Instance processing method for serverless architecture, and apparatus

Abstract

The present application relates to the field of data processing. Disclosed are an instance processing method for a serverless architecture, and an apparatus. The method comprises: acquiring a performance indicator data set of a plurality of first instances in a serverless architecture, the first instances being used for executing tasks of a target object, and performance indicator data in the performance indicator data set being used for representing current operating statuses of the first instances; on the basis of the performance indicator data set, determining whether there is an abnormal instance among the plurality of first instances, the abnormal instance being an instance the current operating status of which has had a problem; and in case that there is an abnormal instance among the plurality of first instances, determining the abnormity level of the abnormal instance, and on the basis of the abnormity level of the abnormal instance, determining a processing policy for the abnormal instance. The present application solves the technical problem of the low resource utilization rate caused by failures to process abnormal instances in the prior art.

Description

[0001] This application claims priority to Chinese patent application number 202410320540.1, filed with the Chinese Patent Office on March 19, 2024, entitled "Instance Processing Method and Apparatus for Serverless Architecture," the entire contents of which are incorporated herein by reference. Technical Field: This application relates to the field of data processing, and more specifically, to an instance processing method and apparatus for serverless architecture. Background: A serverless architecture (Serverless architecture) is a computing model that allows developers to build and run applications without managing servers. In this architecture, cloud service providers are responsible for running servers and dynamically managing resource allocation. Serverless architectures typically rely on ephemeral, stateless computing instances that are launched on demand and released after processing requests. Although Serverless architectures offer high scalability and flexibility, in actual operation, instances may become unstable or unable to respond to new requests due to programming errors, resource limitations, or other reasons. These abnormal instances not only consume resources but also may affect the performance and reliability of the entire service. Existing technologies often lack effective automated processing mechanisms. In such cases, abnormal instances may continue to occupy resources without being promptly discovered and addressed, resulting in wasted resources and reduced service quality. Currently, no effective solution has been proposed to address the problem of low resource utilization caused by the inability to address abnormal instances in the aforementioned related technologies. Summary of the Invention: The present invention provides an instance processing method and apparatus for a serverless architecture, at least to address the technical problem of low resource utilization caused by the inability to address abnormal instances in the related technologies. According to one aspect of an embodiment of the present application, a method for instance processing in a serverless architecture is provided, comprising: obtaining a performance indicator data set of multiple first instances in the serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set are used to characterize the current operating status of the multiple first instances; judging whether there are abnormal instances in the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; if the abnormal instance exists in the multiple first instances, determining the abnormality level of the abnormal instance, and determining a processing strategy for the abnormal instance based on the abnormality level of the abnormal instance.Furthermore, determining whether an abnormal instance exists in the plurality of first instances based on the performance indicator data set includes: determining whether the abnormal instance exists in the plurality of first instances based on a preset abnormal instance determination rule and the performance indicator data set; or determining whether the abnormal instance exists in the plurality of first instances based on a machine learning model and the performance indicator data set. Furthermore, determining whether the abnormal instance exists in the plurality of first instances based on the preset abnormal instance determination rule and the performance indicator data set includes: determining whether performance indicator data in the performance indicator data set satisfies a performance indicator condition in the abnormal instance determination rule, and obtaining a determination result; if the determination result indicates that performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, then determining that the abnormal instance exists in the plurality of first instances, and determining the first instance corresponding to the performance indicator data that satisfies the performance indicator condition as the abnormal instance; and if the determination result indicates that no performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, then determining that the abnormal instance does not exist in the plurality of first instances. Furthermore, determining whether an abnormal instance exists among the multiple first instances based on the machine learning model and the performance indicator data set includes: extracting features from the performance indicator data in the performance indicator data set using the machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector representing the operating status of the multiple first instances; predicting an abnormality probability value of a first instance among the multiple first instances based on the target feature vector using the machine learning model to obtain an abnormality probability value; and determining whether the abnormal instance exists among the multiple first instances based on the abnormality probability value. Furthermore, if the abnormal instance exists among the multiple first instances, determining the abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining, from the multiple first instances, the number of second instances associated with the abnormal instance based on the currently processed task information and the pending task information; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances.Furthermore, determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance includes: if the abnormality level of the abnormal instance is level 1, determining the handling strategy for the abnormal instance to re-determine the current operating status of the abnormal instance after a preset time, and restarting the abnormal instance if there is a problem with the current operating status; if the abnormality level of the abnormal instance is level 2, determining the handling strategy for the abnormal instance to restart the abnormal instance, wherein the abnormality impact of the second level is higher than the abnormality impact of the first level; and if the abnormality level of the abnormal instance is level 3, determining the handling strategy for the abnormal instance to remove the abnormal instance, wherein the abnormality impact of the third level is higher than the abnormality impact of the second level. Furthermore, after determining that the handling strategy for the abnormal instance is to remove the abnormal instance, the method further includes: removing the abnormal instance from an instance service pool; creating a third instance based on the currently processed task information and the pending task information of the abnormal instance, and executing the currently processed task information and the pending task information according to the third instance. Furthermore, after determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance, the method further includes: writing the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the handling strategy into a log file; and optimizing the abnormal instance determination rule and / or the machine learning model based on the log file. According to another aspect of an embodiment of the present application, a method for handling instances in a serverless architecture is provided, comprising: obtaining a performance indicator data set of multiple first instances in the serverless architecture uploaded by a client, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances; determining, on a cloud server, whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in its current operating status; if the abnormal instance exists among the multiple first instances, determining the abnormality level of the abnormal instance and, based on the abnormality level of the abnormal instance, determining a handling strategy for the abnormal instance; and returning the handling strategy to the client.According to another aspect of an embodiment of the present application, an instance processing device for a serverless architecture is provided, comprising: a first acquisition unit configured to acquire a performance indicator data set for multiple first instances in the serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances; a first judgment unit configured to determine whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; and a first determination unit configured to determine the abnormality level of the abnormal instance if the abnormal instance exists among the multiple first instances, and determine a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. Furthermore, the first judgment unit comprises: a first judgment module configured to determine whether there is an abnormal instance among the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or a second judgment module configured to determine whether there is an abnormal instance among the multiple first instances based on a machine learning model and the performance indicator data set. Further, the first judgment module includes: a first judgment submodule, configured to judge whether the performance indicator data in the performance indicator data set meets the performance indicator condition in the abnormal instance determination rule, and obtain a judgment result; a first determination submodule, configured to, if the judgment result indicates that there is performance indicator data in the performance indicator data set that meets the performance indicator condition in the abnormal instance determination rule, indicate that the abnormal instance exists in the multiple first instances, and determine the first instance corresponding to the performance indicator data that meets the performance indicator condition as the abnormal instance; and a second determination submodule, configured to, if the judgment result indicates that there is no performance indicator data in the performance indicator data set that meets the performance indicator condition in the abnormal instance determination rule, indicate that the abnormal instance does not exist in the multiple first instances. Furthermore, the second judgment module includes: an extraction submodule, configured to perform feature extraction on the performance indicator data in the performance indicator data set through the machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector characterizing the operating status of the multiple first instances; a prediction submodule, configured to predict the abnormal probability value of the first instance in the multiple first instances based on the target feature vector through the machine learning model to obtain an abnormal probability value; and a second judgment submodule, configured to judge whether the abnormal instance exists in the multiple first instances based on the abnormal probability value.Furthermore, the first determination unit includes: an acquisition module, configured to acquire currently processed task information and to-be-processed task information of the abnormal instance; a first determination module, configured to determine, from the multiple first instances, a number of second instances associated with the abnormal instance based on the currently processed task information and the to-be-processed task information; and a second determination module, configured to determine an abnormality level of the abnormal instance based on the performance indicator data set and the number of the second instances. Further, the first determination unit includes: a third determination module, configured to, if the abnormality level of the abnormal instance is the first level, determine that a processing strategy for the abnormal instance is to re-determine the current running status of the abnormal instance after a preset time, and restart the abnormal instance if there is a problem with the current running status; a fourth determination module, configured to, if the abnormality level of the abnormal instance is the second level, determine that a processing strategy for the abnormal instance is to restart the abnormal instance, wherein the abnormality impact of the second level is higher than the abnormality impact of the first level; and a fifth determination module, configured to, if the abnormality level of the abnormal instance is the third level, determine that a processing strategy for the abnormal instance is to eliminate the abnormal instance, wherein the abnormality impact of the third level is higher than the abnormality impact of the second level. Furthermore, the apparatus further includes: a removal unit configured to, after determining that the processing strategy for the abnormal instance is to remove the abnormal instance, remove the abnormal instance from the instance service pool; a creation unit configured to, based on the currently processed task information and pending task information of the abnormal instance, create a third instance, and execute the currently processed task information and pending task information according to the third instance. Furthermore, the apparatus further includes: a writing unit configured to, after determining the processing strategy for the abnormal instance based on the abnormality level of the abnormal instance, write the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy into a log file; and an optimization unit configured to optimize the abnormal instance determination rules and / or the machine learning model based on the log file.According to another aspect of an embodiment of the present application, an instance processing apparatus for a serverless architecture is provided, comprising: a second acquisition unit configured to acquire a performance indicator data set of multiple first instances in a serverless architecture uploaded by a client, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances; a second determination unit configured to determine, on a cloud server, whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in its current operating status; a second determination unit configured to determine the abnormality level of the abnormal instance if the abnormal instance exists among the multiple first instances, and determine a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance; and a return unit configured to return the handling strategy to the client. According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the storage medium stores a program, wherein, when the program is executed, the device containing the storage medium is controlled to execute any one of the aforementioned instance processing methods for a serverless architecture. According to another embodiment of the present application, an electronic device is provided, comprising a memory storing an executable program; and a processor configured to execute the program, wherein the program, when executed, executes any one of the aforementioned instance processing methods for a serverless architecture. According to another embodiment of the present application, a computer program product is provided, comprising a stored computer program, wherein when executed by the processor, the computer program implements any one of the aforementioned instance processing methods for a serverless architecture. In this embodiment of the present application, the following steps are employed: obtaining a performance indicator data set for multiple first instances in a serverless architecture, wherein the first instances are configured to execute tasks of a target object, and the performance indicator data in the performance indicator data set represents the current operating status of the first instances; determining, based on the performance indicator data set, whether there are abnormal instances among the multiple first instances, wherein the abnormal instance is an instance with a problem in its current operating status; and if there are abnormal instances among the multiple first instances, determining an abnormality level of the abnormal instance, and determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. This solves the technical problem in related arts of being unable to handle abnormal instances, resulting in low resource utilization.In this solution, a performance indicator data set of the current first instance in the serverless architecture is collected. The performance indicator data in the performance indicator data set can be used to accurately assess the current operating status of the first instance. Therefore, the performance indicator data set can be used to determine whether there are abnormal instances among the multiple first instances, that is, whether there are instances with abnormal operating status. If an abnormal instance exists, the abnormality level of the abnormal instance can be determined, and a handling strategy for the abnormal instance can be determined based on the abnormality level. Through the above steps, the operating status of instances in the serverless architecture can be automatically detected, preventing abnormal instances from affecting the overall service. If an abnormal instance is determined to exist, a handling strategy for the abnormal instance is determined based on the abnormality level of the abnormal instance, which can reduce resource waste caused by abnormal instances and thereby improve resource utilization of the serverless architecture. BRIEF DESCRIPTION OF THE DRAWINGS The drawings described herein are provided to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their description are provided for the purpose of explaining the present application and are not intended to unduly limit the present application. In the accompanying drawings: Figure 1 is a hardware block diagram of a computer terminal according to the first embodiment of the present application; Figure 2 is a flowchart of a method for processing instances in a serverless architecture according to the first embodiment of the present application; Figure 3 is a flowchart of a method for processing instances in a serverless architecture according to the first embodiment of the present application; Figure 4 is a flowchart of a method for processing instances in a serverless architecture according to the second embodiment of the present application; Figure 5 is a schematic diagram of an apparatus for processing instances in a serverless architecture according to the third embodiment of the present application; Figure 6 is a schematic diagram of an apparatus for processing instances in a serverless architecture according to the fourth embodiment of the present application; Figure 7 is a schematic diagram of an apparatus for processing instances in a serverless architecture according to the fifth embodiment of the present application; and Figure 8 is a hardware block diagram of a computer terminal according to the sixth embodiment of the present application. To help those skilled in the art better understand the present invention, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. It should be understood that the described embodiments represent only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present application without inventive effort should fall within the scope of protection of the present application. It should be noted that the terms "first," "second," and so on, in the specification and claims of this application and the aforementioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.Furthermore, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus. First, some nouns or terms appearing in the description of the embodiments of this application are subject to the following interpretation:

[0002] Serverless computing is a cloud computing model based on Platform as a Service (PaaS). Serverless computing provides a micro-architecture where end customers do not need to deploy, configure, or manage server services. All server services required for code execution are provided by the cloud platform. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, storage, and display) involved in this application are all authorized by the user or fully authorized by all parties. The collection, use, and processing of relevant data must comply with relevant laws, regulations, and standards in the relevant region, and corresponding operation portals are provided for users to choose to authorize or reject. Example 1 According to an embodiment of the present application, an example processing method for a serverless architecture is also provided. It should be noted that the steps shown in the flowchart of the accompanying figure can be executed in a computer system such as a set of computer-executable instructions. Although the flowchart shows a logical order, in some cases, the steps shown or described can be executed in a different order. The method embodiment provided in Example 1 of the present application can be executed in a mobile terminal, computer terminal, or similar computing device. Figure 1 shows a hardware block diagram of a computer terminal (or mobile device) for implementing an example processing method for a serverless architecture. As shown in Figure 1 , the computer terminal (or mobile device) 10 may include a processor assembly 102 (processor assembly 102 may include, but is not limited to, a processing device such as a microprocessor (MCU) or a programmable logic device (FPGA), and the processor assembly 102 may include a processor assembly, as shown in Figure 1 as 102a, 102b, ..., 102n), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, the computer terminal (or mobile device) 100 may include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS), a network interface, a power supply, and / or a camera. Those skilled in the art will appreciate that the structure shown in Figure 1 is merely illustrative and does not limit the structure of the electronic device. For example, the computer terminal 10 may include more or fewer components than shown in FIG1 , or have a configuration different from that shown in FIG1 . It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as “data processing circuits”.The data processing circuit can be embodied in whole or in part as software, hardware, firmware, or any other combination. Furthermore, the data processing circuit can be a single, independent processing module, or fully or partially integrated into any other component of the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selecting a path for a variable resistor terminal connected to an interface). Memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the example processing method for a serverless architecture in the embodiments of the present application. Processor 102 executes the software programs and modules stored in memory 104 to execute various functional applications and data processing, thereby implementing the aforementioned example processing method for a serverless architecture. Memory 104 can include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some embodiments, memory 104 may further include memory remotely located from processor 102, which can be connected to computer terminal 10 via a network. Examples of the aforementioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. Transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by the communications provider of computer terminal 10. In one example, transmission device 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to enable communication with the Internet. In one example, transmission device 106 can be a radio frequency (RF) module for wireless communication with the Internet. The display can be a touchscreen liquid crystal display that enables a user to interact with the user interface of computer terminal 10 (or mobile device). In the aforementioned operating environment, the present application provides an example processing method for a serverless architecture, as shown in FIG2 . FIG2 is a flowchart of an example processing method for a serverless architecture according to the first embodiment of the present application. As shown in FIG2 , the instance processing method includes: Step S201, obtaining a performance indicator data set of multiple first instances in a serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances.Optionally, all running first instances in the current Serverless instance pool of the Serverless architecture are determined, and then, using an instance detection component, performance indicator data of all first instances is collected in real time to obtain the aforementioned performance indicator data set. It should be noted that the instance detection component can be deployed around the Serverless instance pool to promptly obtain instance performance indicator data. It should be noted that performance indicator data includes, but is not limited to, key indicators such as instance response time, error rate, and memory usage. This performance indicator data can accurately represent the current operating status of the first instances. In step S202, based on the performance indicator data set, it is determined whether there are any abnormal instances among the multiple first instances. An abnormal instance is an instance with a problem in its current operating status. Optionally, after collecting and obtaining the aforementioned performance indicator data set, the instance detection component transmits the performance indicator data set to an anomaly detection model, and the anomaly detection module determines whether there are any abnormal instances among the multiple first instances. It should be noted that the anomaly detection module can determine whether there are any abnormal instances based on whether the performance indicator data exceeds a threshold. It should be noted that the aforementioned abnormal instances are instances experiencing problems with the current operating state, such as instances that cannot correctly process data or instances with excessively high memory usage. In step S203, if an abnormal instance exists among the multiple first instances, the abnormality level of the abnormal instance is determined, and a handling strategy for the abnormal instance is determined based on the abnormality level of the abnormal instance. Alternatively, if an abnormal instance with an abnormal operation exists among the multiple first instances, the abnormality level of the abnormal instance can be further determined based on the performance indicator data of the abnormal instance. It should be noted that the abnormality level is used to indicate the abnormal impact of the current abnormal instance. After determining the abnormality level of the abnormal instance, a handling strategy for the abnormal instance is determined based on the abnormality level, such as restarting or removing the abnormal instance.In summary, a performance indicator data set of the current first instance of the serverless architecture is collected, and the performance indicator data in the performance indicator data set can be used to accurately evaluate the current operating status of the first instance. Therefore, the performance indicator data set can be used to determine whether there are abnormal instances among the multiple first instances, that is, whether there are instances with abnormal operating status. If there are abnormal instances, the abnormality level of the abnormal instance can be determined, and the processing strategy for the abnormal instance can be determined according to the abnormality level. Through the above steps, the operating status of the instances in the serverless architecture can be automatically detected, avoiding the situation where the abnormal instance affects the overall service. In addition, when it is determined that there is an abnormal instance, the processing strategy for the abnormal instance is determined according to the abnormality level of the abnormal instance, which can reduce the waste of resources caused by the abnormal instance, thereby achieving the effect of improving the resource utilization of the serverless architecture. To improve the accuracy of determining abnormal instances, in the instance processing method for a serverless architecture provided in the first embodiment of the present application, determining whether there are abnormal instances in the multiple first instances based on a performance indicator data set includes: determining whether there are abnormal instances in the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or determining whether there are abnormal instances in the multiple first instances based on a machine learning model and the performance indicator data set. Optionally, the following two methods can be used to determine whether there are abnormal instances in the first instances: First, obtain an empirically determined abnormal instance determination rule. For example, the abnormal instance determination rule may define an instance as an abnormal instance if performance indicator data such as response time, error rate, or memory usage exceeds a defined threshold. Determine whether there are abnormal instances in the multiple first instances based on the abnormal instance determination rule and the performance indicator data set. Second, input the performance indicator data set into a machine learning model, and determine whether there are abnormal instances in the first instances using the machine learning model. In an optional embodiment, the first and second methods can be combined to determine whether there are abnormal instances in the first instances. For example, different weights can be assigned to the determination results of the first and second methods, and finally, whether an abnormal instance exists is determined based on the weights and the determination results of the first and second methods. These two methods allow for flexible determination of abnormal instance existence, adapting to complex and diverse application scenarios. In an optional embodiment, a machine learning model and historical instance performance data can be used to predict potential instance issues and intervene before they actually occur, further improving the operational stability of the serverless system.To improve the efficiency of determining whether an abnormal instance exists, in the instance processing method for a serverless architecture provided in the first embodiment of the present application, based on a preset abnormal instance determination rule and a performance indicator data set, determining whether an abnormal instance exists among multiple first instances includes: determining whether performance indicator data in the performance indicator data set satisfies a performance indicator condition in the abnormal instance determination rule, and obtaining a determination result; if the determination result indicates that performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, then an abnormal instance exists among the multiple first instances; if the determination result indicates that no performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, then an abnormal instance does not exist among the multiple first instances. Optionally, determining whether an abnormal instance exists using the preset abnormal instance determination rule includes the following steps: determining whether the performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule. For example, the performance indicator condition in the abnormal instance determination rule may include whether the error rate of the instance exceeds a threshold, or whether the response time exceeds a time threshold. If performance indicator data in the performance indicator data set satisfies the performance indicator conditions in the abnormal instance determination rule, then an abnormal instance is detected in the multiple first instances, and the first instance corresponding to the performance indicator data that satisfies the performance indicator conditions is determined to be an abnormal instance. If no performance indicator data in the performance indicator data set satisfies the performance indicator conditions in the abnormal instance determination rule, then an abnormal instance is detected in the multiple first instances. The performance indicator conditions in the abnormal instance determination rule can be used to quickly and directly determine whether an abnormal instance exists in the first instance, and the result of determining whether an abnormal instance exists in the first instance is interpretable. To improve the accuracy of determining whether an abnormal instance exists, in the instance processing method for a serverless architecture provided in the first embodiment of the present application, determining whether an abnormal instance exists in the multiple first instances based on a machine learning model and the performance indicator data set includes: extracting features from the performance indicator data in the performance indicator data set using the machine learning model to obtain a target feature vector, where the target feature vector is a feature vector representing the operating status of the multiple first instances; predicting an abnormality probability value for a first instance in the multiple first instances based on the target feature vector using the machine learning model to obtain an abnormality probability value; and determining whether an abnormal instance exists in the multiple first instances based on the abnormality probability value.Optionally, determining whether an abnormal instance exists using a machine learning model includes the following steps: inputting performance indicator data from a performance indicator data set into the machine learning model; performing feature extraction on the performance indicator data in the performance indicator data set using the machine learning model to obtain a corresponding target feature vector. For example, the machine learning model performs a convolution calculation on all performance indicator data to obtain an initial feature vector; then, feature vectors in the initial feature vector are added or multiplied to obtain a target feature vector. The target feature vector is then used to predict an abnormality probability value for a first instance among multiple first instances to obtain an abnormality probability value. Finally, determining whether an abnormal instance exists in the first instance based on the abnormality probability value. For example, if the abnormality probability value exceeds 60%, the instance is determined to be an abnormal instance. It should be noted that the aforementioned machine learning model is trained based on a training set constructed from historical instance sample data and the true labels corresponding to the historical instance sample data. The machine learning module can mine potential features in the performance indicator data, thereby effectively improving the accuracy of determining whether an abnormal instance exists. To improve the rationality of handling abnormal instances, in the instance handling method for a serverless architecture provided in the first embodiment of the present application, if an abnormal instance exists among multiple first instances, determining the abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining the number of second instances associated with the abnormal instance from the multiple first instances based on the currently processed task information and pending task information; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances. Optionally, after the abnormal instance is determined to exist, the currently processed task information and pending task information of the abnormal instance are obtained, and then the number of second instances associated with the abnormal instance in the first instance is determined based on the currently processed task information and pending task information. It should be noted that the second instances associated with the abnormal instance here can be instances that require further processing based on the output results of the abnormal instance, or instances that output results to the abnormal instance for further processing. After obtaining the number of second instances associated with the abnormal instance, the abnormality level of the abnormal instance can be determined based on the performance indicator data set and the number of second instances. For example, the greater the number of second instances, or the more indicator data in the performance indicator dataset exceeds a threshold, the higher the abnormality level of the corresponding abnormal instance. In an optional embodiment, a deep learning model can be used to predict the abnormality level of the abnormal instance based on the number of second instances and the indicator data in the performance indicator dataset to obtain the corresponding abnormality level.The number of second instances associated with an abnormal instance and the performance indicator data of the abnormal instance can be used to accurately assess the impact of the abnormal instance and the severity of the abnormality, thereby improving the accuracy of determining the abnormality level. To achieve flexible handling of abnormal instances, the instance handling method for a serverless architecture provided in the first embodiment of the present application determines a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. The method includes: if the abnormality level of the abnormal instance is level 1, determining the handling strategy for the abnormal instance to re-determine the current operating status of the abnormal instance after a preset time, and restarting the abnormal instance if there is a problem with the current operating status; if the abnormality level of the abnormal instance is level 2, determining the handling strategy for the abnormal instance to restart the abnormal instance, where the abnormality impact of the second level is higher than the abnormality impact of the first level; and if the abnormality level of the abnormal instance is level 3, determining the handling strategy for the abnormal instance to eliminate the abnormal instance, where the abnormality impact of the third level is higher than the abnormality impact of the second level. Optionally, if the abnormal instance's abnormality level is level 1, the operating status of the abnormal instance can be rechecked after a period of time (i.e., the aforementioned preset time). If the abnormal instance is still abnormal, the abnormal instance is restarted. If the recheck result indicates no abnormality, the abnormal instance is not processed. For example, if the abnormal instance's response time is only slightly above the response threshold, and the number of second instances associated with the abnormal instance is small, the abnormal instance's abnormality level is level 1. If the abnormal instance's abnormality level is level 2, the handling strategy for the abnormal instance is to restart the abnormal instance so that it can continue to process data. For example, if the abnormal instance's response time is above the response threshold, the error rate is above the threshold, and the number of second instances associated with the abnormal instance is greater than a first preset number, the abnormal instance's abnormality level is level 2. If the abnormal instance's abnormality level is level 3, the abnormal instance's impact is relatively high. Therefore, the handling strategy for the abnormal instance is to eliminate the abnormal instance to ensure the normal operation of the Serverless architecture. For example, if the error rate of an abnormal instance exceeds 90%, and the number of second instances associated with the abnormal instance exceeds a second preset number, the abnormal instance is assigned a third abnormality level. By setting different handling strategies for different abnormality levels, the flexibility of abnormal instance handling is improved, avoiding the problem of Serverless architecture malfunctioning due to excessive instance removal.To ensure the normal operation of the serverless architecture after removing abnormal instances, the instance processing method for a serverless architecture provided in the first embodiment of the present application, after determining that the processing strategy for abnormal instances is to remove abnormal instances, further includes: removing the abnormal instance from the instance service pool; creating a third instance based on the abnormal instance's currently processed task information and pending task information, and executing the currently processed task information and pending task information using the third instance. Optionally, the abnormal instance is removed from the instance service pool, and then a third instance is created based on the normal instance's currently processed task information and pending task information, so that the currently processed task information and pending task information can subsequently be executed by the third instance. To further improve the accuracy of abnormal instance identification, the instance processing method for a serverless architecture provided in the first embodiment of the present application, after determining the processing strategy for the abnormal instance based on its abnormality level, further includes: writing the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy to a log file; and optimizing the abnormal instance identification rules and / or machine learning model based on the log file. Optionally, all operations performed during instance identification and the corresponding handling strategies are recorded in a log file. Notifications can also be sent to administrators based on the log file, reporting key events and operations. Furthermore, the abnormal instance identification rules and machine learning models can be continuously optimized based on the log file to improve the accuracy of subsequent processing. In an optional embodiment, the flow chart shown in FIG3 can be used to implement abnormal instance identification and handling: Step 1: Real-time monitoring of the Serverless instance's status and performance indicators, including key indicators such as response time, error rate, and memory usage. Step 2: Analyze the Serverless instance's status and performance indicators, using preset rules or machine learning models to identify possible abnormal instances. Step 3: Once an abnormal instance is detected, a determination is made as to whether removal is necessary based on factors such as the severity, duration, and impact on other instances. Step 4: Once the instance is determined to be necessary, it is removed from the service pool, and a new instance is launched as needed to maintain service capacity. All operations and decisions are recorded for audit and subsequent analysis. The system may also send notifications to administrators, reporting key events and operations. A self-learning mechanism can be used to continuously optimize anomaly detection algorithms and handling strategies, continuously improving performance based on historical data and feedback. This solution implements an automated detection and anomaly handling mechanism, overcoming the inefficiency and scalability issues of traditional manual intervention methods.The status and performance of instances are continuously monitored, and abnormal instances can be identified and automatically handled in real time, significantly reducing reliance on manual operations. By detecting and intelligently handling abnormal instances in real time, this solution significantly improves the availability and stability of Serverless applications. By automatically eliminating abnormal instances and launching new instances as needed, efficient resource utilization is ensured and resource waste is reduced. In the instance processing method for a serverless architecture provided in Example 1 of the present application, a performance indicator data set for multiple first instances in the serverless architecture is obtained. The first instances are used to execute tasks of target objects, and the performance indicator data in the performance indicator data set represents the current operating status of the first instances. Based on the performance indicator data set, it is determined whether any of the multiple first instances contain abnormal instances, where an abnormal instance is an instance with a problem in its current operating status. If any of the multiple first instances contain abnormal instances, the abnormality level of the abnormal instance is determined, and based on the abnormality level, a handling strategy for the abnormal instance is determined. This method solves the technical problem in related technologies of being unable to handle abnormal instances, resulting in relatively low resource utilization. In this solution, a performance indicator data set for all current first instances in the serverless architecture is collected. The performance indicator data in the performance indicator data set can be used to accurately assess the current operating status of the first instances. Therefore, the performance indicator data set can be used to determine whether there are any abnormal instances among the multiple first instances, that is, whether there are any instances with abnormal operating status. If an abnormal instance exists, the abnormality level of the abnormal instance can be determined, and a handling strategy for the abnormal instance can be determined based on the abnormality level. Through the above steps, the operating status of instances in the serverless architecture can be automatically detected, preventing abnormal instances from affecting the overall service. If an abnormal instance is determined to exist, a handling strategy for the abnormal instance can be determined based on the abnormality level of the abnormal instance, thereby reducing resource waste caused by abnormal instances and thereby improving resource utilization of the serverless architecture. It should be noted that for simplicity of description, the aforementioned method embodiments are described as a series of actions. However, those skilled in the art should be aware that this application is not limited to the order of the actions described, as certain steps can be performed in a different order or simultaneously according to this application. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for this application.Through the above description of the embodiments, those skilled in the art will clearly understand that the methods according to the above embodiments can be implemented using software and a necessary general-purpose hardware platform. Hardware implementation is also possible, but in many cases the former is a more preferred implementation method. Based on this understanding, the technical solution of the present application, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, or a network device) to execute the methods of the various embodiments of the present application. Embodiment 2 According to an embodiment of the present application, a method for processing instances in a serverless architecture is also provided. As shown in FIG4 , the method includes: Step S401: Obtaining a performance indicator data set of multiple first instances in the serverless architecture uploaded by a client, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances; Step S402: Determining, on a cloud server, whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein an abnormal instance is an instance with a problem in its current operating status; Step S403: If an abnormal instance exists among the multiple first instances, determining the abnormality level of the abnormal instance and, based on the abnormality level of the abnormal instance, determining a handling strategy for the abnormal instance; Step S404: Returning the handling strategy to the client. On the cloud server, the specific method for processing instances in the serverless architecture is the same as that in Embodiment 1 and will not be further described here. It should be noted that, for simplicity of description, the aforementioned method embodiments are described as a series of combined actions. However, those skilled in the art should be aware that this application is not limited by the order of the actions described, as certain steps may be performed in a different order or simultaneously, depending on the application. Furthermore, those skilled in the art should also be aware that the embodiments described in this specification are preferred embodiments, and the actions and modules involved are not necessarily required for this application. Through the above description of the embodiments, those skilled in the art will clearly understand that the methods according to the aforementioned embodiments can be implemented using software plus a required general-purpose hardware platform, or alternatively, hardware. However, in many cases, the former is the preferred implementation method.Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, can essentially be embodied in the form of a software product. This computer software product is stored in a storage medium (e.g., ROM / RAM, a magnetic disk, or an optical disk) and includes instructions for enabling a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of various embodiments of this application. Example 3 According to an embodiment of this application, an instance processing device for a serverless architecture is also provided for implementing the aforementioned instance processing method for a serverless architecture. As shown in FIG5 , the device includes a first acquisition unit 501, a first judgment unit 502, and a first determination unit 503. oA first acquisition unit 501 is configured to acquire a performance indicator data set for multiple first instances in a serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances. A first judgment unit 502 is configured to determine whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein an abnormal instance is an instance with a problem in its current operating status. A first determination unit 503 is configured to determine the abnormality level of the abnormal instance if an abnormal instance exists among the multiple first instances, and determine a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. Optionally, in the instance processing device for a serverless architecture provided in Embodiment 3 of the present application, the first judgment unit includes: a first judgment module configured to determine whether there are abnormal instances among the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or a second judgment module configured to determine whether there are abnormal instances among the multiple first instances based on a machine learning model and the performance indicator data set. Optionally, in the instance processing device for a serverless architecture provided in Example 3 of the present application, the first judgment module includes: a first judgment sub-module, configured to judge whether the performance indicator data in the performance indicator data set meets the performance indicator conditions in the abnormal instance determination rule, and obtain a judgment result; a first determination sub-module, configured to, if the judgment result indicates that there is performance indicator data in the performance indicator data set that meets the performance indicator conditions in the abnormal instance determination rule, then it indicates that there are abnormal instances in multiple first instances, and the first instance corresponding to the performance indicator data that meets the performance indicator conditions is determined as an abnormal instance; a second determination sub-module, configured to, if the judgment result indicates that there is no performance indicator data in the performance indicator data set that meets the performance indicator conditions in the abnormal instance determination rule, then it indicates that there are no abnormal instances in multiple first instances. Optionally, in the instance processing device for a serverless architecture provided in Example 3 of the present application, the second judgment module includes: an extraction submodule, configured to perform feature extraction on the performance indicator data in the performance indicator data set through a machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector characterizing the operating status of the multiple first instances; a prediction submodule, configured to predict the abnormal probability value of the first instance in the multiple first instances based on the target feature vector through a machine learning model to obtain an abnormal probability value; and a second judgment submodule, configured to judge whether there is an abnormal instance in the multiple first instances based on the abnormal probability value.Optionally, in the instance processing device for a serverless architecture provided in Example 3 of the present application, the first determination unit includes: an acquisition module, configured to obtain currently processed task information and to-be-processed task information of the abnormal instance; a first determination module, configured to determine the number of second instances associated with the abnormal instance from multiple first instances based on the currently processed task information and to-be-processed task information; and a second determination module, configured to determine the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances. Optionally, in the instance processing device for a serverless architecture provided in Example 3 of the present application, the first determination unit includes: a third determination module, which is configured to, if the abnormality level of the abnormal instance is the first level, determine that the processing strategy for the abnormal instance is to re-determine the current running status of the abnormal instance after a preset time, and restart the abnormal instance if there is a problem with the current running status; a fourth determination module, which is configured to, if the abnormality level of the abnormal instance is the second level, determine that the processing strategy for the abnormal instance is to restart the abnormal instance, wherein the abnormal impact of the second level is higher than the abnormal impact of the first level; and a fifth determination module, which is configured to, if the abnormality level of the abnormal instance is the third level, determine that the processing strategy for the abnormal instance is to eliminate the abnormal instance, wherein the abnormal impact of the third level is higher than the abnormal impact of the second level. Optionally, in the instance processing device for a serverless architecture provided in the third embodiment of the present application, the device further includes: a removal unit configured to, after determining that the processing strategy for the abnormal instance is to remove the abnormal instance, remove the abnormal instance from the instance service pool; a creation unit configured to, based on the currently processed task information and pending task information of the abnormal instance, create a third instance, and execute the currently processed task information and pending task information according to the third instance. Optionally, in the instance processing device for a serverless architecture provided in the third embodiment of the present application, the device further includes: a writing unit configured to, after determining the processing strategy for the abnormal instance based on the abnormality level of the abnormal instance, write the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy into a log file; and an optimization unit configured to optimize the abnormal instance determination rules and / or machine learning model based on the log file. It should be noted that the first acquisition unit 501, first judgment unit 502, and first determination unit 503 described above correspond to steps S201 to S203 in Example 1. These three units and corresponding steps implement the same examples and application scenarios, but are not limited to those disclosed in Example 1. It should be noted that the aforementioned modules, as part of the apparatus, can be run in the computer terminal 10 provided in Example 1.It should be noted that the preferred implementation schemes involved in the above embodiments of this application are the same as the solution, application scenarios, and implementation processes provided in Example 1, but are not limited to the solution provided in Example 1. Example 4 According to an embodiment of this application, an instance processing device for a serverless architecture is also provided for implementing the above instance processing method for a serverless architecture. As shown in Figure 6, the device includes: a second obtaining unit 601, a second judging unit 602, a second determining unit 603, and a returning unit 604. A second acquisition unit 601 is configured to acquire a performance indicator data set of multiple first instances in a serverless architecture uploaded by a client, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances. A second judgment unit 602 is configured to determine, on a cloud server, whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status. A second determination unit 603 is configured to, if the abnormal instance exists among the multiple first instances, determine the abnormality level of the abnormal instance and, based on the abnormality level of the abnormal instance, determine a handling strategy for the abnormal instance. A return unit 604 is configured to return the handling strategy to the client. It should be noted that the second acquisition unit 601, second judgment unit 602, second determination unit 603, and return unit 604 described above correspond to steps S401 to S404 in Example 2. The examples and application scenarios implemented by these four units and the corresponding steps are the same, but are not limited to the contents disclosed in Example 1. It should be noted that the above-mentioned module, as part of the device, can be run in the computer terminal 10 provided in Example 1. It should be noted that the preferred implementation schemes involved in the above-mentioned embodiments of this application are the same as the scheme provided in Example 2, as well as the application scenarios and implementation processes, but are not limited to the scheme provided in Example 2. Example 5 According to an embodiment of this application, an instance processing system for a serverless architecture is also provided for implementing the above-mentioned instance processing method for a serverless architecture. As shown in FIG7 , the system includes: an instance detection component 70, an anomaly detection module 71, a decision engine 72, and a log management module 73. As shown in FIG7 , a large Serverless instance pool is located in the center, representing all running instances. Surrounding the instance pool are instance detection components 70, which collect data from each instance in real time. Instance detection components 70 transmit data to anomaly detection module 71, and the output of anomaly detection module 71 is connected to decision engine 72.The decision engine 72 controls the addition or removal of managed instances based on analysis results. The log management module 73 receives information from the decision engine and instance management module. At the core of the entire system is a self-learning mechanism that receives feedback from various components to optimize overall system performance. Instance Detection Component: This component monitors the status and performance metrics of all Serverless instances in real time, including key indicators such as response time, error rate, and memory usage. Anomaly Detection Module: The anomaly detection algorithm is the core of the solution. It analyzes detection data and uses pre-set rules or machine learning models to identify possible anomalies. Decision Engine: Once an anomaly instance is detected, the decision engine determines whether to remove it based on factors such as the severity, duration, and impact on other instances. If an instance is determined to be suitable for removal, it is automatically removed from the service pool and a new instance is launched as needed to maintain service capacity. Log Management Module: All operations and decisions are recorded for auditing and subsequent analysis. The system may also send notifications to administrators to report key events and operations. Self-Learning Mechanism: A self-learning mechanism can be used to continuously optimize anomaly detection algorithms and handling strategies. In summary, the aforementioned instance processing system solves the technical problem in related technologies of being unable to handle abnormal instances, resulting in low resource utilization. In this solution, a performance indicator data set is collected for the current first instance of the serverless architecture. The performance indicator data in the performance indicator data set can be used to accurately assess the current operating status of the first instance. Therefore, the performance indicator data set can be used to determine whether there are abnormal instances among the multiple first instances, that is, whether there are instances with abnormal operating status. If an abnormal instance exists, the abnormality level of the abnormal instance can be determined, and a handling strategy for the abnormal instance can be determined based on the abnormality level. Through the above steps, the operating status of instances in the serverless architecture can be automatically detected, preventing abnormal instances from affecting the overall service. Furthermore, if an abnormal instance is determined, a handling strategy for the abnormal instance is determined based on the abnormality level of the abnormal instance, reducing resource waste caused by abnormal instances and thereby improving resource utilization in the serverless architecture. It should be noted that, for simplicity of description, the aforementioned method embodiments are described as a series of actions. However, those skilled in the art should be aware that this application is not limited by the order of the actions described, as certain steps may be performed in a different order or simultaneously, depending on the application. Furthermore, those skilled in the art should also be aware that the embodiments described in this specification are preferred embodiments, and the actions and modules involved are not necessarily required for this application.Through the above description of the embodiments, those skilled in the art will clearly understand that the methods according to the above embodiments can be implemented using software and the necessary general-purpose hardware platform. Of course, hardware can also be used, but in many cases the former is the preferred implementation method. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk) and includes instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of the various embodiments of this application. Example 6: The embodiments of this application may provide a computer terminal, which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal can be replaced by a terminal device such as a mobile terminal. Optionally, in this embodiment, the computer terminal can be located in at least one of multiple network devices in a computer network. In this embodiment, the computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: obtaining a performance indicator data set for multiple first instances in the serverless architecture, wherein the first instances are used to execute a task of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the first instances; determining whether there are abnormal instances in the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; if there are abnormal instances in the multiple first instances, determining the abnormality level of the abnormal instance, and determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. The computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: determining whether there are abnormal instances in the multiple first instances based on the performance indicator data set includes: determining whether there are abnormal instances in the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or determining whether there are abnormal instances in the multiple first instances based on a machine learning model and the performance indicator data set.The above-mentioned computer terminal can execute the program code of the following steps in the instance processing method for the serverless architecture: Based on the preset abnormal instance judgment rule and the performance indicator data set, judging whether there are abnormal instances in multiple first instances includes: judging whether the performance indicator data in the performance indicator data set meets the performance indicator conditions in the abnormal instance judgment rule, and obtaining a judgment result; if the judgment result indicates that there are performance indicator data in the performance indicator data set that meet the performance indicator conditions in the abnormal instance judgment rule, then it indicates that there are abnormal instances in the multiple first instances, and the first instance corresponding to the performance indicator data that meets the performance indicator conditions is determined as an abnormal instance; if the judgment result indicates that there are no performance indicator data in the performance indicator data set that meet the performance indicator conditions in the abnormal instance judgment rule, then it indicates that there are no abnormal instances in the multiple first instances. The computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: Determining whether there are abnormal instances among the multiple first instances based on a machine learning model and a performance indicator data set includes: extracting features from performance indicator data in the performance indicator data set using the machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector representing the operating status of the multiple first instances; predicting an abnormality probability value of a first instance among the multiple first instances based on the target feature vector using the machine learning model to obtain an abnormality probability value; and determining whether there are abnormal instances among the multiple first instances based on the abnormality probability value. The computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: If an abnormal instance exists among the multiple first instances, determining an abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining, from the multiple first instances, the number of second instances associated with the abnormal instance based on the currently processed task information and the pending task information; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances.The above-mentioned computer terminal can execute the program code of the following steps in the instance processing method for the serverless architecture: Determining the processing strategy for the abnormal instance based on the abnormal level of the abnormal instance includes: if the abnormal level of the abnormal instance is the first level, determining the processing strategy for the abnormal instance is to re-determine the current running status of the abnormal instance after a preset time, and restart the abnormal instance if there is a problem with the current running status; if the abnormal level of the abnormal instance is the second level, determining the processing strategy for the abnormal instance is to restart the abnormal instance, wherein the abnormal impact of the second level is higher than the abnormal impact of the first level; if the abnormal level of the abnormal instance is the third level, determining the processing strategy for the abnormal instance is to eliminate the abnormal instance, wherein the abnormal impact of the third level is higher than the abnormal impact of the second level. The computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: After determining that the processing strategy for an abnormal instance is to remove the abnormal instance, the method further includes: removing the abnormal instance from the instance service pool; creating a third instance based on the currently processed task information and pending task information of the abnormal instance, and executing the currently processed task information and pending task information based on the third instance. The computer terminal can execute program code for the following steps in the instance processing method for a serverless architecture: After determining the processing strategy for the abnormal instance based on the abnormality level of the abnormal instance, the method further includes: writing the abnormal instance, performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy to a log file; and optimizing the abnormal instance determination rules and / or machine learning model based on the log file. Optionally, FIG8 is a block diagram of the structure of a computer terminal according to an embodiment of the present application. As shown in FIG8 , the computer terminal 10 may include one or more (only one is shown in FIG8 ) processors 102 and a memory 104. oThe computing terminal 10 may also include a storage controller to control and manage the memory 104. The computing terminal 10 may also include a peripheral interface to connect to a radio frequency module, an audio module, a display screen, and the like. The memory may be used to store software programs and modules, such as the program instructions / modules corresponding to the example processing method for a serverless architecture in the embodiments of the present application. The processor executes the software programs and modules stored in the memory to execute various functional applications and data processing, thereby implementing the example processing method for a serverless architecture. The memory may include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some embodiments, the memory may further include memory located remotely from the processor, which may be connected to the terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. The processor may access information and applications stored in the memory through a transmission device to perform the following steps: obtaining a performance indicator data set for multiple first instances in a serverless architecture, wherein the first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the first instances; determining whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; if there are abnormal instances among the multiple first instances, determining an abnormality level of the abnormal instance, and determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. Optionally, the processor may further execute program code for the following steps: determining whether there are abnormal instances among the multiple first instances based on the performance indicator data set includes: determining whether there are abnormal instances among the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or determining whether there are abnormal instances among the multiple first instances based on a machine learning model and the performance indicator data set.Optionally, the processor may further execute program code of the following steps: based on a preset abnormal instance determination rule and a performance indicator data set, determining whether there is an abnormal instance in the multiple first instances includes: determining whether the performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, and obtaining a judgment result; if the judgment result indicates that there is performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that there is an abnormal instance in the multiple first instances, and the first instance corresponding to the performance indicator data that satisfies the performance indicator condition is determined as an abnormal instance; if the judgment result indicates that there is no performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that there is no abnormal instance in the multiple first instances. Optionally, the processor may further execute program code for the following steps: determining whether an abnormal instance exists among the multiple first instances based on the machine learning model and the performance indicator data set includes: extracting features from the performance indicator data in the performance indicator data set using the machine learning model to obtain a target feature vector, where the target feature vector is a feature vector representing the operating status of the multiple first instances; predicting an abnormality probability value for a first instance among the multiple first instances based on the target feature vector using the machine learning model to obtain an abnormality probability value; and determining whether an abnormal instance exists among the multiple first instances based on the abnormality probability value. Optionally, the processor may further execute program code for the following steps: if an abnormal instance exists among the multiple first instances, determining an abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining, from the multiple first instances, the number of second instances associated with the abnormal instance based on the currently processed task information and the pending task information; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances. Optionally, the processor may further execute program code of the following steps: determining, based on the abnormality level of the abnormal instance, a processing strategy for the abnormal instance, including: if the abnormality level of the abnormal instance is the first level, determining the processing strategy for the abnormal instance to re-determine the current operating status of the abnormal instance after a preset time, and restarting the abnormal instance if there is a problem with the current operating status; if the abnormality level of the abnormal instance is the second level, determining the processing strategy for the abnormal instance to restart the abnormal instance, wherein the abnormal impact of the second level is higher than the abnormal impact of the first level; if the abnormality level of the abnormal instance is the third level, determining the processing strategy for the abnormal instance to eliminate the abnormal instance, wherein the abnormal impact of the third level is higher than the abnormal impact of the second level.Optionally, the processor may further execute program code for the following steps: After determining that the processing strategy for the abnormal instance is to remove the abnormal instance, the method further includes: removing the abnormal instance from the instance service pool; creating a third instance based on the currently processed task information and pending task information of the abnormal instance, and executing the currently processed task information and pending task information based on the third instance. Optionally, the processor may further execute program code for the following steps: After determining the processing strategy for the abnormal instance based on the abnormality level of the abnormal instance, the method further includes: writing the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy to a log file; and optimizing the abnormal instance determination rules and / or machine learning model based on the log file. Those skilled in the art will appreciate that the structure shown in FIG. 7 is merely illustrative, and the computer terminal 10 may also be a smartphone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile internet device (MID), a PAD, or other terminal device. FIG. 8 does not limit the structure of the electronic device. For example, the computer terminal 10 may include more or fewer components (such as a network interface, a display device, etc.) than those shown in FIG8 , or may have a configuration different from that shown in FIG8 . Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware associated with the terminal device through a program. The program can be stored in a computer-readable storage medium, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. Example 7 The present application also provides a computer-readable storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the example processing method for a serverless architecture provided in Example 1. Optionally, in this embodiment, the storage medium can be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: obtaining a performance indicator data set for multiple first instances in a serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to represent the current operating status of the multiple first instances; determining whether there are abnormal instances in the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; if there are abnormal instances in the multiple first instances, determining an abnormality level of the abnormal instance, and determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance. The storage medium is configured to store program code for executing the following steps: determining whether there are abnormal instances in the multiple first instances based on the performance indicator data set includes: determining whether there are abnormal instances in the multiple first instances based on a preset abnormal instance determination rule and the performance indicator data set; or determining whether there are abnormal instances in the multiple first instances based on a machine learning model and the performance indicator data set. The above-mentioned storage medium is configured to store program code for executing the following steps: based on a preset abnormal instance determination rule and a performance indicator data set, determining whether there is an abnormal instance in multiple first instances includes: determining whether the performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, and obtaining a judgment result; if the judgment result indicates that there is performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that there is an abnormal instance in the multiple first instances, and the first instance corresponding to the performance indicator data that satisfies the performance indicator condition is determined as an abnormal instance; if the judgment result indicates that there is no performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that there is no abnormal instance in the multiple first instances. The above-mentioned storage medium is configured to store program code for executing the following steps: judging whether there are abnormal instances in multiple first instances based on the machine learning model and the performance indicator data set, including: extracting features of the performance indicator data in the performance indicator data set through the machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector that characterizes the operating status of the multiple first instances; predicting the abnormal probability value of the first instance in the multiple first instances based on the target feature vector through the machine learning model to obtain an abnormal probability value; judging whether there are abnormal instances in the multiple first instances based on the abnormal probability value.The storage medium is configured to store program code for executing the following steps: if an abnormal instance exists among multiple first instances, determining the abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining the number of second instances associated with the abnormal instance from the multiple first instances based on the currently processed task information and the pending task information; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of second instances. The storage medium is configured to store program code for executing the following steps: Determining a handling strategy for the abnormal instance based on the abnormality level of the abnormal instance includes: if the abnormality level of the abnormal instance is level 1, determining the handling strategy for the abnormal instance to re-determine the current operating status of the abnormal instance after a preset time, and restarting the abnormal instance if there is a problem with the current operating status; if the abnormality level of the abnormal instance is level 2, determining the handling strategy for the abnormal instance to restart the abnormal instance, wherein the abnormality impact of the second level is higher than the abnormality impact of the first level; if the abnormality level of the abnormal instance is level 3, determining the handling strategy for the abnormal instance to remove the abnormal instance, wherein the abnormality impact of the third level is higher than the abnormality impact of the second level. The storage medium is configured to store program code for executing the following steps: After determining the handling strategy for the abnormal instance to remove the abnormal instance, the method further includes: removing the abnormal instance from the instance service pool; creating a third instance based on the currently processed task information and the pending task information of the abnormal instance, and executing the currently processed task information and the pending task information according to the third instance. The storage medium is configured to store program code for executing the following steps: after determining a handling strategy for an abnormal instance based on its abnormality level, the method further includes: writing the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the handling strategy to a log file; and optimizing the abnormal instance determination rules and / or machine learning model based on the log file. The serial numbers of the embodiments of this application are for descriptive purposes only and do not represent the merits of the embodiments. In the embodiments of this application, the description of each embodiment has its own emphasis. For portions not described in detail in a particular embodiment, reference should be made to the relevant descriptions of other embodiments. It should be understood that the disclosed technical content in the several embodiments provided herein can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is merely a logical functional division. In actual implementation, other divisions may be used. For example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not implemented.Furthermore, the coupling, direct coupling, or communication connection shown or discussed may be through interfaces, indirect coupling, or communication connection between units or modules, and may be electrical or otherwise. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the objectives of the present embodiment as needed. Furthermore, the functional units in the various embodiments of the present application may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. These integrated units may be implemented in either hardware or software functional units. If the integrated units are implemented as software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for causing a computer device (such as a personal computer, server, or network device) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), a removable hard drive, a magnetic disk, or an optical disk. The above are merely preferred embodiments of the present application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present application, and such improvements and modifications should also be considered within the scope of protection of the present application.Industrial Applicability The instance processing method and device for a serverless architecture provided in the embodiments of the present application collect a performance indicator data set of the current first instance of the serverless architecture. The performance indicator data in the performance indicator data set can be used to accurately evaluate the current operating status of the first instance. Therefore, the performance indicator data set can be used to determine whether there are abnormal instances among multiple first instances, that is, whether there are instances with abnormal operating status. If there are abnormal instances, the abnormality level of the abnormal instance can be determined, and the processing strategy for the abnormal instance can be determined according to the abnormality level. The above steps can automatically detect the operating status of instances in the serverless architecture, avoid the situation where abnormal instances affect the overall service, and when it is determined that there are abnormal instances, the processing strategy for the abnormal instance is determined according to the abnormality level of the abnormal instance, which can reduce the waste of resources caused by abnormal instances, thereby achieving the effect of improving the resource utilization of the serverless architecture.

Claims

Claims 1. A method for processing instances in a serverless architecture, comprising: Obtain a performance indicator data set of multiple first instances in a serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set are used to characterize the current operating status of the multiple first instances; determine whether there are abnormal instances among the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; if there are abnormal instances among the multiple first instances, determine the abnormality level of the abnormal instance, and determine a processing strategy for the abnormal instance based on the abnormality level of the abnormal instance.

2. The method according to claim 1, wherein: The determining, based on the performance indicator data set, whether there is an abnormal instance in the multiple first instances includes: determining, based on a preset abnormal instance determination rule and the performance indicator data set, whether there is the abnormal instance in the multiple first instances; or determining, based on a machine learning model and the performance indicator data set, whether there is an abnormal instance in the multiple first instances.

3. The method according to claim 2, wherein: The determining, based on a preset abnormal instance determination rule and the performance indicator data set, whether the abnormal instance exists in the multiple first instances includes: determining whether the performance indicator data in the performance indicator data set satisfies the performance indicator condition in the abnormal instance determination rule, and obtaining a determination result; if the determination result indicates that there is performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that the abnormal instance exists in the multiple first instances, and the first instance corresponding to the performance indicator data that satisfies the performance indicator condition is determined as the abnormal instance; if the judgment result indicates that there is no performance indicator data in the performance indicator data set that satisfies the performance indicator condition in the abnormal instance determination rule, then it indicates that the abnormal instance does not exist in the multiple first instances.

4. The method according to claim 2, wherein: The determining, based on the machine learning model and the performance indicator data set, whether there is an abnormal instance among the plurality of first instances includes: 24 Performing feature extraction on the performance indicator data in the performance indicator data set using the machine learning model to obtain a target feature vector, wherein the target feature vector is a feature vector representing the operating status of the multiple first instances; predicting an abnormality probability value of a first instance in the multiple first instances based on the target feature vector using the machine learning model to obtain an abnormality probability value; and determining whether the abnormal instance exists in the multiple first instances based on the abnormality probability value.

5. The method according to claim 1, wherein: Determining the abnormality level of the abnormal instance includes: obtaining currently processed task information and pending task information of the abnormal instance; determining, based on the currently processed task information and the pending task information, the number of second instances associated with the abnormal instance from the multiple first instances; and determining the abnormality level of the abnormal instance based on the performance indicator data set and the number of the second instances.

6. The method according to claim 1, wherein: The determining, based on the abnormality level of the abnormal instance, a handling strategy for the abnormal instance includes: if the abnormality level of the abnormal instance is a first level, determining that the handling strategy for the abnormal instance is to re-determine the current operating status of the abnormal instance after a preset time, and restart the abnormal instance if there is a problem with the current operating status; if the abnormality level of the abnormal instance is a second level, determining that the handling strategy for the abnormal instance is to restart the abnormal instance, wherein the abnormality impact of the second level is higher than the abnormality impact of the first level; and if the abnormality level of the abnormal instance is a third level, determining that the handling strategy for the abnormal instance is to eliminate the abnormal instance, wherein the abnormality impact of the third level is higher than the abnormality impact of the second level.

7. The method according to claim 6, wherein: After determining that the processing strategy for the abnormal instance is to remove the abnormal instance, the method further includes: removing the abnormal instance from the instance service pool; creating a third instance based on the currently processed task information and the pending task information of the abnormal instance, and executing the currently processed task information and the pending task information based on the third instance.

8. The method according to claim 2, wherein: According to the abnormal level of the abnormal instance, determine After determining the processing strategy for the abnormal instance, the method further includes: writing the abnormal instance, the performance indicator data corresponding to the abnormal instance, the abnormality level of the abnormal instance, and the processing strategy into a log file; and optimizing the abnormal instance determination rule and / or the machine learning model based on the log file.

9. A method for processing instances in a serverless architecture, comprising: Obtain a performance indicator data set of multiple first instances in a serverless architecture uploaded by a client, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set are used to characterize the current operating status of the multiple first instances; determine, in a cloud server, based on the performance indicator data set whether there are abnormal instances among the multiple first instances, wherein the abnormal instance is an instance with a problem in the current operating status; if the abnormal instance exists among the multiple first instances, determine the abnormality level of the abnormal instance, and determine a processing strategy for the abnormal instance based on the abnormality level of the abnormal instance; and return the processing strategy to the client.

10. An instance processing device for a serverless architecture, comprising: A first acquisition unit is configured to acquire a performance indicator data set of multiple first instances in a serverless architecture, wherein the multiple first instances are used to execute tasks of a target object, and the performance indicator data in the performance indicator data set is used to characterize the current operating status of the multiple first instances; a first judgment unit is configured to judge whether there are abnormal instances in the multiple first instances based on the performance indicator data set, wherein the abnormal instance is an instance with a problem in the current operating status; and a first determination unit is configured to determine the abnormality level of the abnormal instance if the abnormal instance exists in the multiple first instances, and determine a processing strategy for the abnormal instance based on the abnormality level of the abnormal instance.

11. A method for processing instances in a serverless architecture, comprising: The second acquisition unit is configured to acquire a performance indicator data set of multiple first instances in the serverless architecture uploaded by the client, wherein the multiple first instances are used to execute the tasks of the target object, and the performance indicator data in the performance indicator data set are used to characterize the current running status of the multiple first instances; the second judgment unit is configured to judge whether there is an abnormal instance in the multiple first instances based on the performance indicator data set in the cloud server, wherein the abnormal instance is an abnormal instance in the current running status. a second determining unit configured to determine an abnormality level of the abnormal instance if the abnormal instance exists in the multiple first instances, and determine a processing strategy for the abnormal instance based on the abnormality level of the abnormal instance; and a returning unit configured to return the processing strategy to the client.

12. A computer-readable storage medium comprising a stored program, wherein: When the program is running, the device where the storage medium is located is controlled to execute the instance processing method for the serverless architecture as described in any one of claims 1 to 9.

13. An electronic device, comprising: a memory storing an executable program; A processor is used to run the program, wherein the program executes the instance processing method for the serverless architecture described in any one of claims 1 to 9 when running.

14. A computer program product, comprising a stored computer program, wherein when the computer program is executed by a processor, the computer program implements the instance processing method for a serverless architecture according to any one of claims 1 to 9. 27

Images