Resource object management method and device and storage medium

Abstract

The embodiment of the invention provides a resource object management method and device and a storage medium. In the embodiment of the invention, the management method corresponding to the management requirement of the target resource object is determined by utilizing the powerful processing capability of the language model; on the basis of the powerful processing capacity of the language model, the predefined resource object management strategy and the resource object management method determined by the language model are used for fusion processing, and the predefined resource object management strategy is adjusted in time according to the strategy definition rule defined by the user, so that the user experience is improved. The resource object management method determined by the language model is fused by using the predefined resource object management strategy, so that the defect that the language model cannot respond to the adjustment of the resource object management strategy in time can be made up, and the flexibility and reliability of resource object management can be improved.

Description

Technical Field

[0001] The present application relates to the field of cloud service technology, and in particular to a resource object management method, device and storage medium. Background Art

[0002] Container cluster management systems, such as Kubernetes (abbreviated as K8s), are used to manage containerized workloads and services. It provides a powerful application programming interface (API) for deploying and managing applications, making it easier to maintain and expand applications.

[0003] The operation of applications deployed on the container cluster management system depends on the resource objects in the container cluster management system. The performance of resource objects has a significant impact on the performance of applications. For the container cluster management system, the introduction of artificial intelligence (AI) technology can help users monitor and diagnose applications deployed on the container cluster and provide management suggestions for resource objects in the container cluster management system.

[0004] However, the AI ​​model is trained based on historical data and cannot correctly handle the resource object management policy defined after the model is trained. The resource object management policy itself is uncertain or changeable and is defined by the user. Therefore, the traditional method of using AI models to determine the management policy for resource objects has poor flexibility and reliability and cannot adapt to the changing resource object management policy adjustments. Summary of the invention

[0005] Multiple aspects of the present application provide a resource object management method, device and storage medium to improve the flexibility and reliability of resource object management.

[0006] The present application provides a resource object management method, including:

[0007] Obtain management requirements for target resource objects in the container cluster management system;

[0008] Determining text description information of a management method for the target resource object according to the management requirement by using a language model;

[0009] Generating a first executable command corresponding to the management method according to the text description information;

[0010] According to a predefined resource object management policy, the first executable command is fused to obtain a fused second executable command;

[0011] The target resource object is managed according to the second executable command.

[0012] The embodiment of the present application further provides a computing device, comprising: a memory and a processor; wherein the memory is used to store a computer program;

[0013] The processor is coupled to the memory and is used to execute the computer program to perform the steps in the resource object management method.

[0014] An embodiment of the present application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the steps in the above-mentioned resource object management method.

[0015] In an embodiment of the present application, the powerful processing capability of the language model is utilized to determine the management method corresponding to the management demand for the target resource object; and based on the powerful processing capability of the language model, the resource object management method determined by the language model is integrated with a predefined resource object management strategy. Since the predefined resource object management strategy is adjusted in time according to the user-defined policy definition rules, therefore, the resource object management method determined by the language model is integrated with the predefined resource object management strategy, which can make up for the defect that the language model cannot respond to the adjustment of the resource object management strategy in time, and help to improve the flexibility and reliability of resource object management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used 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 descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 Define a process diagram for resource object management strategy;

[0018] Figure 2 and Figure 3 A schematic diagram of the internal architecture of the resource object management system provided in an embodiment of the present application;

[0019] Figure 4 A schematic diagram of a process flow of a resource object management method provided in an embodiment of the present application;

[0020] Figure 5 A flowchart of another resource object management method provided in an embodiment of the present application;

[0021] Figure 6 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0023] In view of the technical problem that the traditional method of using AI models to determine the management strategy for resource objects has poor flexibility and reliability and cannot adapt to the changing resource object management strategy adjustments, some embodiments of the present application provide a solution, the main ideas of which include: using the powerful processing capabilities of the language model to determine the management method corresponding to the management needs of the target resource object; and based on the powerful processing capabilities of the language model, using pre-defined resource object management strategies to integrate the resource object management methods determined by the language model. Since the pre-defined resource object management strategies are adjusted in time according to user-defined policy definition rules, therefore, using pre-defined resource object management strategies to integrate the resource object management methods determined by the language model can make up for the defect that the language model cannot respond to resource object management strategy adjustments in time, which helps to improve the flexibility and reliability of resource object management.

[0024] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0025] It should be noted that the same reference numerals denote the same objects in the following drawings and embodiments, and therefore, once an object is defined in one drawing or embodiment, it does not need to be further discussed in the subsequent drawings and embodiments.

[0026] In the embodiment of the present application, the Platform as a Service (PaaS) platform can support PaaS products through key capabilities such as unified application management, heterogeneous resource management, and intelligent operation and maintenance. An application instance is a unit of application release in the PaaS platform. It is an instance after a Charts package is deployed to the platform, which contains various resources.

[0027] A resource is an endpoint in the container cluster management system API, which stores a collection of API objects of a certain category (also called resource objects). Objects (resource objects) in the container cluster management system are persistent entities in the container cluster management system. Resource objects can include: resource objects natively supported by the container cluster management system and custom resource (CR) objects.

[0028] Among them, the resource objects natively supported by the container cluster management system may include: container groups (pods, etc.), services, deployments, namespaces, events, Statefulset and other workload resources.

[0029] The Custom Resource (CR) object is an extension of the container cluster management system API. The CR object is a resource object entity defined by the Custom Resource Definition (CRD). Users can define custom resources through CRD. The CR object can be stored in the API Server component of the container cluster management system in the form of a YAML file. The CR defines the attribute information of the CR object.

[0030] Since an application is an instance of a Charts package deployed to the platform, it contains various resource objects. Therefore, the operation of the application depends on the resource objects. The management and operation of the application can be achieved through the management and operation of the resource objects.

[0031] The resource object management method provided in the embodiment of the present application is mainly divided into two stages: one is the resource object management strategy definition stage, and the other is the resource object management stage of integrating the language model to implement the resource object management strategy. The resource object management strategy definition stage is exemplarily described below.

[0032] Figure 1 A schematic diagram of the process of defining resource object management strategies. Figure 1 As shown, in order to define resource object management policies, the metadata of resource objects in the container cluster management system and the access permissions associated with the resource objects can be obtained. The metadata of resource objects includes metadata of multiple dimensions, including but not limited to: infrastructure description information, metadata related to the application corresponding to the resource object, operation and maintenance data of the resource object, observable data, traffic management related data, and model definitions related to delivery and security.

[0033] Among them, the infrastructure description information is used to describe the computing environment that the resource object itself relies on, such as the machine model used by the resource object, the rack it is located in, and the region where it is located. The metadata related to the application corresponding to the resource object is used to describe the resource preference type of the application, such as whether it is central processing unit (CPU) intensive or memory intensive. The operation and maintenance data of the resource object includes: the node where the resource object runs, the computing resources used (such as CPU) and memory, etc. Observable data includes the fusion data formed by the log data, monitoring data, and tracking data of the resource object.

[0034] The access rights associated with the above-mentioned resource objects mainly refer to the access rights of resource objects, and different resource objects have different access rights. In some embodiments, the access rights associated with resource objects may include: role-based access rights control (i.e., Role Based Access Control, RBAC) and / or attribute-based access rights control (Attribute Based Access Control, ABAC) based on the attributes of resource objects. Among them, the role refers to the user role, and roles can be created to grant roles access rights to resource objects. A role can correspond to one or more users. Multiple refers to two or more. Different roles can have the same or different access rights to the same resource object. The same role can also have the same or different access rights to different resource objects, which can be flexibly set according to user needs.

[0035] Access rights control based on resource object attributes is a permission definition based on resource object attributes. It can grant access rights to resource object attributes at the resource object granularity. The same attribute of different resource objects can have the same or different access rights; different attributes of the same resource object can also have the same or different access rights. The specific settings can be flexibly made according to user needs.

[0036] The attributes of a resource object refer to information that reflects the attributes of the resource object, including but not limited to: status information of the resource object, workload resource information (also referred to as sub-resource information), hardware resource demand information, and scheduling attribute information.

[0037] Among them, the status information of the resource object can reflect the health status of the resource object, and whether the resource object has reached the final state (i.e., the ready state) can be determined based on the status information of the resource object. The workload of the resource object refers to the application running on the resource object on the container cluster management system, and the workload can be run in a container group (such as a Pod) in the container cluster management system. In the container cluster management system, there is no need to directly manage each Pod, and a group of Pods can be managed using load resources. These workload resource configuration controllers ensure that the state of the workload resources reaches the desired state.

[0038] Hardware resource demand information reflects the demand of resource objects for hardware resources. Hardware resources include but are not limited to: computing resources (such as the number of CPU cores), storage resources (such as memory) and disk resources.

[0039] The scheduling attribute information of resource objects is used to reflect the scheduling requirements of resource objects, which can affect the deployment of resource objects on worker nodes. The scheduling attribute information of resource objects includes but is not limited to: the label of the container group running the workload of the resource object, scheduling affinity rules, tolerance rules, and node selector configuration, etc.

[0040] Scheduling affinity rules refer to the affinity rules between container groups and other container groups. Application instances with affinity can be deployed on the same worker node. Container groups without affinity cannot be deployed on the same compute node. Labels are used to indicate the affinity between container groups and other container groups. Tolerance is a key-value attribute data defined on a container group (such as a pod) and is used to configure the taints of the compute nodes that it can tolerate. The scheduler can only schedule container groups to compute nodes that can tolerate node taints for the application instance.

[0041] The node selector configuration is used to select a working node. The scheduler schedules the node selector configuration label matching and schedules the container group to the target working node.

[0042] Among them, the metadata of the resource object and the access rights associated with the resource object are saved in different documents, so that a group of documents can be obtained. The metadata of different resource objects can also be saved in multiple documents. In an embodiment of the present application, in order to facilitate description and distinction, the above-mentioned document that saves the metadata of the resource object is defined as the first document; and the document that saves the access rights associated with the resource object is defined as the second document. Among them, the first document can be one or more. The second document can also be one or more. Multiple refers to 2 or more. The access rights associated with the resource object may include: role-based access rights control and access rights control based on the attributes of the resource object. The role-based access rights control and access rights control based on the attributes of the resource object can also be saved in different second documents.

[0043] Based on the first document storing the metadata of the resource object and the second document storing the access rights associated with the resource object, a reference relationship between the first document and the second document can be established, and the first document and the second document can be aggregated according to the reference relationship between the first document and the second document to obtain a context model of the resource object. Figure 1 The context model generation unit generates a context model.

[0044] That is, the context model of a resource object is obtained by establishing a reference relationship between documents for a group of documents corresponding to the resource object and aggregating the group of documents. The group of documents stores the metadata of the resource object and / or the access rights associated with the resource object. Therefore, the context model of a resource object can also be implemented as a document, which may include at least one of the following data: the association relationship between different resource objects, the metadata of the resource object, and the access rights associated with the resource object. Preferably, the document corresponding to the context model includes: the association relationship between different resource objects, the metadata of the resource object, and the access rights associated with the resource object.

[0045] Further, if Figure 1 As shown, a resource object management strategy can be defined based on the context model of the resource object. The resource object management strategy refers to a strategy for managing resource objects, including but not limited to the following management strategies: (1) container group scaling strategy, which is used to determine the number of container group replicas required for the effective operation of the application; (2) load balancing strategy, which performs load balancing by analyzing network traffic and determining to route traffic to different container groups; (3) resource allocation optimization strategy for container groups; (4) resource object abnormality recovery strategy; and (5) resource object abnormality identification strategy, etc.

[0046] Among them, the management strategies used for resource object operation and maintenance include but are not limited to: (1) resource object anomaly detection and alarm strategy; (2) resource object anomaly recovery strategy; (3) container group resource allocation optimization strategy; (4) container cluster capacity planning strategy; (5) container group integration and deployment strategy corresponding to the application; and (6) resource object security detection and vulnerability identification strategy, etc.

[0047] In the embodiments of the present application, the specific implementation method of defining the resource object management policy according to the context model of the resource object is not limited. In some embodiments, the resource object management policy can be manually defined according to the context model of the resource object and the user's management requirements for the resource object. Alternatively, the definition rules of the resource object management policy can also be set on the computing device, and the computing device can define the resource object management policy according to the context model of the resource object and the definition rules of the resource object management policy, etc.

[0048] Specifically, according to the definition rule of the resource object management policy, the resource object involved in the definition rule can be obtained from the context model of the resource object, and the management policy of the resource object involved in the definition rule can be defined according to the definition rule.

[0049] In the above embodiment, the resource object management policy is defined based on the context model of the resource object, which can take into account more dimensional information of the resource object, has stronger policy reasoning capabilities, and can optimize the resource object management policy, thereby helping to optimize the resource object performance when the resource object management policy is applied to resource object management.

[0050] The resource object management policy obtained in the above embodiment can be introduced into the policy engine. The policy engine is a tool or platform that can manage resource objects in the container cluster management system, which can help users of the container cluster management system ensure the security, reliability and stability of the container cluster. It can manage resource objects according to the resource object management policy defined by the user, and perform abnormal repair when necessary.

[0051] In an embodiment of the present application, in order to solve the technical problem that the traditional method of using AI models to determine the management strategy for resource objects has poor flexibility and cannot adapt to the changing resource object management strategy adjustments, a resource object management method that integrates a policy engine and a language model is proposed. Based on the powerful processing capabilities of the language model, a policy engine based on real-time changes and changes is introduced, which makes up for the defect that the language model cannot respond to the resource object management strategy adjustments in a timely manner, and helps to improve the flexibility and reliability of resource object management.

[0052] In the embodiment of the present application, the language model is mainly used in the field of natural language processing (NLP). The goal of these models is to understand and generate human language. In order to achieve this goal, the model needs to be trained on a large amount of text data to learn various patterns and structures of the language. In this embodiment, the language model can output text information adapted to the prompt word under the trigger of the input prompt word.

[0053] In an embodiment of the present application, the language model may be a neural network model. In some embodiments, when the number of model parameters of the language model is large, for example, the number of parameters of the language model is millions, hundreds of millions, billions or even more, the language model may also be referred to as a large language model (LLM). In an embodiment of the present application, a large language model is defined as a neural network model whose number of model parameters meets a preset parameter number range. Among them, the number of model parameters corresponding to the preset parameter number range is very large, which can be millions, hundreds of millions, billions or even more, and the specific value can be determined by the standard in the field of artificial intelligence (AI).

[0054] Among them, the large language model can be trained through two stages: pre-training and fine-tuning. In the pre-training stage, the model is trained on large-scale general text data to learn the basic structure of the language and various common sense. Then, in the fine-tuning stage, the model is further trained on smaller and more specific field data sets. Fine-tuning allows the model to better understand and generate the language of this specific field, so as to better complete specific tasks.

[0055] In this embodiment, the large language model can use the historical management requirements in the container cluster management system (such as the K8s system) as the model input and the text description information of the historical management method corresponding to the historical management requirements as the actual results corresponding to the model input to perform model training in the fine-tuning stage to obtain the language model provided in this embodiment, that is, the large language model. The loss function of the language model can be determined by the difference between the text description prediction information of the historical management method corresponding to the historical management requirements output by the language model in the model training stage and the actual results corresponding to the above model input (that is, the text description information of the historical management method corresponding to the historical management requirements).

[0056] In the embodiments of the present application, the specific implementation architecture of the large language model is not limited. The large language model can adopt a language model architecture with permission to use and open source, or a self-developed language model architecture. Optionally, the large language model can be a generative pre-training (GPT) model, a general language model (GLM), or a variation of these models.

[0057] Combine the following Figure 2 and Figure 3 The internal architecture diagram of the resource object management system provided is used to exemplify the resource object management method provided in the embodiment of the present application.

[0058] Combination Figure 2 and Figure 3 When a user needs to manage a resource object, he or she can provide management requirement description information for the target resource object. The management requirement description information is used to describe the user's management requirement for the target resource object and can be text information.

[0059] Combination Figure 2 and Figure 3 , the resource object management system can obtain the management requirement description information of the target resource object provided by the user. Optionally, the resource object management system can pre-process the management requirement description information. Figure 3 As shown, the management requirement description information can be preprocessed by the prompt processing unit in the resource object management system, wherein the management requirement description information is the prompt of the language model.

[0060] Optionally, preprocessing the management requirement description information may include: removing unnecessary formats in the management requirement description information; performing sentence segmentation and word segmentation on the management requirement description information to obtain words corresponding to the management requirement description information; further, normalizing the words to convert the words into a standard form and removing unnecessary words in the standard form of the words, etc.

[0061] In this embodiment, the management requirement description information provided by the user can be used as management requirements and input into the language model, and the language model can be used to determine the text description information of the management method for the target resource object based on the management requirement description information. During the language model training phase, the language model is trained using the historical management requirements in the container cluster management system (such as the K8s system) as the model input and the text description information of the historical management method corresponding to the historical management requirements as the model output. Therefore, in this embodiment, the management requirement description information can be input into the trained language model, and the output result is the text description information of the management method for the target resource object.

[0062] Specifically, Figure 3 As shown, an agent execution unit may be provided. The agent execution unit is used to call the language model and input the management requirements into the language model.

[0063] Since the language model outputs the text description information of the management method for the target resource object, the computing device needs to execute a computer program or command to manage the target resource object. Therefore, in this embodiment, an executable command corresponding to the management method for the target resource object can also be generated based on the text description information of the management method for the target resource object. An executable command refers to a computer instruction that can be executed by a computing device. Figure 3 As shown, this step can be done by Figure 3 The command generator in is completed. The command generator refers to a software function module, plug-in or tool that converts text into computer instructions.

[0064] In the embodiments of the present application, the specific implementation form of generating the executable command corresponding to the management method for the target resource object according to the text description information of the management method for the target resource object is not limited. In some embodiments, a code example of the management method can be obtained from the text description information of the management method for the target resource object; the code example can be modified according to the management requirements for the target resource object to obtain an executable command, etc. For example, the identifier of the target resource object can be obtained from the management requirements for the target resource object; and the identifier of the resource object in the code example can be modified to the identifier of the target object, thereby obtaining an executable command, etc.

[0065] The executable command is a computer instruction converted from the text description information of the management method for the target resource object output by the language model. The text description information of the management method for the target resource object determined by the language model is learned by the language model during the training phase. The language model has not learned the resource object management strategy after the language model training is completed. Therefore, the management method for the target resource object determined by the language model has certain limitations and may be different from the resource object management strategy defined by the user for the target resource object.

[0066] To solve this problem, combined Figure 2 and Figure 3 , the executable command can be input into the policy engine. The policy engine can fuse the executable command according to the predefined resource object management policy to obtain the executable command after fusion processing. In the embodiment of the present application, fusing the executable command means fusing the resource object management policy into the executable command, including but not limited to: deleting some instructions of the executable command according to the resource object management policy, supplementing the executable command, and modifying some instructions of the executable command, etc. One or more. Multiple refers to 2 or more.

[0067] Optionally, the policy engine may obtain the management policy of the target resource object from the predefined resource object management policy. Further, it may be determined whether the executable command is consistent with the management policy of the target object. If not, the executable command is fused according to the management policy of the target resource object to obtain an executable command after fusion. In the embodiment of the present application, for the convenience of description and distinction, the executable command before fusion is defined as the first executable command; the executable command after fusion is defined as the second executable command.

[0068] In some embodiments, the policy engine may also fuse the first executable command according to a predefined resource object management policy and a context model of the resource object in the container cluster management system to obtain a second executable command after the fusion processing. For the description of the context model of the resource object, please refer to the relevant content of the above embodiment, which will not be repeated here. Since the context of the resource object provides the real-time status of the resource object, the execution of the resource object management policy is adjusted in time according to the definition rules defined by the user. Therefore, according to the context model of the resource object and the predefined resource object management policy, the executable command is fused and processed. The obtained executable command after the fusion processing responds to the adjustment of the resource object management policy and also takes into account the real-time status of the resource object. Therefore, it can make up for the defect that the language model cannot respond to the resource object management policy and the status adjustment of the resource object in time, which helps to improve the flexibility and reliability of resource object management.

[0069] Some optional implementation methods of fusing the first executable command according to a predefined resource object management policy and a context model of a resource object in a container cluster management system are exemplarily described below.

[0070] Implementation method 1: It is possible to determine whether it is necessary to jointly manage other resource objects associated with the target resource object during the management of the target resource object based on the association relationship between different resource objects and the predefined resource object management policy. Specifically, it is possible to determine other resource objects associated with the target resource object based on the association relationship between resource objects; and determine whether there is a policy for jointly managing the target resource object and other resource objects associated with it in the predefined resource object management policy; if the judgment result is yes, it is determined whether it is necessary to jointly manage other resource objects associated with the target resource object during the management of the target resource object. Further, a management policy for jointly managing the target resource object and other resource objects associated with it can be obtained from the predefined resource object management policy. Further, based on the management policy for jointly managing the target resource object and other resource objects associated with it, a command for jointly managing the target resource object and other resource objects associated with it can be supplemented in the first executable command to obtain a second executable command after fusion processing.

[0071] Implementation method 2: The access rights to the attributes of the target resource object can be obtained from the access rights associated with the resource object; based on the access rights to the attributes of the target resource object, the commands corresponding to the attributes without access rights are deleted from the first executable command to obtain a second executable command after fusion processing.

[0072] Implementation method 3: From the access rights associated with the resource object, obtain the access rights of the role of the user who makes the management request to the target resource object; and based on the access rights of the role of the user to the target resource object, select the command corresponding to the management item to which the role of the user has no access rights in the first executable command to obtain the second executable command after fusion processing.

[0073] In implementation mode 3, management items refer to management items of target resource objects, which can be distinguished by metadata and / or attributes of target resource objects. For example, different attributes of target resource objects are different management items, and different states of target resource objects are different management items.

[0074] Implementation method 4: Obtain the metadata of the target resource object from the metadata of the resource object; obtain the management policy corresponding to the metadata of the target resource object from the resource object management policy; and fuse the first executable command according to the management policy corresponding to the metadata of the target resource object to obtain a second executable command after fusion.

[0075] The above embodiments 1-4 only exemplarily give some optional implementation methods for fusing the first executable command according to the predefined resource object management strategy and the context model of the resource object in the container cluster management system, and are not limited to the above four implementation methods, nor are they limited to the above four implementation methods. The above embodiments 1-4 can be implemented one by one, or multiple of them can be implemented in combination. Multiple refers to 2 or more. When multiple methods are implemented in combination, multiple implementation methods are used to fuse the first executable command together to obtain a second executable command after fusion processing.

[0076] Furthermore, the target resource object can be managed according to the second executable command after the fusion process. The specific implementation method of managing the target resource object is determined by the second executable command after the fusion process.

[0077] Specifically, the target management path for managing the target resource object can be determined according to the second executable command after the fusion processing. If the target management path involves an existing management module, the management module involved in the target management path can be called to manage the target resource object. Figure 3 As shown, if the target management path involves a diagnostic analyzer, the diagnostic analyzer can be called to perform diagnostic analysis on the target resource object to obtain a diagnostic analysis result; and / or, if the target management path involves an audit analyzer, the audit analyzer can be called to perform audit processing on the log data of the target resource object to obtain an audit analysis result, etc.

[0078] If the target management path does not involve an existing management module, the executable command after fusion processing can be directly executed to achieve the management of the target resource object.

[0079] The above executable command execution steps can be performed by Figure 3 The command executor shown is completed. The command executor is a software function module, plug-in or tool for executing computer instructions. Figure 3 As shown, the second executable command after fusion processing can be input into the command executor; the command executor can determine the target management path for managing the target resource object according to the second executable command after fusion processing; and call the management module involved in the target management path to manage the target resource object.

[0080] In this embodiment, a resource object management method that integrates a policy engine and a language model is proposed, and the powerful processing capability of the language model is used to determine the management method corresponding to the management demand for the target resource object; and based on the powerful processing capability of the language model, the resource object management method determined by the language model is integrated with a predefined resource object management policy. Since the predefined resource object management policy is adjusted in time according to the user-defined policy definition rules, the resource object management method determined by the language model is integrated with the predefined resource object management policy, which can make up for the defect that the language model cannot respond to the adjustment of the resource object management policy in time, and help to improve the flexibility and reliability of resource object management.

[0081] In an embodiment of the present application, in order to further improve the reliability of resource object management, the management result of resource management of the target resource object according to the second executable command after fusion processing can be fed back to the input of the language model to supplement the resource demand description information provided by the user, thereby improving the prompts of the language model and improving the reliability of the text description information of the management method output by the language model.

[0082] Specifically, before inputting the management result description information of resource management of the target resource object according to the second executable command after fusion processing in the current round and the resource requirement description information provided by the user as the management requirement for the target resource management object into the language model, it is also possible to judge whether the management result of managing the target resource object according to the second executable command after fusion processing meets the management requirement corresponding to the current round based on the management requirement description information and the management result description information; if not, the management result description information of resource management of the target resource object according to the second executable command after fusion processing in the current round and the resource requirement description information provided by the user are input into the language model as the management requirement for the target resource management object, and the above language model is cyclically executed to determine the text description information of the management method for the target resource object according to the management requirement; generate an executable command corresponding to the management method according to the text description information; fuse the executable command according to the pre-defined resource object management policy to obtain the executable command after fusion processing; and manage the target resource object according to the executable command after fusion processing, until the management result of managing the target resource object according to the executable command after fusion processing meets the management requirement.

[0083] Therefore, the management requirements input into the language model in the first round may be the management requirements description information for the target resource object provided by the user; the management requirements input into the language model in other rounds may be the management requirements description information for the target resource object provided by the user and the management result description information obtained in the previous round. Accordingly, assuming that the above-mentioned current round is the Nth round, N≥2, and is an integer, and the management result of managing the target resource object obtained in the (N-1)th round does not meet the management requirements corresponding to the (N-1)th round, then in the process of executing the resource object management method provided by the embodiment of the present application in the Nth round, obtaining the management requirements for the target resource object can be implemented as follows: obtaining the management requirements description information for the target resource object provided by the user; and obtaining the management result description information obtained in the (N-1)th round. Among them, the management result description information obtained in the (N-1)th round is: using the fused executable command (defined as the third executable command) obtained in the process of executing the resource object management method in the (N-1)th round to manage the target resource object. Furthermore, the management requirement description information provided by the user and the management result description information obtained in the (N-1)th round may be used as management requirements.

[0084] In some embodiments of the present application, Figure 3 As shown, the management requirement description information provided by the user and the text description information of the management method for the target resource object output by the language model can also be saved. Figure 3 The proxy execution unit can call the history component to save and maintain the history, for example, to save and maintain the management requirement description information provided by the user and the text description information of the management method for the target resource object output by the language model. In this way, the next time a user provides the same resource requirement description information, it will no longer respond by calling the language model, but can determine the text description information of the management method for the target resource object by retrieving the history, which can improve the speed of determining the management method for the target resource object. In the policy execution stage, the text description information of the historical management method corresponding to the management requirement description information can be directly used to generate executable commands to manage the target resource object.

[0085] The above embodiment provides an exemplary description of the resource object management method provided by the embodiment of the present application from the perspective of the internal architecture of the resource object management system. Each module or component in the above resource object management system can be deployed on the same computing device or on different computing devices. The computing device can be any device with data processing functions, such as a server, a computer, a workstation, etc., and of course it can also be a cloud-based server array, or a virtual machine (Virtual Machine, VM) running in a cloud-based server array, etc. The following is an exemplary description of the resource object management method provided by the embodiment of the present application from the perspective of a computing device.

[0086] Figure 4 The following is a flow chart of the resource object management method provided in the embodiment of the present application. Figure 4 As shown, the resource object management method mainly includes:

[0087] 401. Obtain management requirements for a target resource object in a container cluster management system.

[0088] 402. Determine text description information of the management method for the target resource object according to management requirements using the language model.

[0089] 403. Generate a first executable command corresponding to the management method according to the text description information.

[0090] 404. According to a predefined resource object management policy, the executable commands are fused to obtain a fused second executable command.

[0091] 405. Manage the target resource object according to the second executable command.

[0092] The resource object management method provided in the embodiment of the present application is mainly divided into two stages: one is the resource object management strategy definition stage, and the other is the resource object management stage of integrating the language model to implement the resource object management strategy. The resource object management strategy definition stage is exemplarily described below.

[0093] In order to define resource object management policies, metadata of resource objects and access rights associated with resource objects in the container cluster management system may be obtained. For descriptions of metadata of resource objects and access rights associated with resource objects, please refer to the relevant contents of the above system embodiments, which will not be repeated here.

[0094] Among them, the metadata of the resource object and the access rights associated with the resource object are saved in different documents, so that a group of documents can be obtained. The metadata of different resource objects can also be saved in multiple documents. In an embodiment of the present application, in order to facilitate description and distinction, the above-mentioned document that saves the metadata of the resource object is defined as the first document; and the document that saves the access rights associated with the resource object is defined as the second document. Among them, the first document can be one or more. The second document can also be one or more. Multiple refers to 2 or more. The access rights associated with the resource object may include: role-based access rights control and access rights control based on the attributes of the resource object. The role-based access rights control and access rights control based on the attributes of the resource object can also be saved in different second documents.

[0095] Based on the first document storing the metadata of the resource object and the second document storing the access rights associated with the resource object, a reference relationship between the first document and the second document can be established, and the first document and the second document can be aggregated according to the reference relationship between the first document and the second document to obtain a context model of the resource object. Figure 1 The context model generation unit generates a context model.

[0096] That is, the context model of a resource object is obtained by establishing a reference relationship between documents for a group of documents corresponding to the resource object and aggregating the group of documents. Among them, the group of documents stores the metadata of the resource object and / or the access rights associated with the resource object. Therefore, the context model of a resource object can also be implemented as a document, which may include: at least one of the association relationships between different resource objects, the metadata of the resource object and the access rights associated with the resource object. Preferably, the document corresponding to the context model includes: the association relationships between different resource objects, the metadata of the resource object and the access rights associated with the resource object. Further, the resource object management strategy can be defined based on the context model of the resource object.

[0097] In the embodiments of the present application, the specific implementation method of defining the resource object management policy according to the context model of the resource object is not limited. In some embodiments, the resource object management policy can be manually defined according to the context model of the resource object and the user's management requirements for the resource object. Alternatively, the definition rules of the resource object management policy can also be set on the computing device, and the computing device can define the resource object management policy according to the context model of the resource object and the definition rules of the resource object management policy, etc.

[0098] In the above embodiment, the resource object management policy is defined based on the context model of the resource object, which can take into account more dimensional information of the resource object, has stronger policy reasoning capabilities, and can optimize the resource object management policy, thereby helping to optimize the resource object performance when the resource object management policy is applied to resource object management.

[0099] In the embodiment of the present application, in order to solve the technical problem that the traditional method of using AI models to determine the management strategy for resource objects is poor in flexibility and cannot adapt to the ever-changing resource object management strategy adjustment, a resource object management method that integrates a strategy engine and a language model is proposed. On the basis of the powerful processing capability of the language model, a strategy engine based on real-time changes and changes is introduced to make up for the defect that the language model cannot respond to the resource object management strategy adjustment in a timely manner, which helps to improve the flexibility and reliability of resource object management. For the description of the language model, please refer to the relevant content of the above-mentioned system embodiment, which will not be repeated here.

[0100] When a user needs to manage a resource object, he or she may provide management requirement description information for the target resource object. The management requirement description information is used to describe the user's management requirement for the target resource object and may be text information.

[0101] Accordingly, in step 401, the management requirements for the target resource object may be obtained. The management requirements for the target resource object may be the management requirements description information of the target resource object provided by the user. Accordingly, step 401 may be implemented as: obtaining the management requirements description information of the target resource object provided by the user as the management requirements for the target resource object.

[0102] In some embodiments, in order to improve the reliability of resource object management, the management results of the previous round of resource management of the target resource object according to the executable commands after fusion processing can be fed back to the input of the language model to supplement the resource demand description information provided by the user, which can improve the prompts of the language model and thereby improve the reliability of the text description information of the management method output by the language model.

[0103] Specifically, before inputting the management result description information of resource management of the target resource object according to the fused executable command in the current round and the resource requirement description information provided by the user as the management requirement for the target resource management object into the language model, it is also possible to judge whether the management result of managing the target resource object according to the fused executable command meets the management requirement corresponding to the current round based on the management requirement description information and the management result description information; if not, the management result description information of resource management of the target resource object according to the fused executable command in the current round and the resource requirement description information provided by the user are input into the language model as the management requirement for the target resource management object, and the above language model is cyclically executed to determine the text description information of the management method for the target resource object according to the management requirement; generate an executable command corresponding to the management method according to the text description information; fuse the executable command according to the pre-defined resource object management policy to obtain the fused executable command; and perform the step of managing the target resource object according to the fused executable command until the management result of managing the target resource object according to the fused executable command meets the management requirement.

[0104] Therefore, the management requirements input into the language model in the first round can be the management requirements description information for the target resource object provided by the user; the management requirements input into the language model in other rounds can be the management requirements description information for the target resource object provided by the user and the management result description information obtained in the previous round. For other rounds, step 401 can be implemented as: obtaining the management result description information obtained in the previous round and the management requirements description information for the target resource object provided by the user as the management requirements for the target resource object.

[0105] That is, assuming that the above-mentioned current round is the Nth round, N≥2, and is an integer, and the management result of managing the target resource object obtained in the (N-1)th round does not meet the management requirements corresponding to the (N-1)th round, then in the process of executing the resource object management method provided by the embodiment of the present application in the Nth round, obtaining the management requirements for the target resource object can be implemented as: obtaining the management requirement description information for the target resource object provided by the user; and obtaining the management result description information obtained in the (N-1)th round. Among them, the management result description information obtained in the (N-1)th round is: using the fused executable command (defined as the third executable command) obtained in the process of executing the resource object management method in the (N-1)th round to manage the target resource object. Description information. Furthermore, the management requirement description information provided by the user and the management result description information obtained in the (N-1)th round can be used as management requirements.

[0106] Optionally, the management requirements may be preprocessed. The management requirements are the prompts of the language model. Optionally, the preprocessing of the management requirements may include: removing unnecessary formats in the management requirements; performing sentence segmentation and word segmentation on the management requirements to obtain words corresponding to the management requirements; further, the words may be normalized to convert the words into a normalized form, and unnecessary words in the normalized form of the words may be removed.

[0107] In this embodiment, the management requirements can be input into the language model, and in step 402, the language model is used to determine the text description information of the management method for the target resource object based on the management requirements description information. During the language model training phase, the language model is trained using the historical management requirements in the container cluster management system (such as the K8s system) as the model input and the text description information of the historical management method corresponding to the historical management requirements as the model output. Therefore, in this embodiment, the management requirements description information can be input into the trained language model, and the output result is the text description information of the management method for the target resource object.

[0108] Since the language model outputs the text description information of the management method for the target resource object, the computing device needs to execute a computer program or command to manage the target resource object. Therefore, in step 403, an executable command (defined as a first executable command) corresponding to the management method for the target resource object can also be generated based on the text description information of the management method for the target resource object. An executable command refers to a computer instruction that can be executed by a computing device.

[0109] In the embodiments of the present application, the specific implementation form of generating the executable command corresponding to the management method for the target resource object according to the text description information of the management method for the target resource object is not limited. In some embodiments, a code example of the management method can be obtained from the text description information of the management method for the target resource object; the code example can be modified according to the management requirements for the target resource object to obtain an executable command, etc. For example, the identifier of the target resource object can be obtained from the management requirements for the target resource object; and the identifier of the resource object in the code example can be modified to the identifier of the target object, thereby obtaining an executable command, etc.

[0110] The executable command is a computer instruction converted from the text description information of the management method for the target resource object output by the language model. The text description information of the management method for the target resource object determined by the language model is learned by the language model during the training phase. The language model has not learned the resource object management strategy after the language model training is completed. Therefore, the management method for the target resource object determined by the language model has certain limitations and may be different from the resource object management strategy defined by the user for the target resource object.

[0111] To solve this problem, in step 404, the first executable command may be fused according to a predefined resource object management policy to obtain a fused executable command (defined as a second executable command).

[0112] Optionally, the management policy of the target resource object can be obtained from predefined resource object management policies. Further, it can be determined whether the first executable command is consistent with the management policy of the target object. If not, the first executable command is fused according to the management policy of the target resource object to obtain a fused second executable command.

[0113] In some embodiments, the first executable command may be fused according to a predefined resource object management strategy and a context model of a resource object in a container cluster management system to obtain a fused second executable command. For a description of the context model of a resource object, please refer to the relevant content of the above embodiment, which will not be repeated here.

[0114] Some optional implementation methods for fusing executable commands according to a predefined resource object management policy and a context model of a resource object in a container cluster management system are described below by way of example.

[0115] Implementation method 1: It is possible to determine whether it is necessary to jointly manage other resource objects associated with the target resource object during the management of the target resource object based on the association relationship between different resource objects and the predefined resource object management policy. Specifically, it is possible to determine other resource objects associated with the target resource object based on the association relationship between resource objects; and determine whether there is a policy for jointly managing the target resource object and other resource objects associated with it in the predefined resource object management policy; if the judgment result is yes, it is determined whether it is necessary to jointly manage other resource objects associated with the target resource object during the management of the target resource object. Further, a management policy for jointly managing the target resource object and other resource objects associated with it can be obtained from the predefined resource object management policy. Further, based on the management policy for jointly managing the target resource object and other resource objects associated with it, a command for jointly managing the target resource object and other resource objects associated with it can be supplemented in the first executable command to obtain a second executable command after fusion processing.

[0116] Implementation method 2: The access rights to the attributes of the target resource object can be obtained from the access rights associated with the resource object; based on the access rights to the attributes of the target resource object, the commands corresponding to the attributes without access rights are deleted from the first executable command to obtain a second executable command after fusion processing.

[0117] Implementation method 3: From the access rights associated with the resource object, obtain the access rights of the role of the user who makes the management request to the target resource object; and based on the access rights of the role of the user to the target resource object, select the command corresponding to the management item to which the role of the user has no access rights in the first executable command to obtain the second executable command after fusion processing.

[0118] In implementation mode 3, management items refer to management items of target resource objects, which can be distinguished by metadata and / or attributes of target resource objects. For example, different attributes of target resource objects are different management items, and different states of target resource objects are different management items.

[0119] Implementation method 4: Obtain the metadata of the target resource object from the metadata of the resource object; obtain the management policy corresponding to the metadata of the target resource object from the resource object management policy; and fuse the first executable command according to the management policy corresponding to the metadata of the target resource object to obtain a second executable command after fusion.

[0120] The above embodiments 1-4 only exemplarily give some optional implementation methods for fusing executable commands according to the predefined resource object management strategy and the context model of the resource object in the container cluster management system, and are not limited to the above four implementation methods, nor are they limited to the above four implementation methods. The above embodiments 1-4 can be implemented one by one, or multiple of them can be implemented in combination. Multiple refers to 2 or more. When multiple methods are implemented in combination, multiple implementation methods are used to fuse the first executable command together to obtain a second executable command after fusion processing.

[0121] Furthermore, in step 405, the target resource object may be managed according to the second executable command after the fusion processing. The specific implementation method of managing the target resource object is determined by the executable command after the fusion processing.

[0122] Specifically, the target management path for managing the target resource object can be determined according to the second executable command after the fusion processing. If the target management path involves an existing management module, the management module involved in the target management path can be called to manage the target resource object. For example, if the target management path involves a diagnostic analyzer, the diagnostic analyzer can be called to perform diagnostic analysis on the target resource object to obtain a diagnostic analysis result; and / or, if the target management path involves an audit analyzer, the audit analyzer can be called to perform audit processing on the log data of the target resource object to obtain an audit analysis result, etc.

[0123] If the target management path does not involve an existing management module, the executable command after fusion processing can be directly executed to achieve the management of the target resource object.

[0124] In this embodiment, the powerful processing capability of the language model is utilized to determine the management method corresponding to the management demand for the target resource object; and based on the powerful processing capability of the language model, the resource object management method determined by the language model is integrated with the predefined resource object management strategy. Since the predefined resource object management strategy is timely adjusted according to the user-defined policy definition rules, therefore, the resource object management method determined by the language model is integrated with the predefined resource object management strategy, which can make up for the defect that the language model cannot respond to the adjustment of the resource object management strategy in time, and help to improve the flexibility and reliability of resource object management.

[0125] In the embodiment of the present application, in order to further improve the reliability of resource object management, the management result of resource management of the target resource object according to the executable command after fusion processing can be fed back to the input of the language model to supplement the resource demand description information provided by the user, which can improve the prompt of the language model and thus improve the reliability of the text description information of the management method output by the language model. Figure 5 An example is given. Figure 5 As shown, the resource management method provided in the embodiment of the present application may include:

[0126] S1. Obtain management requirement description information for the target resource object provided by the user.

[0127] S2. Input the management requirement description information into the language model as the management requirement.

[0128] S3. Determine text description information of the management method for the target resource object according to management requirements using the language model.

[0129] S4. Generate a first executable command corresponding to the management method according to the text description information.

[0130] S5. According to a predefined resource object management policy, the first executable command is fused to obtain a fused second executable command.

[0131] S6. Manage the target resource object according to the second executable command after fusion processing to obtain a management result.

[0132] S7: According to the management requirement description information and the management result description information, determine whether the management result meets the management requirement. If the determination result is yes, stop the resource object management process; if the determination result is no, execute step S8.

[0133] S8. Input the management requirement description information and the management result description information into the language model as management requirements, and return to execute step S3.

[0134] The above-mentioned embodiment will feed back the management results of resource management of the target resource object according to the executable command after fusion processing to the input of the language model, which is used to supplement the resource demand description information provided by the user, and can improve the prompts of the language model, thereby improving the reliability of the text description information of the management method output by the language model.

[0135] It should be noted that the execution subject of each step of the method provided in the above embodiment can be the same device, or the method can be executed by different devices. For example, the execution subject of steps 401 and 402 can be device A; for another example, the execution subject of step 401 can be device A, and the execution subject of step 402 can be device B; and so on.

[0136] In addition, in some of the processes described in the above embodiments and the accompanying drawings, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel, and the sequence numbers of the operations, such as 401, 402, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0137] Accordingly, an embodiment of the present application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the steps in the above-mentioned resource object management method.

[0138] Figure 6 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present application. Figure 6 As shown, the computing device includes: a memory 60a and a processor 60b. The memory 60a is used to store computer programs.

[0139] The processor 60b is coupled to the memory 60a, and is used to execute the computer program to execute the steps in the resource object management method provided in the above embodiments. The detailed description of each method step can be found in the above embodiments, and will not be repeated here.

[0140] In some optional embodiments, such as Figure 6 As shown, the computing device may also include optional components such as a communication component 60c, a power component 60d, a display component 60e, and an audio component 60f. Figure 6 The components are shown schematically only and do not necessarily include Figure 6 The components shown do not necessarily mean that the computing device can only include Figure 6 Components shown.

[0141] in addition, Figure 6 The components in the dashed box are optional components, not mandatory components, and may depend on the product form of the electronic device. The computing device of this embodiment can be implemented as a terminal device such as a desktop computer, a laptop computer, a mobile phone, or an Internet of Things device; it can also be a traditional server, a cloud server, or a server cluster.

[0142] In an embodiment of the present application, the memory is used to store a computer program and can be configured to store various other data to support operations on the device where it is located. Among them, the processor can execute the computer program stored in the memory to implement the corresponding control logic. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random-Access Memory, SRAM), electrically erasable programmable read only memory (Electrically Erasable Programmable Read Only Memory, EEPROM), erasable programmable read only memory (Electrical Programmable Read Only Memory, EPROM), programmable read only memory (Programmable Read Only Memory, PROM), read only memory (Read Only Memory, ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0143] In the embodiment of the present application, the processor can be any hardware processing device that can execute the logic of the above method. Optionally, the processor can be a central processing unit (CPU), a graphics processing unit (GPU) or a microcontroller unit (MCU); it can also be a field programmable gate array (FPGA), a programmable array logic device (PAL), a general array logic device (GAL), a complex programmable logic device (CPLD) and other programmable devices; or an advanced reduced instruction set (RISC) processor (Advanced RISC Machines, ARM) or a system on chip (SoC), etc., but not limited to this.

[0144] In an embodiment of the present application, the communication component is configured to facilitate wired or wireless communication between the device in which it is located and other devices. The device in which the communication component is located can access a wireless network based on a communication standard, such as Wireless Fidelity (WiFi), 2G or 3G, 4G, 5G or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component can also be based on Near Field Communication (NFC) technology, Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology or other technologies.

[0145] In an embodiment of the present application, the display component may include a liquid crystal display (LCD) and a touch panel (TP). If the display component includes a touch panel, the display component may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0146] In an embodiment of the present application, a power supply component is configured to provide power to various components of the device in which it is located. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which the power supply component is located.

[0147] In an embodiment of the present application, the audio component may be configured to output and / or input audio signals. For example, the audio component includes a microphone (Microphone, MIC), and when the device where the audio component is located is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal may be further stored in a memory or sent via a communication component. In some embodiments, the audio component also includes a speaker for outputting an audio signal. For example, for a device with a language interaction function, voice interaction with a user can be achieved through an audio component.

[0148] 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, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0149] It should also be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do they limit "first" and "second" to different types.

[0150] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage, etc.) that contain computer-usable program code.

[0151] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0152] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0154] In a typical configuration, a computing device includes one or more processors (CPU, etc.), input / output interfaces, network interfaces, and memory.

[0155] Memory may include non-permanent storage in a computer-readable medium, random-access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0156] The storage medium of a computer is a readable storage medium, which may also be referred to as a readable medium. The readable storage medium includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be a computer-readable instruction, a data structure, a module of a program, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0157] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the above elements.

[0158] The above contents are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A resource object management method, characterized in that: include: Obtain management requirements for target resource objects in the container cluster management system; Determining text description information of a management method for the target resource object according to the management requirement by using a language model; Generating a first executable command corresponding to the management method according to the text description information; According to a predefined resource object management policy, the first executable command is fused to obtain a fused second executable command; The target resource object is managed according to the second executable command.

2. The method according to claim 1, characterized in that: The obtaining of management requirements for target resource objects in the container cluster management system includes: Acquire management requirement description information for the target resource object provided by a user as the management requirement; or, The resource object management method is executed for the Nth round; N≥2 and is an integer; the obtaining of management requirements for the target resource object in the container cluster management system includes: Acquire management requirement description information for the target resource object provided by the user; and acquire management result description information of managing the target resource object according to the third executable command after fusion processing; and use the management requirement description information and the management result description information as the management requirement; The third executable command is an executable command obtained after fusion processing during the (N-1)th round of executing the resource object management method.

3. The method according to claim 2, characterized in that Before using the management requirement description information and the management result description information as the management requirement, the method further includes: According to the management requirement description information and the management result description information, determining whether the management result of managing the target resource object according to the executable command after the fusion processing meets the management requirement corresponding to the (N-1)th round; If not, the management requirement description information and the management result description information are used as the management requirement.

4. The method according to claim 1, characterized in that: The step of generating a first executable command corresponding to the management method according to the text description information includes: Obtaining a code example of the management method from the text description information; According to the management requirement, the code example is modified to obtain the first executable command.

5. The method according to claim 4, characterized in that The step of modifying the code example according to the management requirement to obtain the first executable command includes: Obtaining an identifier of the target resource object from the management requirement; The identifier of the resource object in the code example is modified to the identifier of the target resource object to obtain the first executable command.

6. The method according to claim 1, characterized in that The step of fusing the first executable command according to a predefined resource object management policy to obtain a second executable command after fusing includes: The executable command is fused according to the predefined resource object management strategy and the context model of the resource object in the container cluster management system to obtain the second executable command; the resource object includes the target resource object; the context model is obtained by establishing a document reference relationship between a group of documents corresponding to the resource object and aggregating the group of documents; the group of documents stores the metadata of the resource object and / or the access rights associated with the resource object.

7. The method according to claim 6, characterized in that The context model of the resource object includes at least one of the following data: association relationships between different resource objects, metadata of resource objects, and access rights associated with resource objects; The fusing the first executable command according to the context model of the resource object in the container cluster management system and the predefined resource object management policy to obtain a second executable command after the fusing process includes: According to the association relationship between the different resource objects and the predefined resource object management policy, it is judged whether it is necessary to jointly manage other resource objects associated with the target resource object during the management of the target resource object; if the judgment result is yes, a management policy for jointly managing the target resource object and the other resource objects is obtained from the predefined resource object management policy; and according to the management policy for joint management, a command for jointly managing the target resource object and the other resource objects is supplemented in the first executable command to obtain the second executable command; and / or, Obtaining access rights for the attributes of the target resource object from the access rights associated with the resource object; deleting commands corresponding to the attributes without access rights from the first executable commands according to the access rights for the attributes of the target resource object, so as to obtain the second executable command; and / or, Obtaining the access rights of the role of the user who has proposed the management requirement to the target resource object from the access rights associated with the resource object; obtaining the second executable command from the command corresponding to the management item to which the role of the user has no access rights in the first executable command according to the access rights of the role of the user to the target resource object; and / or, Obtain the metadata of the target resource object from the metadata of the resource object; obtain the management policy corresponding to the metadata of the target resource object from the resource object management policy; and fuse the first executable command according to the management policy corresponding to the metadata of the target resource object to obtain the second executable command.

8. The method according to claim 6, characterized in that Before fusing the first executable command according to the predefined resource object management policy and the context model of the resource object in the container cluster management system, the method further includes: Acquire a first document storing metadata of the resource object and a second document storing access rights associated with the resource object; Establishing a reference relationship between the first document and the second document, and aggregating the first document and the second document according to the reference relationship to obtain the context model; The resource object management strategy is defined according to the context model.

9. The method according to any one of claims 1 to 8, characterized in that: The language model is a neural network model whose number of model parameters conforms to a preset parameter number range; the language model is obtained by training the model using historical management requirements in the container cluster management system as model input and using text description information of historical management methods corresponding to the historical management requirements as actual results corresponding to the model input.

10. The method according to claim 1, characterized in that The managing the target resource object according to the second executable command includes: determining, according to the second executable command, a target management path for managing the target resource object; The management module involved in the target management path is called to manage the target resource object.

11. The method according to claim 10, characterized in that The calling of the management module involved in the target management path to manage the target resource object includes: Calling the diagnostic analyzer to perform diagnostic analysis on the target resource object to obtain the diagnostic analysis result; and / or, The audit analyzer is called to perform audit processing on the log data of the target resource object to obtain the audit analysis results.

12. A computing device, characterized in that: include: A memory and a processor; wherein the memory is used to store a computer program; The processor is coupled to the memory and configured to execute the computer program to perform the steps of the method according to any one of claims 1 to 11.

13. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the steps in the method according to any one of claims 1 to 11.

Images