependency failure (such as an abnormal registration center), it is necessary to analyze the service dependency topology and service. Service availability parameters; Data transmission bottlenecks may occur at the DaaS layer, such as insufficient network bandwidth or excessive data conversion overhead. In this case, data transmission rate and bandwidth utilization will be used for detection. Data integrity issues involve data loss or damage during transmission, which need to be identified through data integrity checks and data packet loss rate. At the SaaS layer, application service failures may manifest as increased application service error rates or long response times. User behavior anomalies can be analyzed by monitoring application service response time and error rates. For example, if frequent request timeouts lead to service overload, the user request frequency and number of timed-out requests need to be monitored to locate the problem.

[0103] Furthermore, based on predicted task start and completion data, the Task Anomaly Analysis and Prediction module dynamically predicts whether there will be task backlogs or execution delays within the predicted timeframe. Once a critical bottleneck or task delay is predicted, the Task Anomaly Analysis and Prediction module provides data updates to the Critical Path and Task Discovery modules to update the critical path. This module also initiates a root cause analysis process to provide optimization recommendations for cross-domain task computing. The results of the anomaly analysis and vulnerability prediction analysis are stored in the cross-domain computing task flow data warehouse, providing a data access interface for external monitoring and alarm analysis devices to access.

[0104] In addition, you can refer to Figure 7 , Figure 7 This is a diagram of the task end-to-end flow visualization module for the abnormal node location method of this application. The task end-to-end flow visualization module can integrate multiple visualization tools and analysis methods based on the abnormality monitoring and prediction module to help operation and maintenance personnel fully understand the execution status of the task and identify potential bottlenecks and abnormalities. Figure 7As shown, the end-to-end task flow visualization module first generates a report listing abnormal subtasks within specific cross-domain computing tasks based on anomaly analysis results at different business layers. Combining this with the knowledge graph of cross-domain cloud computing task operation and maintenance data, it obtains a task dependency topology diagram of the task execution path. By analyzing node attributes and dependencies, it demarcates anomalies end-to-end. Finally, the module provides a tracking diagram for specific task nodes, displaying the status of each link in the subtask flow, such as start time, end time, and resource usage, helping operations personnel clearly understand the root cause of the problem.

[0105] This embodiment determines the subtask attributes and subtask dependencies corresponding to the cross-domain computing task based on the target data; encapsulates the target data based on the subtask attributes and the subtask dependencies to obtain encapsulated data, which is in the form of a triple; performs node analysis on the critical path based on the time parameters of each node in the critical path, and calculates the criticality score of each node based on the analysis result; determines the node corresponding to the highest score in the criticality score as the abnormal node of the cross-domain computing task, and the criticality score is used to represent the degree of influence of the node on the time of the critical path. The above method of this embodiment encapsulates the target data in the form of a triple based on the subtask attributes and subtask dependencies corresponding to the cross-domain computing task, thereby unifying the data format and better capturing and managing the complex dependencies between the various subtasks in the cross-domain computing task; at the same time, by calculating the criticality score of each node in the critical path to quantify the degree of influence of each node on the time of the critical path, it is possible to more clearly locate the abnormal node that has the greatest impact on the flow time of the entire cross-domain computing task.

[0106] In particular, in order to more intuitively demonstrate the overall technical solution of this application, this application provides a solution example, and the specific steps of the solution example are referred to as follows.

[0107] The first step is to monitor business-layer data. For a cross-domain task A, business data from the IaaS, PaaS, DaaS, and SaaS layers is collected. The IaaS layer collects data from the "Start VM Instance" and "Initialize VM" subtasks; the PaaS layer collects data from the "Container Orchestration" and "Microservice Construction" subtasks; the DaaS layer collects data from the "Data Conversion" and "Data Reporting" jobs; and the SaaS layer collects data from the "Monitor Transaction Request" subtasks. For example, the properties of the "Microservice Registration" node are {"Business Layer":"PaaS","Data Type":"Log","Start Time":"2024-08-18 20:00:00","End Time":"2024-08-18 20:30:00","Resource Utilization":"75%"}.

[0108] The second step is to standardize data collection. The data collection module uses a message queue to collect real-time data sent by the monitoring data generation device. It is uniformly encapsulated in JSON format. The attribute values include the subtask name, subtask data type, and node-specific attributes (such as resource utilization rate and execution status), and then provided to the data encapsulation module.

[0109] The third step is to encapsulate data into nodes. The data encapsulation module parses the JSON data, extracts the dependencies between subtasks based on attribute fields, and encapsulates the subtask attributes and indicator characteristics into nodes. For example, the content of the "VM Instance Start" node includes: {"Node Name":"VM Instance Start","Business Tier":"IaaS","Node Attributes":{"Start Time":"2024-08-18 20:00:00","End Time":"2024-08-18 20:15:00","Resource Utilization":"65%","Status":"Success"}}. The node also stores the dependencies associated with it, such as the dependency with the "Initialize VM" subtask: {"Dependency Node":"Initialize VM","Dependency Type":"Sequential Dependency"}.

[0110] The fourth step is cross-domain computing task flow data storage. After the data encapsulation module processes the data, the encapsulated data is transmitted through the data storage module and stored in the cross-domain computing task flow data warehouse. In actual operation, node attributes and related indicators are stored in a distributed MPP database to support efficient ad hoc data query and analysis. The database tables are partitioned and indexed by fields such as business level, task flow, and node attributes to ensure efficient data query.

[0111] The fifth step is to build a task knowledge graph. In order to efficiently store the dependencies between nodes, the task knowledge graph can be constructed through the data processed by the data encapsulation module. Assuming that the cross-domain task A includes node A virtual machine instance startup (IaaS), node B initialization virtual machine (IaaS), node C container orchestration (PaaS), node D microservice construction (PaaS), node E microservice registration (PaaS), node F data conversion job (DaaS), node G listening transaction request (SaaS), node H data reporting (DaaS), the system will build the task knowledge graph through subtask attributes and dependencies. In the graph database, the dependencies between these nodes will be stored and presented in the form of a graph, for example: node A->node B (sequential dependency), node C->node D (sequential dependency), node E->node G (sequential dependency), node F->node H (sequential dependency).

[0112] Step 6: Discover the critical path and subtasks. During the execution of cross-domain task A, the critical path and task discovery module can discover possible bottleneck paths (i.e., critical paths) through historical and real-time data analysis. For example, in the above task process, the "microservice construction" subtask is delayed. Subsequent impact analysis reveals that the completion time of the entire task is subject to the critical path composed of this node, and the critical path is ultimately determined to be A->B->C->D->E->G. The critical path and task discovery module can then mark this node as a bottleneck node (i.e., an abnormal node) and prompt that it requires special attention.

[0113] Step 7: Task Anomaly Analysis and Prediction. When the system detects that the bottleneck node, "Microservice Build," is taking too long to execute or consuming too many resources, the Task Anomaly Analysis and Prediction module conducts a root cause analysis. By combining real-time metrics and historical data, the system infers that the registration center is overloaded, preventing service registration, and that this resource bottleneck is causing delays in dependent tasks. The Task Anomaly Analysis and Prediction module also issues alerts predicting possible future anomalies and provides optimization suggestions, such as allocating more resources or adjusting the order of task execution.

[0114] Step 8: Visualize the end-to-end flow of tasks. If an anomaly is detected during task execution, the module will generate a task list report to show which subtasks have problems. The task list report can include detailed information such as the task name, the data interface to which it belongs, and the data process stage to which it belongs. Operation and maintenance personnel can also use the data flow task dependency topology diagram to show the data flow direction of a specific part of the task during the entire task flow process. For details, please refer to Figure 8 , Figure 8 This is an example diagram of the key task data flow topology of the abnormal node positioning method in this application. Operation and maintenance personnel can use the following Figure 8 The topology diagram shown determines that the critical path is starting virtual machine instances, initialization, container orchestration, microservice construction, microservice registration, listening to transaction requests, completion, and quickly locating bottleneck nodes.

[0115] In addition, the task end-to-end flow visualization module can also provide a task tracking diagram to show the propagation path of the task between different business layers, and provide detailed attribute display function of subtask nodes to help operation and maintenance personnel understand the order and duration of subtask execution, as well as the dependencies between different domains. Figure 9 , Figure 9 This is the task tracking diagram for the abnormal node location method in this application. By viewing the task tracking diagram of the microservice registration node, operations and maintenance personnel can deeply explore possible bottlenecks during task execution, intuitively see the task execution path, and analyze the impact range and propagation chain of the bottleneck task.

[0116] Reference Figure 10, Figure 10 This is a structural block diagram of the first embodiment of the abnormal node locating device of this application.

[0117] like Figure 10 As shown, the abnormal node locating device proposed in the embodiment of the present application includes:

[0118] The data encapsulation module 101 is used to encapsulate the target data to obtain encapsulated data, wherein the target data is the operation and maintenance data included in the cross-domain computing task in the big data cloud platform;

[0119] A critical path determining module 102 is configured to determine a critical path based on the encapsulated data, wherein the critical path is a path that takes the longest time to complete a subtask in the cross-domain computing task;

[0120] The abnormal node locating module 103 is configured to perform node analysis on the critical path and locate abnormal nodes of the cross-domain computing task according to the analysis results.

[0121] This embodiment encapsulates the target data to obtain encapsulated data, wherein the target data is the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform; determines the critical path based on the encapsulated data, wherein the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task; performs node analysis on the critical path, and locates the abnormal nodes of the cross-domain computing task based on the analysis results. After the above method of this embodiment uniformly encapsulates the operation and maintenance data contained in the cross-domain computing task in the big data cloud platform, it determines the critical path of the cross-domain computing task based on the encapsulated data. Since the critical path is the path with the longest time required to complete the subtasks in the cross-domain computing task, that is, the critical path contains the most critical task that determines the flow cycle of the cross-domain computing task, this embodiment analyzes each node in the critical path, thereby being able to more clearly locate the abnormal node that has the greatest impact on the flow time of the entire cross-domain computing task.

[0122] Based on the first embodiment of the abnormal node locating device of the present application, a second embodiment of the abnormal node locating device of the present application is proposed.

[0123] In this embodiment, the critical path determination module 102 is also used to construct a knowledge graph corresponding to the cross-domain computing task based on the encapsulated data; predict the execution completion time of the task execution path contained in the cross-domain computing task based on the knowledge graph, and determine the critical path according to the prediction result.

[0124] Furthermore, the critical path determination module 102 is also used to obtain the task dependency topology graph of the subtasks in the cross-domain computing task from the knowledge graph, and determine the predicted start time and predicted end time of the subtask; predict the execution completion time of the task execution path contained in the cross-domain computing task based on the task dependency topology graph, the predicted start time and the predicted end time; and determine the path in the task execution path that takes the longest time to complete the subtask as the critical path based on the prediction result.

[0125] Furthermore, the critical path determination module 102 is also used to, if the subtask belongs to the first node, perform time prediction on the subtask using the first prediction principle to obtain a predicted start time and a predicted end time; if the subtask belongs to a non-first node, perform time prediction on the subtask using the second prediction principle to obtain a predicted start time and a predicted end time.

[0126] Furthermore, the data encapsulation module 101 is also used to determine the subtask attributes and subtask dependencies corresponding to the cross-domain computing task based on the target data; encapsulate the target data based on the subtask attributes and the subtask dependencies to obtain encapsulated data, and the encapsulated data is in the form of triples.

[0127] Furthermore, the abnormal node positioning module 103 is also used to perform node analysis on the critical path according to the time parameters of each node in the critical path, and calculate the criticality score of each node according to the analysis results; the node corresponding to the highest score in the criticality score is determined as the abnormal node of the cross-domain computing task, and the criticality score is used to indicate the degree of time influence of the node on the critical path.

[0128] Other embodiments or specific implementations of the abnormal node locating device of the present application can refer to the above-mentioned method embodiments and will not be repeated here.

[0129] The present application provides an abnormal node locating device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the abnormal node locating method in the above-mentioned embodiment 1.

[0130] Reference below Figure 11, which shows a schematic diagram of the structure of an abnormal node location device suitable for implementing an embodiment of the present application. The abnormal node location device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 11 The abnormal node locating device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0131] like Figure 11 As shown, the abnormal node location device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory 1002 or programs loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the abnormal node location device. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems may be connected to the input / output interface 1006: an input device 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the abnormal node location device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an abnormal node location device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems can be implemented or have alternatively.

[0132] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising 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 via a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are performed.

[0133] The abnormal node locating device provided in this application utilizes the abnormal node locating method described in the aforementioned embodiment to address the technical issues surrounding abnormal node locating. Compared to related technologies, the beneficial effects of the abnormal node locating device provided in this application are the same as those of the abnormal node locating method described in the aforementioned embodiment. Other technical features of the abnormal node locating device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0134] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0135] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0136] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer program) stored thereon, and the computer-readable program instructions are used to execute the abnormal node location method in the above embodiment.

[0137] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an 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 embodiment, the 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 or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0138] The computer-readable storage medium may be included in the abnormal node locating device, or may exist independently without being assembled into the abnormal node locating device.

[0139] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the abnormal node locating device, the abnormal node locating device can write computer program codes for performing the operations of the present application in one or more programming languages or a combination thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, and C++; and also include conventional procedural programming languages such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (for example, connected through the Internet using an Internet service provider).

[0140] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. 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 by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0141] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0142] The computer-readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described abnormal node location method. This computer-readable storage medium can address the technical problem of related art in locating abnormal nodes in cross-domain computing tasks. Compared with related art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the abnormal node location method provided in the above-described embodiment, and are not further elaborated here.

[0143] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned abnormal node locating method when executed by a processor.

[0144] The computer program product provided in this application can solve the technical problem of abnormal node location. Compared with the related art, the beneficial effects of the computer program product provided in this application are the same as those of the abnormal node location method provided in the above embodiment, and will not be repeated here.

[0145] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for locating abnormal nodes, characterized in that: The method comprises the following steps: Encapsulating target data to obtain encapsulated data, wherein the target data is operation and maintenance data included in a cross-domain computing task in a big data cloud platform; Determine a critical path based on the encapsulated data, where the critical path is the path that takes the longest time to complete a subtask in the cross-domain computing task; Perform node analysis on the critical path, and locate abnormal nodes of the cross-domain computing task based on the analysis results.

2. The abnormal node location method according to claim 1, characterized in that: The step of determining the critical path based on the package data includes: Constructing a knowledge graph corresponding to the cross-domain computing task according to the encapsulated data; Based on the knowledge graph, the execution completion time of the task execution path included in the cross-domain computing task is predicted, and the critical path is determined according to the prediction result.

3. The abnormal node location method according to claim 2, characterized in that: The step of predicting the execution completion time of the task execution path included in the cross-domain computing task based on the knowledge graph and determining the critical path according to the prediction result includes: Obtaining a task dependency topology graph of subtasks in the cross-domain computing task from the knowledge graph, and determining a predicted start time and a predicted end time of the subtasks; Predicting the execution completion time of the task execution path included in the cross-domain computing task based on the task dependency topology graph, the predicted start time, and the predicted end time; According to the prediction result, the path in the task execution path that takes the longest time to complete the subtask is determined as the critical path.

4. The abnormal node location method according to claim 3, characterized in that: The step of determining the predicted start time and predicted end time of the subtask includes: If the subtask belongs to the first node, the time of the subtask is predicted using the first prediction principle to obtain the predicted start time and the predicted end time; If the subtask belongs to a non-first node, a time prediction is performed on the subtask using a second prediction principle to obtain a predicted start time and a predicted end time.

5. The abnormal node location method according to claim 1, wherein: The step of encapsulating the target data to obtain encapsulated data includes: Determine the subtask attributes and subtask dependencies corresponding to the cross-domain computing task based on the target data; The target data is encapsulated based on the subtask attributes and the subtask dependencies to obtain encapsulated data, where the encapsulated data is in a triple form.

6. The abnormal node location method according to claim 1, wherein: The step of performing node analysis on the critical path and locating abnormal nodes of the cross-domain computing task according to the analysis results includes: Performing node analysis on the critical path according to the time parameters of each node in the critical path, and calculating the criticality score of each node according to the analysis results; A node corresponding to a highest score among the criticality scores is determined as an abnormal node of the cross-domain computing task, wherein the criticality score is used to indicate the degree of time influence of the node on the critical path.

7. An abnormal node locating device, characterized in that: The abnormal node locating device includes: A data encapsulation module is used to encapsulate target data to obtain encapsulated data, where the target data is the operation and maintenance data included in the cross-domain computing task in the big data cloud platform; A critical path determination module, configured to determine a critical path based on the encapsulated data, wherein the critical path is a path that takes the longest time to complete a subtask in the cross-domain computing task; The abnormal node locating module is used to perform node analysis on the critical path and locate abnormal nodes of the cross-domain computing task according to the analysis results.

8. An abnormal node locating device, characterized in that: The device includes: a memory, a processor, and an abnormal node locating program stored in the memory and executable on the processor, wherein the abnormal node locating program is configured to implement the steps of the abnormal node locating method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and an abnormal node locating program is stored on the storage medium. When the abnormal node locating program is executed by a processor, the steps of the abnormal node locating method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes an abnormal node locating program, which, when executed by a processor, implements the steps of the abnormal node locating method according to any one of claims 1 to 6.

Images