A node processing method, apparatus, device, storage medium, and program product.

Abstract

This disclosure provides a node processing method applicable to the fields of cloud computing and fintech. The method includes: during application operation, monitoring the first and second dimension indicator information of N service nodes in the service chain topology used to implement the application, according to service evaluation indicators; identifying nodes to be adjusted from the N service nodes; determining a target strategy from M application service assurance strategies based on the second dimension indicator information of the node to be adjusted; and processing the node to be adjusted according to the target strategy to obtain the adjusted target service node. This disclosure also provides a node processing apparatus, device, storage medium, and program product.

Description

Technical Field

[0001] This disclosure relates to the fields of cloud computing technology and fintech, and specifically to a node processing method, apparatus, device, medium, and program product. Background Technology

[0002] In the field of cloud computing, with the popularization of containers, microservices and function computing, applications can be broken down into multiple microservices and functions for containerized deployment.

[0003] Based on this, since each microservice has unique operating characteristics, it is necessary to ensure the overall SLA (Service Level Agreement) of the application in terms of application operation, that is, to ensure the stable operation of the application.

[0004] In the process of realizing the inventive concept disclosed herein, the inventors discovered that in related technologies, there are problems such as difficulty in guaranteeing the SLA of the application and low user experience. Summary of the Invention

[0005] In view of the above problems, this disclosure provides a node processing method, apparatus, device, medium and program product.

[0006] According to a first aspect of this disclosure, a node processing method is provided, comprising: during application operation, for N service nodes in a service chain topology used to implement the application, monitoring first-dimensional indicator information and second-dimensional indicator information of the N service nodes according to a service evaluation indicator dimension, wherein the first-dimensional indicator information is used to characterize the request reception status of the service node, and the second-dimensional indicator information is used to characterize the request processing error status of the service node, and N is a positive integer greater than 1; determining a node to be adjusted from the N service nodes, wherein the node to be adjusted is a service node whose first-dimensional indicator information meets a first predetermined condition; determining a target strategy from M application service assurance strategies based on the second-dimensional indicator information of the node to be adjusted, wherein the application service assurance strategy is a strategy used to ensure the stable operation of the application by adjusting nodes, and the adjustment methods of the M nodes corresponding to the M application service assurance strategies are different, and M is a positive integer greater than 1; and processing the node to be adjusted according to the target strategy to obtain an adjusted target service node.

[0007] According to embodiments of this disclosure, processing a node to be adjusted according to a target strategy to obtain an adjusted target service node includes: when the target strategy is a node isolation strategy, determining an interaction node that interacts with the node to be adjusted from the service chain topology; calling a substitute service node corresponding to the node to be adjusted to interact with the interaction node, so as to process the request corresponding to the node to be adjusted through the substitute service node and isolate the node to be adjusted; and determining the substitute service node as the target service node.

[0008] According to embodiments of this disclosure, the above-mentioned node processing method further includes: displaying a graph of the service chain topology on a visualization interface, wherein the graph includes image identifiers of N service nodes and node call paths corresponding to the N service nodes; updating the graph based on the image identifiers of the substitute service nodes and the image identifiers of the nodes to be adjusted, to obtain an updated graph.

[0009] According to embodiments of this disclosure, processing a node to be adjusted according to a target strategy to obtain an adjusted target service node includes: when the target strategy is a node capacity adjustment strategy, predicting target indicator information of the node to be adjusted in a future target time period based on indicator information of the node to be adjusted in a first predetermined time period; and adjusting the capacity of the node to be adjusted based on the target indicator information to obtain the adjusted target service node.

[0010] According to an embodiment of this disclosure, when the target strategy is a node capacity adjustment strategy, based on the indicator information of the node to be adjusted in a first predetermined time period, the target indicator information of the node to be adjusted in a future target time period is predicted, including: identifying the node type information of the node to be adjusted; determining a target prediction model corresponding to the node type information from K trained prediction models, wherein the types of service nodes corresponding to the training samples of the K trained prediction models are all different, and K is a positive integer greater than 1; inputting the indicator information in the first predetermined time period into the target prediction model, and outputting the target indicator information in the target time period.

[0011] According to embodiments of this disclosure, training samples are obtained through the following method: obtaining sample indicator information of service nodes within I historical time periods, where I is an integer greater than 1; for the i-th historical time period within the I historical time periods, generating label information corresponding to the sample indicator information of the i-th historical time period based on the time period identifiers of the Ii historical time periods following the i-th historical time period and the sample indicator information of the Ii historical time periods, where i is a positive integer less than 1; generating training samples based on the sample indicator information of the I-1 historical time periods other than the i-th historical time period and the label information corresponding to the sample indicator information of the I-1 historical time periods.

[0012] According to embodiments of this disclosure, the first dimension indicator information includes request throughput; monitoring the first dimension indicator information and the second dimension indicator information of N service nodes includes: monitoring the request throughput of each of the N service nodes within a second predetermined time period; determining the node to be adjusted from the N service nodes includes: determining the throughput change of each of the N service nodes based on the request throughput of the N service nodes within the second predetermined time period; determining the total throughput change based on the request throughput of the N service nodes within the second predetermined time period, wherein the total throughput change is the change in the total throughput of the N service nodes within the second predetermined time period; determining the node to be adjusted from the N service nodes based on the throughput change of each of the N service nodes and the total throughput change, wherein the throughput change of the node to be adjusted has the greatest impact on the degree of change in the total throughput change.

[0013] According to embodiments of this disclosure, the second dimension indicator information includes the request processing error rate; multiple application service assurance strategies include a node isolation strategy and a node capacity adjustment strategy; based on the second dimension indicator information of the node to be adjusted, a target strategy is determined from the multiple application service assurance strategies, including: determining the node isolation strategy as the target strategy when the request processing error rate is greater than or equal to the error rate threshold; and determining the node capacity adjustment strategy as the target strategy when the request processing error rate is less than the error rate threshold.

[0014] According to embodiments of this disclosure, a service node is a function or microservice node used to implement service functions; the first dimension indicator information includes at least one of request throughput and request response time; the second dimension indicator information includes request processing error rate.

[0015] The second aspect of this disclosure provides a node processing apparatus, comprising: a monitoring module, configured to monitor, during application operation, N service nodes in a service chain topology used to implement the application, according to a service evaluation index dimension, first-dimensional index information and second-dimensional index information of the N service nodes, wherein the first-dimensional index information is used to characterize the request reception status of the service node, and the second-dimensional index information is used to characterize the request processing error status of the service node, and N is a positive integer greater than 1; a first determining module, configured to determine a node to be adjusted from the N service nodes, wherein the node to be adjusted is a service node whose first-dimensional index information satisfies a first predetermined condition; a second determining module, configured to determine a target strategy from M application service assurance strategies based on the second-dimensional index information of the node to be adjusted, wherein the application service assurance strategy is a strategy used to ensure the stable operation of the application by adjusting nodes, and the adjustment methods of the M nodes corresponding to the M application service assurance strategies are different, and M is a positive integer greater than 1; and a processing module, configured to process the node to be adjusted according to the target strategy to obtain the adjusted target service node.

[0016] A third aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0017] A fourth aspect of this disclosure also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0018] The fifth aspect of this disclosure also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0019] According to embodiments of this disclosure, by monitoring first-dimensional indicator information during application operation, nodes to be adjusted can be determined from N service nodes based on request response status. Furthermore, based on the second-dimensional indicator information indicating request processing errors in the nodes to be adjusted, a target strategy for handling these nodes can be determined. Therefore, by using a target strategy to handle nodes with abnormal conditions, dynamic handling of application anomalies is achieved, ensuring the application's SLA, preventing application anomalies while providing services to users, and improving user experience.

[0020] Furthermore, by monitoring the first and second dimension indicator information of service nodes in real time, the limitations imposed by the indicator information collection cycle can be avoided, thereby improving the system versatility of the node processing method in this embodiment.

[0021] Furthermore, by using first-dimensional and second-dimensional indicator information, the influence of other indicator information can be avoided, thus improving the accuracy of processing service nodes. Attached Figure Description

[0022] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0023] Figure 1 The diagram illustrates an application scenario of the node processing method according to an embodiment of the present disclosure.

[0024] Figure 2 A flowchart illustrating a node processing method according to an embodiment of the present disclosure is shown schematically.

[0025] Figure 3 A schematic diagram of a processing node according to an embodiment of the present disclosure is shown.

[0026] Figure 4 A schematic diagram of an outdated atlas according to an embodiment of the present disclosure is shown.

[0027] Figure 5 A schematic diagram of an updated atlas according to an embodiment of the present disclosure is shown.

[0028] Figure 6 A schematic diagram of the architecture of a prediction model according to an embodiment of the present disclosure is shown.

[0029] Figure 7 A schematic diagram of a feature processing unit in a prediction model according to an embodiment of the present disclosure is shown.

[0030] Figure 8 A flowchart illustrating a node capacity adjustment method according to an embodiment of the present disclosure is shown schematically.

[0031] Figure 9 A flowchart illustrating a model training method according to an embodiment of the present disclosure is shown schematically.

[0032] Figure 10 A schematic block diagram of a node processing apparatus according to an embodiment of the present disclosure is shown.

[0033] Figure 11 A block diagram schematically illustrates an electronic device suitable for implementing a node processing method according to an embodiment of the present disclosure. Detailed Implementation

[0034] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0035] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0036] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0037] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0038] In the technical solution disclosed herein, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0039] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0040] In one embodiment of this disclosure, the resource consumption of microservices can be analyzed by collecting conventional system performance metrics such as CPU (Central Processing Unit) and memory, and configuring metric thresholds. Alternatively, the resource consumption of microservices can be analyzed using data prediction methods. Based on this, resources can be scaled up or down according to the resource consumption of microservices to ensure the stable operation of applications.

[0041] The inventors discovered that the above method can only solve the resource consumption bottleneck problem of a single microservice, and cannot solve the problem of error handling that occurs during application runtime. For example, even if the metrics corresponding to the application resources are normal, the application data may still show a large number of errors. In this case, the application's SLA is difficult to guarantee.

[0042] In another embodiment of this disclosure, various intelligent methods can be used to perform root cause analysis or fault location in applications to handle online faults. However, there are few universally applicable root cause analysis methods, and intelligent analysis technologies are limited to fixed scenarios. Furthermore, the entire process from problem discovery and troubleshooting to problem resolution is time-consuming, which can affect the SLA of online applications.

[0043] In view of this, embodiments of this disclosure provide a node processing method, comprising: during application operation, for N service nodes in the service chain topology used to implement the application, monitoring first-dimensional indicator information and second-dimensional indicator information of the N service nodes according to the service evaluation indicator dimension, wherein the first-dimensional indicator information is used to characterize the request reception status of the service node, and the second-dimensional indicator information is used to characterize the request processing error status of the service node, and N is a positive integer greater than 1. Determining a node to be adjusted from the N service nodes, wherein the node to be adjusted is a service node whose first-dimensional indicator information meets a first predetermined condition. Based on the second-dimensional indicator information of the node to be adjusted, determining a target strategy from M application service guarantee strategies, wherein the application service guarantee strategy is a strategy used to ensure the stable operation of the application by adjusting nodes, and the adjustment methods of the M nodes corresponding to the M application service guarantee strategies are different, and M is a positive integer greater than 1. And processing the node to be adjusted according to the target strategy to obtain the adjusted target service node.

[0044] Figure 1 The diagram illustrates an application scenario of the node processing method according to an embodiment of the present disclosure.

[0045] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, and a network 104. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0046] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0047] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0048] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0049] It should be noted that the node processing method provided in this embodiment can generally be executed by server 105. Correspondingly, the node processing device provided in this embodiment can generally be located in server 105. The node processing method provided in this embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the node processing device provided in this embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0050] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0051] The following will be based on Figure 1 The described scene, through Figures 2-9 The node processing method of the disclosed embodiments will be described in detail.

[0052] Figure 2 A flowchart illustrating a node processing method according to an embodiment of the present disclosure is shown schematically.

[0053] like Figure 2 As shown, the node processing method in this embodiment includes operations S210 to S240.

[0054] During operation S210, during application execution, for N service nodes in the service chain topology used to implement the application, according to the service evaluation index dimension, the first dimension index information and the second dimension index information of the N service nodes are monitored. The first dimension index information is used to characterize the request reception status of the service node, and the second dimension index information is used to characterize the request processing error status of the service node. N is a positive integer greater than 1.

[0055] In operation S220, a node to be adjusted is determined from N service nodes, wherein the node to be adjusted is a service node whose first dimension indicator information meets the first predetermined condition.

[0056] In operation S230, based on the second-dimensional indicator information of the node to be adjusted, a target strategy is determined from M application service assurance strategies. The application service assurance strategy is a strategy used to ensure the stable operation of the application by adjusting the node. The adjustment methods of the M nodes corresponding to the M application service assurance strategies are different, and M is a positive integer greater than 1.

[0057] In operation S240, the node to be adjusted is processed according to the target strategy to obtain the adjusted target service node.

[0058] According to embodiments of this disclosure, a service node is a function or microservice node used to implement service functionality. N service nodes can invoke each other according to a node invocation path. The node invocation path can be the request sending path of a service node. For example, a service node can invoke another node by sending a request to it. The type of request can include data retrieval requests and data write requests, etc.

[0059] In one embodiment of this disclosure, a call chain analysis tool can be used to analyze the business links used to implement the application, identifying the node call paths between service nodes in the business links. Then, based on the service nodes and node call paths in the business links, a service chain topology is generated. In another embodiment of this disclosure, the service chain topology can also be constructed by writing code based on specific requirements.

[0060] According to embodiments of this disclosure, the service evaluation indicator dimension can be a dimension used to evaluate the service quality of a node. Correspondingly, the first dimension indicator information and the second dimension indicator information can be key indicators used to evaluate the service quality of a node, namely, the SLI (Services Level Indicator) indicator.

[0061] Specifically, service evaluation metrics can include throughput, response time, and error rate. Correspondingly, the first dimension metric information includes at least one of request throughput and request response time. The second dimension metric information includes the request processing error rate. By using request throughput, request response time, and request processing error rate to describe the application's operational status from multiple key dimensions, the stability of the application can be accurately determined. This ensures the overall application's SLO (Service Level Objective). The SLO specifies the expected state of service functionality. For example, the SLO can be described as: 99% of nodes achieving an average QPS (Queries Per Second) greater than 100 k / s, 99% of nodes achieving a response latency of less than 500 ms, and the application's corresponding bandwidth greater than 200 MB / s per minute.

[0062] According to embodiments of this disclosure, during application operation, the first-dimensional indicator information and the second-dimensional indicator information of N service nodes can be monitored in real time by using target routines, according to the service evaluation indicator dimension.

[0063] For example, by calling the target routine, the entry and exit times of the request are recorded, and the request response time is determined based on the time difference between the entry and exit times.

[0064] For example, by invoking a target routine to attach it to a hook on a network device, or by tracing system calls, the number of packets or requests within a certain time period can be determined, and the request throughput can be determined based on the number of packets and requests. This time period can be set according to requirements, and this disclosure does not limit it.

[0065] For example, by calling the target routine, the return values ​​of system calls or network requests can be traced and inspected. If an error code is detected, an error counter is incremented. Based on the value of the error counter, the request processing error rate can be determined.

[0066] It should be noted that this disclosure does not limit the implementation method of the target routine. For example, technologies that can implement the target routine of this disclosure include eBPF (extended Berkeley Packet Filter) technology. Taking eBPF as an example, eBPF is a kernel technology that allows programs to run in the Linux system kernel without modifying the kernel source code or loading kernel modules. This technology originated from BPF (Berkeley Packet Filter) technology and was mainly used for network packet filtering. However, with the development of technology, eBPF has been extended to performance monitoring, security, networking, and many other application scenarios.

[0067] According to embodiments of this disclosure, the node to be adjusted is a service node that affects the stability of application operation. The first predetermined condition may be set based on at least one of a predetermined throughput threshold and a predetermined request response time.

[0068] In the embodiments of this disclosure, nodes to be adjusted that meet the first predetermined conditions can be selected from N service nodes according to the first dimension indicator information.

[0069] For example, if the request throughput of the node to be adjusted is less than a predetermined throughput threshold, it can be determined that the first dimension indicator information of the service node meets the first predetermined condition. The predetermined throughput threshold can be 50,000, 100,000, or 150,000, etc., and this disclosure does not limit it.

[0070] For example, if the request response duration of the node to be adjusted is longer than the predetermined request response duration, it can be determined that the first dimension indicator information of the service node meets the first predetermined condition. The predetermined request response duration can be 90 ms, 100 ms, or 110 ms, etc., and this disclosure does not limit it.

[0071] For example, in at least one of the following situations, it can be determined that the first dimension indicator information of the service node does not meet the first predetermined condition: the request throughput of the node to be adjusted is greater than or equal to a predetermined throughput threshold, and the request response time of the node to be adjusted is less than or equal to a predetermined request response time.

[0072] According to embodiments of this disclosure, the application service assurance strategy may include at least one of a node capacity adjustment strategy and a node isolation strategy. The node capacity adjustment strategy may be a strategy that ensures stable application operation by adjusting the capacity of the node to be adjusted. The node isolation strategy may be a strategy that ensures stable application operation by isolating the node to be adjusted.

[0073] In the embodiments of this disclosure, if the first dimension indicator information meets the first predetermined condition, it can be determined that the service node is experiencing abnormal conditions such as request congestion, which affects the stable operation of the application. Based on this, by using an application service guarantee strategy to process the node, problems such as request congestion of the service node can be resolved, ensuring the stable operation of the application.

[0074] Based on this, and using the second-dimensional indicator information, the request processing error status of the node to be adjusted can be determined. Different application service assurance strategies can be used to handle the node to be adjusted depending on its specific condition.

[0075] For example, the node to be adjusted can be scaled up according to a node capacity adjustment strategy. Another example is that the node to be adjusted can be isolated according to a node isolation strategy, and a substitute node corresponding to the node to be adjusted can be used to handle the requests for that node.

[0076] In embodiments of this disclosure, the resources of the service node can be adaptively adjusted to balance request response time and request throughput, so that the first dimension indicator information of the service node does not meet the first predetermined condition.

[0077] According to embodiments of this disclosure, by monitoring first-dimensional indicator information during application operation, nodes to be adjusted can be determined from N service nodes based on request response status. Furthermore, based on the second-dimensional indicator information indicating request processing errors in the nodes to be adjusted, a target strategy for handling these nodes can be determined. Therefore, by using a target strategy to handle nodes with abnormal conditions, dynamic handling of application anomalies is achieved, ensuring the application's SLA, preventing application anomalies while providing services to users, and improving user experience.

[0078] Furthermore, by monitoring the first and second dimension indicator information of service nodes in real time, the limitations imposed by the indicator information collection cycle can be avoided, thereby improving the system versatility of the node processing method in this embodiment.

[0079] Furthermore, by using first-dimensional and second-dimensional indicator information, the influence of other indicator information can be avoided, thus improving the accuracy of processing service nodes.

[0080] According to embodiments of this disclosure, a node to be adjusted can be determined from N service nodes based on the amount of change in request response time. For example, a service node can be determined as a node to be adjusted if the threshold for the amount of change in request response time of a service node is greater than a predetermined threshold for the amount of change in response time. For example, the amount of change in request response time can be the increase in request response time. The higher the increase, the longer the service node takes to respond to the request.

[0081] According to embodiments of this disclosure, the first dimension indicator information includes request throughput. Monitoring the first dimension indicator information and the second dimension indicator information of N service nodes includes: monitoring the request throughput of each of the N service nodes within a second predetermined time period. Determining the node to be adjusted from the N service nodes includes: determining the throughput change of each of the N service nodes based on the request throughput of the N service nodes within the second predetermined time period; determining the total throughput change based on the request throughput of the N service nodes within the second predetermined time period, wherein the total throughput change is the change in the total throughput of the N service nodes within the second predetermined time period; and determining the node to be adjusted from the N service nodes based on the throughput change of each of the N service nodes and the total throughput change, wherein the throughput change of the node to be adjusted has the greatest impact on the degree of change in the total throughput change.

[0082] According to embodiments of this disclosure, the aforementioned second predetermined time period can be determined based on the historical period in which the change in the request throughput of the service node is greatest, or it can be set according to demand.

[0083] According to embodiments of this disclosure, for each of the N service nodes, the throughput change of each service node can be monitored over a second predetermined time period. The throughput changes of the N service nodes are then summed to obtain the total throughput change of the N service nodes.

[0084] Based on this, variance or standard deviation can be calculated from the throughput changes of each of the N service nodes and the total throughput change. Then, from the throughput changes of each of the N service nodes, the target throughput change that has the greatest impact on the total throughput change is identified. Thus, the service node corresponding to the target throughput change is identified as the node to be adjusted. It should be noted that this disclosure is not limited to variance or standard deviation; any method that achieves the concept of this disclosure is acceptable.

[0085] According to embodiments of this disclosure, based on the change in throughput and the change in total throughput, the service node whose change in throughput has the greatest impact on the change in total throughput is identified from N service nodes as the node to be adjusted, thereby improving the accuracy of identifying service nodes that require adjustment.

[0086] According to the embodiments of this disclosure, when the overall service chain request throughput decreases, the request response time of each service node can be comprehensively analyzed to determine the balance range of request throughput and request response time of each service node, ensuring that there is a suitable request response time even when the request throughput is high, which is beneficial to ensure that the first dimension indicator information of the service node does not meet the first predetermined condition.

[0087] According to embodiments of this disclosure, the second-dimensional indicator information includes the request processing error rate. Multiple application service assurance strategies include a node isolation strategy and a node capacity adjustment strategy. Based on the second-dimensional indicator information of the node to be adjusted, a target strategy is determined from the multiple application service assurance strategies, including: determining the node isolation strategy as the target strategy when the request processing error rate is greater than or equal to an error rate threshold; and determining the node capacity adjustment strategy as the target strategy when the request processing error rate is less than the error rate threshold.

[0088] According to embodiments of this disclosure, if the request processing error rate is less than the error rate threshold, it can be determined that the node to be adjusted has no abnormal conditions. By determining the node capacity adjustment strategy as the target strategy, the stable operation of the application service can be guaranteed. For example, the error rate threshold can be 0.00005%, 0.0001%, or 0.00015%, and this disclosure does not limit it.

[0089] If the request processing error rate is greater than or equal to the error rate threshold, it can be determined that the node to be adjusted is in an abnormal state. Even if the capacity of the node to be adjusted is adjusted, it is difficult to guarantee the stable operation of the application service. Therefore, it is necessary to determine the node isolation strategy as the target strategy to isolate the node to be adjusted in order to prevent the node to be adjusted from affecting the stability of the service operation.

[0090] According to embodiments of this disclosure, by determining the request processing error status of a service node based on the request processing error rate, and then selecting different application service assurance strategies to handle the node to be adjusted based on the request processing error status, the flexibility and accuracy of handling the node to be adjusted are improved, thereby ensuring the application's SLA.

[0091] According to embodiments of this disclosure, processing a node to be adjusted according to a target strategy to obtain an adjusted target service node includes: when the target strategy is a node isolation strategy, determining an interaction node that interacts with the node to be adjusted from the service chain topology; invoking a substitute service node corresponding to the node to be adjusted to interact with the interaction node, so that the request corresponding to the node to be adjusted can be processed through the substitute service node, thus isolating the node to be adjusted; and determining the substitute service node as the target service node.

[0092] According to embodiments of this disclosure, service nodes that will be affected by anomalies in the node to be adjusted can be determined based on the service chain topology. These service nodes can be interaction nodes that directly interact with the node to be adjusted. For example, these interaction nodes can include upstream and downstream nodes of the node to be adjusted in the business chain.

[0093] In the embodiments of this disclosure, the N service nodes can send requests to each other by calling data interfaces. M backup service nodes, where M is a positive integer, can be pre-configured corresponding to each service node. The functions of the backup service nodes are the same as those of the node to be adjusted under normal conditions.

[0094] For example, M substitute service nodes can also interact with other service nodes. When the target strategy is node isolation, calls to the data interface corresponding to the node to be adjusted can be replaced with calls to the data interface corresponding to the substitute service nodes. Based on this, requests sent from upstream nodes to the node to be adjusted can be sent to the substitute service node corresponding to the node to be adjusted for processing, and the substitute node will then send the processed request data to the downstream nodes, thereby achieving isolation of the node to be adjusted.

[0095] According to embodiments of this disclosure, data snapshots can be taken and the isolated node to be adjusted can be taken offline. A fault analysis notification can be generated based on the snapshot data to facilitate the repair of the node. For example, the snapshot information may include the node's storage information, parameter information, the type of error-handling request, etc.

[0096] According to embodiments of this disclosure, if the error rate of the backup service node still exceeds a predetermined error rate threshold, it can be determined that the application is abnormal. Therefore, information about requests that were erroneously handled by the backup service node can be retained, and a fault notification can be generated based on this request information to handle the application's abnormal condition.

[0097] According to embodiments of this disclosure, by identifying the interaction node that interacts with the node to be adjusted from the service chain topology, and then calling the backup service node corresponding to the node to be adjusted to interact with the interaction node, the node to be adjusted is isolated, thereby avoiding the application's SLA from being affected by the abnormal condition of the node to be adjusted, thus ensuring the application's SLA.

[0098] Figure 3 A schematic diagram of a processing node according to an embodiment of the present disclosure is shown.

[0099] like Figure 3 As shown, during application operation, a traffic switching operation 311 can be performed on service nodes 321 and 322 used to implement the application. This traffic switching operation 311 may include traffic control and load balancing. Additionally, a dynamic scaling operation 312 can be performed on a substitute service node 323 used to replace the node to be adjusted, to adjust the capacity of the substitute service node 323. A fault isolation operation 313 can be performed on the replaced service node, and the replaced service node can be identified as a faulty node 324. Data snapshots and lower limit operations can be performed on the faulty node 324, and a notification 314 can be generated to facilitate the repair of the faulty node 324.

[0100] According to embodiments of this disclosure, the node processing method further includes: displaying a graph of the service chain topology on a visual interface, wherein the graph includes image identifiers of N service nodes and node call paths corresponding to the N service nodes. Based on the image identifiers of the substitute service nodes and the image identifiers of the nodes to be adjusted, the graph is updated to obtain an updated graph.

[0101] Figure 4 A schematic diagram of an outdated atlas according to an embodiment of the present disclosure is shown.

[0102] Figure 5 A schematic diagram of an updated atlas according to an embodiment of the present disclosure is shown.

[0103] like Figure 4 and Figure 5 As shown, both the unupdated graph 400 and the updated graph 500 include service node 411, service node 412, service node 421, service node 431, and service node 441. Among them, service node 412 is a substitute service node for service node 411. Figure 4 and Figure 5 The black arrows in the middle illustrate the node call path.

[0104] In the unupdated graph 400, service nodes 411, 421, 431, and 441 interact with each other. However, after service node 411 is replaced by service node 412, the updated graph 500 shows the node call paths between service nodes 412, 421, 431, and 441. It should be noted that the above... Figure 4 and Figure 5 This is for illustrative purposes only and is not intended to limit the scope of this disclosure. In some embodiments, changes in the service chain topology due to changes in service nodes may be illustrated in other ways, which are not limited herein.

[0105] According to embodiments of this disclosure, a service chain topology graph is displayed on a visual interface, and the graph is updated based on the image identifiers of substitute service nodes and nodes to be adjusted. This allows for real-time display of changes in service nodes through the graph, facilitating timely handling of service nodes with abnormal conditions and ensuring the application's Service Level Agreement (SLA).

[0106] According to embodiments of this disclosure, processing a node to be adjusted according to a target strategy to obtain an adjusted target service node includes: when the target strategy is a node capacity adjustment strategy, predicting target indicator information of the node to be adjusted in a future target time period based on indicator information of the node to be adjusted in a first predetermined time period; and adjusting the capacity of the node to be adjusted based on the target indicator information to obtain the adjusted target service node.

[0107] According to embodiments obtained from this disclosure, the target time period can be determined based on the historical time period with the highest request volume. The first predetermined time period can be the current time period. In some embodiments, the target time period and the first predetermined time period can also be set according to demand, which will not be elaborated here. The above-mentioned indicator information includes at least one of the above-mentioned request response time, the above-mentioned request throughput, and the above-mentioned request processing error rate.

[0108] In embodiments of this disclosure, the indicator information of the node to be adjusted within a first predetermined time period can be input into a trained prediction model, which then outputs the target indicator information of the node to be adjusted within a future target time period. For example, the prediction model can be constructed based on networks such as LSTM neural networks.

[0109] In embodiments of this disclosure, the indicator information of the node to be adjusted within a first predetermined time period, the time period identifier of the first predetermined time period, and the time period identifier of the target time period can also be input into a trained prediction model to output the target indicator information of the node to be adjusted within a future target time period.

[0110] In the embodiments of this disclosure, the time difference between the first predetermined time period and the target time period can also be determined, and then the indicator information of the node to be adjusted in the first predetermined time period and the time difference are input into the trained prediction model to output the target indicator information of the node to be adjusted in the future target time period.

[0111] Figure 6 A schematic diagram of the architecture of a prediction model according to an embodiment of the present disclosure is shown.

[0112] Figure 7 A schematic diagram of a feature processing unit in a prediction model according to an embodiment of the present disclosure is shown.

[0113] like Figure 6 and Figure 7 As shown, the prediction model 600 may include connected feature processing units 610, 620, and 630. t-1 It can represent the cell state at the previous moment. t It can represent the current state of a cell. t+1 It can represent the cell state at the next moment. x t-1 It can represent the input features from the previous time step. x t It can represent the input features at the current moment. t+1 It can represent the input features at the next time step. h t-1 It can represent the output characteristics of the previous time step. t It can represent the output characteristics at the current moment. t+1 σ can represent the output characteristics at the next time step. σ can represent the activation function. tanh can represent the hyperbolic tangent function.

[0114] like Figure 7 As shown, the feature processing unit 620 may include a forget gate, an input gate, and an output gate. The forget gate determines the unit state c of the previous time step. t-1 How many are retained as the current cell state c?t The input gate determines the network's input features x at the current time step. t How much is saved to cell state c? t Output gate control unit status c t How much of the output is the current output value h of the prediction model? t The gate can be constructed based on a fully connected layer, with the input feature being a vector and the output feature being a real number vector between 0 and 1. It should be noted that the structure of feature processing unit 610 and feature processing unit 630 can be the same as or different from the structure of feature processing unit 620, and this disclosure does not impose any restrictions.

[0115] It should be noted that, Figure 6 and Figure 7 The structure of the prediction model is shown only and is not intended to limit the structure of the prediction model disclosed herein.

[0116] According to embodiments of this disclosure, the capacity of the node to be adjusted can be adjusted according to target indicator information until the indicator information obtained from the node to be adjusted does not meet the first predetermined condition.

[0117] In the embodiments of this disclosure, when the node to be adjusted is a containerized service node, its capacity can be adjusted by increasing or decreasing the number of replicas. When the node to be adjusted is a virtual machine-based service node, its capacity can be adjusted by adjusting the load balancing configuration. For example, the service node can be scaled up during periods of high request volume and scaled down during periods of low request volume. The request volume of the current period can be determined based on a predetermined request volume threshold, which is not limited herein.

[0118] According to embodiments of this disclosure, by predicting target indicator information within a target time period based on indicator information within a first predetermined time period, and then adjusting the capacity of the node to be adjusted based on the target indicator information, the accuracy of determining the capacity of the node to be adjusted is improved, the accuracy of resource allocation is improved, and the abnormal situation of application services caused by sudden changes in data traffic in future time periods can be avoided, thereby ensuring the SLA of the application.

[0119] According to embodiments of this disclosure, when the target strategy is a node capacity adjustment strategy, based on the indicator information of the node to be adjusted within a first predetermined time period, the target indicator information of the node to be adjusted within a future target time period is predicted, including: identifying the node type information of the node to be adjusted; determining a target prediction model corresponding to the node type information from K trained prediction models, wherein the types of service nodes corresponding to the training samples of the K trained prediction models are all different, and K is a positive integer greater than 1; inputting the indicator information within the first predetermined time period into the target prediction model, and outputting the target indicator information within the target time period.

[0120] According to embodiments of this disclosure, the node type information includes microservice type and function type. Correspondingly, the K trained prediction models include prediction models corresponding to service nodes of the microservice type and prediction models corresponding to service nodes of the function type.

[0121] For example, a prediction model corresponding to a microservice-type service node can be trained using the metric information of the microservice-type service node. Similarly, a prediction model corresponding to a function-type service node can be trained using the metric information of the function-type service node.

[0122] Based on this, when the node type of the node to be adjusted is a microservice type, the indicator information within the first predetermined time period can be input into the target prediction model corresponding to the service node of the microservice type, and the target indicator information within the target time period can be output.

[0123] When the node type of the node to be adjusted is a function type, the indicator information within the first predetermined time period can be input into the target prediction model corresponding to the service node of the function type, and the target indicator information within the target time period can be output.

[0124] According to embodiments of this disclosure, by using a target prediction model corresponding to node type information to predict target indicator information, the accuracy of the predicted target indicator information is improved, thereby enabling accurate adjustment of the resources of the node to be adjusted and ensuring the SLA of the application.

[0125] Figure 8 A flowchart illustrating a node capacity adjustment method according to an embodiment of the present disclosure is shown schematically.

[0126] like Figure 8 As shown, the node capacity adjustment method in this embodiment includes operations S810~S870.

[0127] In operation S810, the target prediction model corresponding to the node type information is determined from K trained prediction models.

[0128] When operating S820, the indicator information of the current time window is preprocessed to obtain the preprocessed indicator information.

[0129] According to embodiments of this disclosure, the current time window may correspond to the current time period. The preprocessing described above may include deduplication and other processing, which is not limited herein.

[0130] When operating S830, the preprocessed indicator information is input into the target prediction model, and the target indicator information for the target time period is output.

[0131] When operating S840, the target index information is processed using a predefined function to obtain the scaling factor.

[0132] According to embodiments of this disclosure, the predetermined function may be constructed as needed, and this disclosure does not limit it.

[0133] In operation S850, it is determined whether the scaling factor is greater than a predetermined coefficient threshold. If yes, operation S860 is executed; if no, operation S870 is executed.

[0134] According to embodiments of this disclosure, the predetermined coefficient threshold can be 1, etc., and this disclosure does not limit it.

[0135] When operating the S860, the node to be adjusted is expanded to obtain the adjusted target service node.

[0136] When operating the S870, the node to be adjusted is scaled down to obtain the adjusted target service node.

[0137] According to embodiments of this disclosure, training samples are obtained as follows: Sample indicator information of the service node within I historical time periods is obtained, where I is an integer greater than 1. For the i-th historical time period within the I historical time periods, label information corresponding to the sample indicator information of the i-th historical time period is generated based on the time period identifiers of the Ii historical time periods following the i-th historical time period and the sample indicator information of the Ii historical time periods, where i is a positive integer less than 1. Training samples are generated based on the sample indicator information of the (I-1)-th historical time periods excluding the i-th historical time period and the label information corresponding to the sample indicator information of the (I-1)-th historical time periods.

[0138] According to embodiments of this disclosure, the j-th historical period can be determined from the i-th historical period following the i-th historical period, where j is a positive integer less than i. Then, the j-th sample indicator information collected within the j-th historical period is determined from the sample indicator information of the aforementioned i-th historical periods.

[0139] Therefore, based on the time period identifier of the j-th historical time period and the j-th sample indicator information, the j-th label information corresponding to the j-th sample indicator information can be generated.

[0140] In some embodiments, the time difference between the j-th and i-th historical time periods can be determined based on the time period identifiers of the j-th and i-th historical time periods. Then, based on this time difference and the j-th sample indicator information, the j-th label information corresponding to the i-th historical time period is generated. Thus, Ii-1 label information corresponding to the i-th historical time period can be obtained.

[0141] Based on this, the i-th historical period is removed from the aforementioned i-th historical period, resulting in i-1 historical periods. These i-1 historical periods correspond to the aforementioned i-1 label information. Then, based on the i-1 label information and the sample index information of the i-1 historical periods, i-1 training samples corresponding to the i-th historical period are generated. Thus, all training samples corresponding to the i-th historical periods can be obtained.

[0142] According to embodiments of this disclosure, training samples are generated by using sample indicator information from I-1 historical time periods and label information from I time periods. This enables the prediction model to accurately predict indicator information for different future time periods, improving the applicability of the prediction model. Consequently, resources of nodes to be adjusted can be adjusted accurately, ensuring the application's SLA.

[0143] Figure 9 A flowchart illustrating a model training method according to an embodiment of the present disclosure is shown schematically.

[0144] like Figure 9 As shown, the model training method in this embodiment includes operations S901 to S909.

[0145] When operating S901, obtain initial sample indicator information according to the service evaluation indicator dimensions of the application.

[0146] In operation S902, the initial sample index information is differentially processed according to the time window to obtain differential index information.

[0147] In operation S903, the differential index information is normalized to obtain the normalized index information.

[0148] When operating S904, the normalized indicator information is processed according to the time window to generate sample indicator information.

[0149] When operating the S905, label information is generated based on sample indicator information.

[0150] When operating S906, the sample indicator information is input into the initial model, and the predicted indicator information is output.

[0151] When operating S907, loss information is calculated based on prediction indicator information and label information.

[0152] In operation S908, determine whether the loss information is less than or equal to the predetermined loss. If not, proceed to operation S909; if yes, proceed to operation S910.

[0153] When operating the S909, the inverse gradient optimization method is used to adjust the model parameters of the trained model and re-output the prediction index information.

[0154] When operating S910, the trained model is identified as the trained prediction model.

[0155] It should be noted that, Figure 9 The model training method of this disclosure is only illustrative. In actual business applications, the model training method of this disclosure can be adjusted according to requirements. For example, sample indicator information and the time period identifier corresponding to the sample indicator information can be input into the initial model to output predicted indicator information. This disclosure does not limit this.

[0156] Based on the above, the node processing method provided in this disclosure enables the management of application microservices or function computations from the overall perspective of the entire application. Furthermore, by analyzing key SLI metrics of microservices or functions and using metrics awareness, policy decision-making, instance online / offline management, fault isolation, and intelligent adjustment and fault tolerance, the overall service level of the application can be guaranteed to meet the SLO target. The operational focus of this disclosure is no longer on intelligent handling centered around faults, but rather on ensuring smooth business operations, isolating the impact of faults, and conducting offline analysis and troubleshooting.

[0157] Based on the above node processing method, this disclosure also provides a node processing apparatus. The following will be combined with... Figure 10 The device is described in detail.

[0158] Figure 10 A schematic block diagram of a node processing apparatus according to an embodiment of the present disclosure is shown.

[0159] like Figure 10 As shown, the node processing device 1000 of this embodiment includes a monitoring module 1010, a first determination module 1020, a second determination module 1030, and a processing module 1040.

[0160] The monitoring module 1010 is used during application operation to monitor, according to service evaluation metrics, the first-dimensional metric information and the second-dimensional metric information of N service nodes in the service chain topology used to implement the application. The first-dimensional metric information characterizes the request reception status of the service node, and the second-dimensional metric information characterizes the request processing error status of the service node. N is a positive integer greater than 1. In one embodiment, the monitoring module 1010 can be used to execute the operation S210 described above, which will not be repeated here.

[0161] The first determining module 1020 is used to determine the node to be adjusted from the N service nodes, wherein the node to be adjusted is a service node whose first dimension indicator information satisfies a first predetermined condition. In one embodiment, the first determining module 1020 can be used to perform the operation S220 described above, which will not be repeated here.

[0162] The second determining module 1030 is used to determine a target strategy from M application service assurance strategies based on the second-dimensional indicator information of the node to be adjusted. The application service assurance strategy is a strategy used to ensure the stable operation of the application by adjusting nodes, and the adjustment methods for the M nodes corresponding to the M application service assurance strategies are different for each strategy. M is a positive integer greater than 1. In one embodiment, the second determining module 1030 can be used to execute the operation S230 described above, which will not be repeated here.

[0163] The processing module 1040 is used to process the node to be adjusted according to the target strategy to obtain the adjusted target service node. In one embodiment, the processing module 1040 can be used to perform the operation S240 described above, which will not be repeated here.

[0164] According to embodiments of this disclosure, the processing module 1040 includes a first determining submodule, a calling submodule, and a second determining submodule. The first determining submodule is used to determine, when the target strategy is a node isolation strategy, an interaction node that interacts with the node to be adjusted from the service chain topology. The calling submodule is used to call a substitute service node corresponding to the node to be adjusted to interact with the interaction node, so that the substitute service node can process the request corresponding to the node to be adjusted and isolate the node to be adjusted. The second determining submodule is used to determine the substitute service node as the target service node.

[0165] According to embodiments of this disclosure, the processing module 1040 further includes a display submodule and an update submodule. The display submodule is used to display a graph of the service chain topology on a visual interface, wherein the graph includes image identifiers of the N service nodes and node call paths corresponding to the N service nodes; the update submodule is used to update the graph based on the image identifiers of the substitute service nodes and the image identifiers of the nodes to be adjusted, thereby obtaining an updated graph.

[0166] According to embodiments of this disclosure, the processing module 1040 further includes a prediction submodule and an adjustment submodule. The prediction submodule is used to predict target indicator information of the node to be adjusted within a future target time period based on indicator information of the node to be adjusted within a first predetermined time period, when the target strategy is a node capacity adjustment strategy. The adjustment submodule is used to adjust the capacity of the node to be adjusted based on the target indicator information to obtain the adjusted target service node.

[0167] According to embodiments of this disclosure, the prediction submodule includes an identification unit, a determination unit, and a prediction unit. The identification unit identifies the node type information of the node to be adjusted; the determination unit determines a target prediction model corresponding to the node type information from K trained prediction models, wherein the types of service nodes corresponding to the training samples of the K trained prediction models are all different, and K is a positive integer greater than 1; the prediction unit inputs the indicator information within the first predetermined time period into the target prediction model and outputs the target indicator information within the target time period.

[0168] According to embodiments of this disclosure, the prediction submodule further includes an acquisition unit, a first generation unit, and a second generation unit. The acquisition unit acquires sample indicator information of the service node within I historical time periods, where I is an integer greater than 1. The first generation unit generates label information corresponding to the sample indicator information of the i-th historical time period, based on the time period identifiers of the Ii historical time periods following the i-th historical time period and the sample indicator information of the Ii historical time periods, where i is a positive integer less than 1. The second generation unit generates the training samples based on the sample indicator information of the (I-1)-th historical time periods excluding the i-th historical time period and the label information corresponding to the sample indicator information of the (I-1)-th historical time periods.

[0169] According to embodiments of this disclosure, the monitoring module 1010 includes a monitoring submodule; the first determining module 1020 includes a third determining submodule, a fourth determining submodule, and a fifth determining submodule. The monitoring submodule is used to monitor the request throughput of each of the N service nodes during a second predetermined time period; the third determining submodule is used to determine the throughput change of each of the N service nodes based on their request throughput during the second predetermined time period; the fourth determining submodule is used to determine the total throughput change based on their request throughput during the second predetermined time period, wherein the total throughput change is the change in the total throughput of the N service nodes during the second predetermined time period; the fifth determining submodule is used to determine the node to be adjusted from the N service nodes based on the throughput changes of each of the N service nodes and the total throughput change, wherein the throughput change of the node to be adjusted has the greatest impact on the degree of change in the total throughput change.

[0170] According to embodiments of this disclosure, the third determining unit 1030 includes a sixth determining submodule and a seventh determining submodule. The sixth determining submodule is used to determine the node isolation strategy as the target strategy when the request processing error rate is greater than or equal to an error rate threshold; the seventh determining submodule is used to determine the node capacity adjustment strategy as the target strategy when the request processing error rate is less than the error rate threshold.

[0171] According to embodiments of this disclosure, any plurality of modules among the monitoring module 1010, the first determining module 1020, the second determining module 1030, and the processing module 1040 may be combined into one module, or any one of these modules may be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules may be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the monitoring module 1010, the first determining module 1020, the second determining module 1030, and the processing module 1040 may be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these methods. Alternatively, at least one of the monitoring module 1010, the first determining module 1020, the second determining module 1030, and the processing module 1040 may be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0172] Figure 11 A block diagram schematically illustrates an electronic device suitable for implementing a node processing method according to an embodiment of the present disclosure.

[0173] like Figure 11 As shown, an electronic device 1100 according to an embodiment of the present disclosure includes a processor 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage portion 1108 into a random access memory (RAM) 1103. The processor 1101 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1101 may also include onboard memory for caching purposes. The processor 1101 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0174] RAM 1103 stores various programs and data required for the operation of electronic device 1100. Processor 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. Processor 1101 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1102 and / or RAM 1103. It should be noted that the programs may also be stored in one or more memories other than ROM 1102 and RAM 1103. Processor 1101 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0175] According to embodiments of this disclosure, the electronic device 1100 may further include an input / output (I / O) interface 1105, which is also connected to a bus 1104. The electronic device 1100 may also include one or more of the following components connected to the input / output (I / O) interface 1105: an input section 1106 including a keyboard, mouse, etc.; an output section 1107 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1108 including a hard disk, etc.; and a communication section 1109 including a network interface card such as a LAN card, modem, etc. The communication section 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the input / output (I / O) interface 1105 as needed. A removable medium 1111, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1110 as needed so that computer programs read from it can be installed into the storage section 1108 as needed.

[0176] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0177] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, 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, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 1102 and / or RAM 1103 and / or one or more memories other than ROM 1102 and RAM 1103 described above.

[0178] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the node processing methods provided in the embodiments of this disclosure.

[0179] When the computer program is executed by the processor 1101, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0180] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1109, and / or installed from the removable medium 1111. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0181] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 1109, and / or installed from removable medium 1111. When the computer program is executed by processor 1101, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0182] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on a user's computing device, partially on a user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0183] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0184] Those skilled in the art will understand that the features described in the various embodiments of this disclosure can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments of this disclosure can be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0185] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A node processing method, comprising: During application operation, for N service nodes in the service chain topology used to implement the application, according to the service evaluation index dimension, the first dimension index information and the second dimension index information of the N service nodes are monitored. The first dimension index information is used to characterize the request receiving status of the service node, and the second dimension index information is used to characterize the request processing error status of the service node. N is a positive integer greater than 1. The node to be adjusted is determined from the N service nodes, wherein the node to be adjusted is a service node whose first dimension indicator information meets the first predetermined condition; Based on the second-dimensional indicator information of the node to be adjusted, a target strategy is determined from M application service assurance strategies. Each of these M application service assurance strategies is used to ensure stable application operation by adjusting nodes, and the adjustment methods for each of the M nodes corresponding to the M application service assurance strategies are different. M is a positive integer greater than 1. The node to be adjusted is processed according to the target strategy to obtain the adjusted target service node; The first dimension of the metric information includes request throughput; The monitoring of the first-dimensional indicator information and the second-dimensional indicator information of the N service nodes includes: During the second predetermined time period, monitor the request throughput of each of the N service nodes; The step of determining the node to be adjusted from the N service nodes includes: Based on the request throughput of the N service nodes during the second predetermined time period, determine the throughput change of each of the N service nodes; Based on the request throughput of the N service nodes during the second predetermined time period, the change in total throughput is determined, wherein the change in total throughput is the change in the total throughput of the N service nodes during the second predetermined time period; Based on the throughput changes of each of the N service nodes and the total throughput change, the node to be adjusted is determined from the N service nodes, wherein the throughput change of the node to be adjusted has the greatest impact on the degree of change of the total throughput change.

2. The method according to claim 1, wherein, The step of processing the node to be adjusted according to the target strategy to obtain the adjusted target service node includes: When the target strategy is a node isolation strategy. From the service chain topology, determine the interaction nodes that will interact with the node to be adjusted; Invoke the backup service node corresponding to the node to be adjusted, and interact with the interaction node to process requests corresponding to the node to be adjusted through the backup service node, thereby isolating the node to be adjusted; and The substitute service node is identified as the target service node.

3. The method according to claim 2, wherein, The method further includes: The service chain topology is displayed in a visualization interface, wherein the graph includes the image identifiers of the N service nodes and the node call paths corresponding to the N service nodes; Based on the image identifiers of the substitute service nodes and the image identifiers of the nodes to be adjusted, the map is updated to obtain the updated map.

4. The method according to claim 1, wherein, The step of processing the node to be adjusted according to the target strategy to obtain the adjusted target service node includes: When the target strategy is a node capacity adjustment strategy, the target indicator information of the node to be adjusted in a future target period is predicted based on the indicator information of the node to be adjusted in the first predetermined period. Based on the target indicator information, the capacity of the node to be adjusted is adjusted to obtain the adjusted target service node.

5. The method according to claim 4, wherein, When the target strategy is a node capacity adjustment strategy, based on the indicator information of the node to be adjusted in the first predetermined time period, the target indicator information of the node to be adjusted in the future target time period is predicted, including: Identify the node type information of the node to be adjusted; From K trained prediction models, determine the target prediction model corresponding to the node type information, wherein the types of service nodes corresponding to the training samples of the K trained prediction models are all different, and K is a positive integer greater than 1. The indicator information within the first predetermined time period is input into the target prediction model, and the target indicator information within the target time period is output.

6. The method according to claim 5, wherein, The training samples were obtained through the following method: Obtain sample indicator information of the service node within I historical time periods, where I is an integer greater than 1; For the i-th historical period within I historical periods, based on the period identifiers of the Ii historical periods following the i-th historical period and the sample indicator information of the Ii historical periods, label information corresponding to the sample indicator information of the i-th historical period is generated, where i is a positive integer less than I; The training samples are generated based on the sample indicator information of I-1 historical periods other than the I-th historical period and the label information corresponding to the sample indicator information of the I-1 historical periods.

7. The method according to claim 1, wherein, The second dimension of the metric information includes the request processing error rate; The multiple application service protection strategies include node isolation strategies and node capacity adjustment strategies; The determination of the target strategy from multiple application service assurance strategies based on the second-dimensional indicator information of the node to be adjusted includes: If the request processing error rate is greater than or equal to the error rate threshold, the node isolation strategy will be determined as the target strategy. as well as If the request processing error rate is less than the error rate threshold, the node capacity adjustment strategy will be determined as the target strategy.

8. The method according to claim 1, wherein, The service node is a function or microservice node used to implement service functions; The first dimension of the metric information also includes request response time; The second dimension of the metric information includes the request processing error rate.

9. A node processing apparatus, comprising: The monitoring module is used to monitor the first dimension indicator information and the second dimension indicator information of N service nodes in the service chain topology used to implement the application during the application operation, according to the service evaluation indicator dimension. The first dimension indicator information is used to characterize the request reception status of the service node, and the second dimension indicator information is used to characterize the request processing error status of the service node. N is a positive integer greater than 1. The first determining module is used to determine the node to be adjusted from the N service nodes, wherein the node to be adjusted is a service node whose first dimension indicator information meets the first predetermined condition. The second determining module is used to determine a target strategy from M application service assurance strategies based on the second-dimensional indicator information of the node to be adjusted. The application service assurance strategy is a strategy used to ensure stable application operation by adjusting nodes, and the adjustment methods for the M nodes corresponding to the M application service assurance strategies are different for each of them. M is a positive integer greater than 1. The processing module is used to process the node to be adjusted according to the target strategy to obtain the adjusted target service node; The first dimension of the metric information includes request throughput; The monitoring of the first-dimensional indicator information and the second-dimensional indicator information of the N service nodes includes: During the second predetermined time period, monitor the request throughput of each of the N service nodes; The step of determining the node to be adjusted from the N service nodes includes: Based on the request throughput of the N service nodes during the second predetermined time period, determine the throughput change of each of the N service nodes; Based on the request throughput of the N service nodes during the second predetermined time period, the change in total throughput is determined, wherein the change in total throughput is the change in the total throughput of the N service nodes during the second predetermined time period; Based on the throughput changes of each of the N service nodes and the total throughput change, the node to be adjusted is determined from the N service nodes, wherein the throughput change of the node to be adjusted has the greatest impact on the degree of change of the total throughput change.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 8.

11. A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 8.

12. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images