Network resource management method, device, equipment and medium

Abstract

The invention discloses a network resource management method and device, equipment and a medium, and relates to the technical field of cloud computing. According to the scheme, a language model special for network resource management is pre-established in a cloud computing platform, a user is supported to submit network resource management demand information to the language model only according to known input related natural language description for a cloud computing environment, and corresponding network resource calling information is generated by the language model; finally, a calling request is generated according to the network resource calling information, the network resource management service responds to the network resource management requirement of the user according to the calling request, cloud computing network management is achieved, the defect that a traditional interactive operation mode is single in form is overcome, higher universality, usability and reliability are achieved, and the user experience is improved. And the large-scale network resource management efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computing, and particularly relates to a network resource management method and device, equipment and medium. BACKGROUND

[0002] Infrastructure as a Service (IaaS) in cloud computing is becoming the focus of the industry, which provides users with basic resources such as computing, storage, and network, and realizes the convenient mode of on-demand access and pay-per-use.

[0003] However, in the network management of cloud computing, the traditional interactive operation mode is still mainly relied on at present, that is, the buttons on the page are used for adding, modifying, deleting, querying and other operations, and the corresponding data is input in the pop-up window to complete the resource management. This way has the problems of single form and inconvenient operation, especially when dealing with large-scale data. Although there are some iterative upgrades, the management mode has not realized fundamental change.

[0004] In view of the above, how to solve the problems of single form and inconvenient operation of the traditional interactive operation mode in the network management of cloud computing, especially the low efficiency when dealing with large-scale data, is a problem to be solved by the technical personnel in the field. SUMMARY

[0005] The present application provides a network resource management method, device, equipment and medium, which at least solves the problems of single form and inconvenient operation of the traditional interactive operation mode in the network management of cloud computing, especially the low efficiency when dealing with large-scale data.

[0006] The present application provides a network resource management method applied to a cloud computing platform; the method comprises:

[0007] When receiving the network resource management demand information of the user, the network resource management demand information is submitted to a pre-constructed language model to generate corresponding network resource calling information; wherein the language model is a model customized and trained based on the basic information of the cloud computing platform and the corresponding resource management interface information;

[0008] According to the network resource calling information, a calling request for the network resource management service is generated;

[0009] The calling request is sent to the network resource management service, so that the network resource management service responds to the network resource management demand of the user.

[0010] The present application also provides a network resource management device applied to a cloud computing platform; the device comprises:

[0011] The submission module is configured to submit the network resource management demand information to a pre-constructed language model to generate corresponding network resource calling information when receiving the network resource management demand information of the user, wherein the language model is a model customized and trained based on basic information of the cloud computing platform and corresponding resource management interface information.

[0012] The generation module is configured to generate a calling request for the network resource management service according to the network resource calling information.

[0013] The sending module is configured to send the calling request to the network resource management service so that the network resource management service responds to the network resource management demand of the user.

[0014] The present application also provides an electronic device, comprising a memory for storing a computer program and a processor for executing the computer program to implement the steps of any of the network resource management methods.

[0015] The present application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of any of the network resource management methods.

[0016] The present application has the advantages that a language model dedicated to network resource management is pre-built in the cloud computing platform, the user can input relevant natural language description according to the existing understanding of the cloud computing environment to submit network resource management demand information to the language model, and the language model generates corresponding network resource calling information; finally, a calling request is generated according to the network resource calling information, and the network resource management service responds to the network resource management demand of the user according to the calling request, cloud computing network management is realized, the form single of the traditional interactive operation mode is overcome, stronger universality, ease of use and reliability are achieved, and the large-scale network resource management efficiency is improved.

[0017] In addition, the present application also provides a network resource management device, equipment and medium, and the effects are the same as above. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 A flowchart of a network resource management method provided by the embodiments of the present application;

[0020] Figure 2A network resource management architecture schematic diagram provided for an embodiment of the present application;

[0021] Figure 3 A network resource management device schematic diagram provided for an embodiment of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0023] It should be noted that, in the description of the present application, the terms “comprise”, “contain” or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. The terms “first”, “second” and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0024] In order for those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0025] At present, in the network management of cloud computing, it still mainly depends on the traditional interactive operation mode, that is, adding, modifying, deleting, querying and other operations are performed through buttons on the page, and corresponding data is input in the pop-up window to complete resource management. This way has the problems of single form and inconvenient operation, especially when dealing with large-scale data. Although there is some iteration and upgrading, the management mode has not been fundamentally changed. In order to solve the above problems, the present application pre-builds a cloud computing network resource management module in the cloud computing platform, which is specially used for cloud computing network resource management. The method provided by the present application is mainly realized based on the module.

[0026] It should be noted that the cloud computing network resource management module can interact with the user interface (User Interface, UI) of the cloud computing system, the neutron component and the language model platform at the same time. The module receives the training instructions of the system administrator on one hand, and receives the application demand instructions of the user in the cloud computing system on the other hand, and can directly output various instructions to the language model and analyze the feedback of the language model, in addition, it can also directly initiate a call request to the neutron to manage the network resources in the cloud computing. The network resource management method based on the module is described in detail as follows:

[0027] Figure 1 A flowchart of a network resource management method provided by an embodiment of the present application is shown in Figure 1. Figure 1 As shown in the figure, the method comprises:

[0028] S10: When receiving the network resource management demand information of the user, the network resource management demand information is submitted to the pre-constructed language model to generate corresponding network resource calling information.

[0029] The language model is a model customized and trained based on the basic information of the cloud computing platform and the corresponding resource management interface information.

[0030] The present scheme pre-integrates a set of back-end system in the cloud computing platform, establishes a conversation with the language model, and maintains the persistence of the conversation; at the same time, the method is exposed to the outside, which can initiate a request to the language model, process and return the response given by the language model. At the same time, a functional system containing front-end and back-end is integrated and developed in the cloud computing platform, which is only open to system administrators and invisible to ordinary users.

[0031] It can be understood that the ordinary training of the language model refers to the process of training the model to understand and generate natural language by using large-scale text data sets through self-supervised learning or supervised learning, in order to improve its language understanding and generation ability. Unlike ordinary training, the main purpose of the above functional system is to provide functions for the system administrator to customize the training of the language model. The customized training content is to make the language model understand the cloud computing environment, be familiar with the management model of different resources in the cloud computing environment, accurately provide the information required by the user in the cloud computing environment, provide accurate access requests for the resource management demand of ordinary users, and continuously optimize the response of the system administrator with the passage of time.

[0032] Figure 2 A network resource management architecture diagram provided by an embodiment of the present application is shown in Figure 2. Figure 2As shown, based on the above system, when a user has a network resource management requirement, the user only needs to submit the natural language description of the network resource management requirement information to the language model through the UI interface function of the front end, and the language model can generate the corresponding network resource calling information.

[0033] It should be noted that the language model is trained by the administrator through the UI interface function by entering the natural language description, the front-end page operation is completed, the calling request is initiated to the back-end interface, the established language model session and the provided method execution are called. In this embodiment, the training data required by the language model and the specific training process are not limited, and are determined according to the specific implementation.

[0034] S11: generating a calling request for a network resource management service according to the network resource calling information.

[0035] Further, the back end receives the network resource calling information generated by the language model, and generates a calling request for a network resource management service according to the network resource calling information. In this embodiment, the specific content of the network resource calling information is not limited, for example, it can include resource identifiers, request parameters, request methods, authentication information, and timestamps. Correspondingly, the calling request is related to the above content of the network resource calling information.

[0036] S12: sending the calling request to the network resource management service, so that the network resource management service responds to the user's network resource management requirement.

[0037] Finally, after generating the calling request for the network resource management service, the back end initiates a calling request to the network resource management service of the cloud computing platform, so that the network resource management service responds to the user's network resource management requirement, and realizes the creation, modification, deletion, query and other operations of the network resource of the cloud computing platform.

[0038] In this embodiment, a language model dedicated to network resource management is pre-built in the cloud computing platform, which supports users to input related natural language descriptions based on the existing understanding of the cloud computing environment, submit network resource management requirement information to the language model, and generate corresponding network resource calling information by the language model. Finally, according to the network resource calling information, a calling request is generated, and the network resource management service responds to the user's network resource management requirement according to the calling request, realizes cloud computing network management, overcomes the form single of the traditional interactive operation mode, has stronger universality, ease of use and reliability, and improves the efficiency of large-scale network resource management.

[0039] On the basis of the above embodiment, in some embodiments, the customized training process of the language model includes:

[0040] S101: Obtain training data; wherein the training data at least contains cloud computing platform basic information, resource management interface information, resource management interface input parameter definition and corresponding return value information, interface call response processing information, interface call exception processing information and response format information.

[0041] S102: Determine the writing format and training input method of the training data.

[0042] S103: Customized training of the language model according to the training data, the writing format and the training input method.

[0043] In order to realize the customized training of the language model, the training data is specifically obtained in the embodiment. It can be understood that the training data is necessary information in the current cloud computing environment, at least containing cloud computing platform basic information, resource management interface information, resource management interface input parameter definition and corresponding return value information, interface call response processing information, interface call exception processing information and response format information. The various types of training data are described below:

[0044] First, in some embodiments, in order to obtain cloud computing platform basic information, cloud platform environment address information, cloud platform login information and cloud platform default system configuration information need to be specifically obtained. Among them, the cloud platform environment address information is an http address or domain name, etc. The cloud platform login information is mainly the account name and password of the user of the cloud computing platform for resource management, and the login information enables the interface call request initiated to the cloud computing platform to pass authentication and authorization. The cloud platform default system configuration information includes all the available partitions and virtual data centers maintained in the system; specifically, each type of information contains their respective names, unique identifiers, etc.; in addition, their default information needs to be specified, because the available partitions, virtual data centers, etc. often contain multiple in the system, if the user does not specify which available partition and which virtual data center to create when creating network resources, the language model can also provide the default available partition and the default virtual data center.

[0045] Secondly, in order to obtain the resource management interface information, in some embodiments, it is particularly necessary to determine the interface provided by the cloud computing platform, including network, router, security group, address group, floating Internet Protocol (Internet Protocol, IP) address, Quality of Service (Quality of Service, QoS), Network Address Translation Gateway, peer connection, firewall, cloud private line, Virtual Private Network (Virtual Private Network, VPN), load balancing, Cloud Domain Name System (Cloud DNS), traffic mirroring, etc.; and obtain the information of each interface, including obtaining the creation network interface information, modifying the network interface information, deleting the network interface information, querying the network interface information, creating the router interface information, modifying the router interface information, deleting the router interface information, querying the router interface information, creating the security group interface information, modifying the security group interface information, deleting the security group interface information, and querying the security group interface information.

[0046] Further, the resource management interface input parameter definition and the corresponding return value information are obtained. Specifically, the information is used to inform the language model how to encapsulate the input parameter information of each type of resource management interface, and to explicitly indicate the attributes and related restrictive explanations and default values of each field of the input parameter, as well as the fields contained in the return value after the interface is called and the explanation of each field. It can be understood that the input parameters and return values of different interfaces of resources are almost different. For example, for the creation of a security group interface, the name, virtual data center identifier, and description need to be transmitted, and the corresponding return value includes the security group unique identifier, name, virtual data center, description, etc. For the interface of creating a subnet, the input parameters include the name, virtual data center, network type, network segment, address pool, maximum transmission unit, and description, and the corresponding return value includes the unique identifier, name, virtual data center, network type, network segment, address pool, maximum transmission unit, and description.

[0047] At the same time, the response processing information after the interface is called is obtained. The information specifically informs the language model how to identify the success or failure flag based on the response information after the interface of each resource is called. For the success case, a relatively unified prompt information is prompted, and for the failure case, an exception handling process is entered. It should be noted that the specific way of identifying the response success or response failure based on the response information in the present embodiment is not limited, for example, the code in the response can be used for identification, and the code is set to 200 when the call is normal without error; when the server cannot understand the syntax or structure of the request, the code is set to 400; when the server reports an error due to other reasons, the code is set to 500.

[0048] Subsequently, interface call exception handling information is acquired. The interface call exception handling information is used to inform the language model how to handle the captured exception in the case of an exception in the call of each resource interface. Specifically, it includes analysis of the interpretation and explanation of each exception return value, making prompts and warnings for different exception return values, assisting users in modifying and correcting natural language descriptions to refine and maintain resources, and the like. For example, assuming that a call request for creating a security group fails because the user-specified name contains illegal characters, the exception handling process is as follows: after the interface for creating a security group fails, the server returns the failure reason, which includes "contains illegal characters" in the name. The language model first needs to analyze the failure reason in the return value. After the analysis is complete, the language model will prompt the user of the root cause, such as making the user fully aware that the name specified when creating the security group contains illegal characters, and the user needs to re-verify the illegal characters; at the same time, after the prompt, the language model informs the user that the user should avoid the name containing illegal characters and re-specify a legal name in the self-defined natural language instruction, so that the user further modifies the natural language instruction, and then uses the revised natural language to maintain the resource using the language model.

[0049] Finally, in order to make the language model clear about the response format it should provide based on the user's question, response format information also needs to be acquired. The response format information is used to make the language model understand that after various configuration item information of the cloud environment is provided, only simple completion information needs to be provided. Specifically, for user behavior, the language model needs to encapsulate a complete Application Programming Interface (API) that can be directly called according to the actual request content and return it; for the analysis of the API call of the cloud platform, according to the success or failure result, a textual explanation is provided.

[0050] The above is the training data that needs to be acquired in the language model customization training process. On this basis, the writing format and training input method of the training data also need to be determined. Specifically, the writing format is the arrangement and structure specification of the training data. In this embodiment, the specific type of the writing format is not limited. The training input method is the specific input method of the training language when input to the language model, for example, the training data can be directly informed to the language model through natural language, or assembled into a word document, a text document, and the like. The content is written in a specific writing format in advance, and then uploaded to the language model, so that the language model understands the training data. Finally, the language model is customized and trained according to the training data, the writing format, and the training input method.

[0051] Therefore, it is ensured that the language model can accurately identify the user's network resource management demand and accurately generate corresponding network resource management information.

[0052] On the basis of the above-mentioned embodiments, in some embodiments, after sending the calling request to the network resource management service, the following is further included:

[0053] S111: Obtain response information of the network resource management service to the calling request.

[0054] S112: Input the response information into the language model to determine the processing result of the calling request.

[0055] After sending the calling request to the network resource management service and processing the calling request by the network resource management service, the cloud computing platform will respond to the calling request. In order to determine whether the current calling is successful, the response information of the network resource management service to the calling request also needs to be obtained. In this embodiment, the specific format of the response information is not limited and is determined according to the specific implementation. Finally, the response information is input into the language model, and the response information is analyzed by the language model, so as to determine the processing result of the current calling request. In this way, the user can more clearly understand whether the current network configuration management demand is met. The specific process of the language model analyzing the response information is described in detail as follows:

[0056] In some embodiments, inputting the response information into the language model to determine the processing result of the calling request includes:

[0057] S121: Based on the interface calling response processing information input during customization training, identify the preset code in the response information; wherein the preset code is a code set in advance to represent the success or failure of the interface calling.

[0058] S122: Determine whether the current network interface calling based on the calling request is successful according to the preset code; if yes, go to step S123; if no, go to step S124.

[0059] S123: Output prompt information representing the success of the operation.

[0060] S124: According to the interface calling exception processing information input during customization training, analyze the failure reason in the response information.

[0061] S125: Output corresponding modification suggestions according to the failure reason, so as to re-determine the network resource management demand information according to the modification suggestions.

[0062] Since the language model uses the interface call response processing information to inform the language model of the interface of each resource after the interface is called and connected, how to identify the success or failure flag based on the response information during the previous customization training process, and uses the interface call exception processing information to inform the language model of how to handle the captured exception and return a prompt and warning for different exception values, assist the user in modifying and correcting the natural language description, and maintain the resource, and the like. Therefore, after the response information is input into the language model, the language model can directly identify the preset code in the response information based on the interface call response processing information input during the customization training. It should be noted that the preset code is a code that is set in advance to represent the success or failure of the interface call, and the specific format of the preset code in this embodiment is not limited, but is determined according to the specific implementation.

[0063] Further, it is determined whether the network interface call based on the call request is successful according to the preset code. If it is confirmed that the network interface call is successful, a prompt information representing the success of the operation is output. If it is confirmed that the network interface call fails, the failure reason in the response information is analyzed according to the interface call exception processing information input during the customization training, and a corresponding modification suggestion is output according to the failure reason, so that the user can determine the network resource management requirement information again according to the modification suggestion and perform a new round of network resource management.

[0064] In this embodiment, the response information is input into the language model, the language model performs abnormality identification and abnormality processing on the network resource management response information according to the strategy learned during the customization training process, and the reliability of the network resource management is improved.

[0065] On the basis of the above-mentioned embodiments, in some embodiments, the network resource call information is generated according to the network resource call information to generate a call request for the network resource management service, including:

[0066] S131: Data shaping is performed on the network resource call information based on the back end.

[0067] S132: A call request for the network resource management service is generated according to the network resource call information after data shaping.

[0068] In order to enable the generated call request to be accurately identified and executed by the network resource management service, when generating the call request, the network resource call information is specifically data shaped based on the backend. It should be noted that the specific process of data shaping in the present embodiment is not limited, for example, it can include data cleaning, format conversion, field screening and reorganization, calculation and aggregation, and verification and checking steps, the purpose is to process the original call information into a format that meets the specific needs or interface requirements. Finally, an accurate and effective call request for the network resource management service is generated according to the data shaped network resource call information.

[0069] In the present embodiment, the call request for the network resource management service is generated according to the data shaped network resource call information, which ensures the standardization and consistency of data transmission.

[0070] On the basis of the above-mentioned embodiments, in some embodiments, after sending the call request to the network resource management service, the method further comprises:

[0071] S141: generating a call log for the network resource management service this time according to the network resource management requirement information and the corresponding call request.

[0072] S142: storing the call log into a storage module of the cloud computing platform.

[0073] In order to better analyze the current call management process, after sending the call request to the network resource management service, a call log for the network resource management service this time can also be generated according to the network resource management requirement information and the corresponding call request, and the call log can be stored into a storage module of the cloud computing platform. In this way, the history record of the call request can be traced through the call log, helping to analyze the time when the problem occurs, and at the same time, the system performance bottleneck can be analyzed to carry out targeted optimization.

[0074] In addition, on the basis of generating the call log, the cloud computing platform can also execute a variety of subsequent tasks based on the response result of the call request to optimize resource allocation and improve system performance. For example, when it is detected that the load of a certain network resource is too high, the resource can be automatically expanded to increase new nodes or interfaces to share the load, ensuring the stability and response speed of the service. Similarly, if a certain interface fails, the system can automatically adjust the load balancing strategy to switch traffic to avoid service interruption. In this way, the cloud computing platform can more intelligently and efficiently manage network resources, improving overall operation efficiency and user experience.

[0075] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment.

[0076] Figure 3 A schematic diagram of a network resource management device is provided for an embodiment of the present application. The device is applied to a cloud computing platform; as shown in the figure, the device comprises: Figure 3

[0077] A submission module 10 is configured to, when receiving network resource management demand information of a user, submit the network resource management demand information to a pre-constructed language model to generate corresponding network resource invocation information; wherein the language model is a model customized and trained based on basic information of the cloud computing platform and corresponding resource management interface information.

[0078] A generation module 11 is configured to generate an invocation request for a network resource management service according to the network resource invocation information.

[0079] A sending module 12 is configured to send the invocation request to the network resource management service so that the network resource management service responds to the network resource management demand of the user.

[0080] In some embodiments, the customized training process of the language model comprises: obtaining training data; wherein the training data at least contains cloud computing platform basic information, resource management interface information, resource management interface input parameter definition and corresponding return value information, interface invocation response processing information, interface invocation exception processing information and response format information; determining the writing format and training input mode of the training data; and customizing and training the language model according to the training data, the writing format and the training input mode.

[0081] In some embodiments, the training data is obtained, specifically comprising obtaining cloud platform environment address information, cloud platform login information and cloud platform default system configuration information; obtaining network interface creation information, network interface modification information, network interface deletion information, network interface query information, router interface creation information, router interface modification information, router interface deletion information, router interface query information, security group interface creation information, security group interface modification information, security group interface deletion information and security group interface query information.

[0082] In some embodiments, the device further comprises:

[0083] A response message obtaining module is configured to obtain response information of the network resource management service to the invocation request;

[0084] A processing result determining module is configured to input the response information into the language model to determine the processing result of the invocation request.

[0085] In some embodiments, the processing result determining module comprises:

[0086] ​The identification module is configured to identify preset codes in the response information based on the interface call response processing information input during customization training.

[0087] The judgment module is configured to determine whether the network interface call based on the call request succeeds or fails according to the preset codes, and trigger the output module if the network interface call succeeds, or trigger the exception processing module if the network interface call fails.

[0088] The output module is configured to output prompt information indicating that the operation succeeds.

[0089] The exception processing module is configured to analyze the failure reason in the response information based on the interface call exception processing information input during customization training.

[0090] The modification suggestion module is configured to output corresponding modification suggestions according to the failure reason, so as to re-determine the network resource management requirement information according to the modification suggestions.

[0091] In some embodiments, the generation module 11 comprises:

[0092] The data shaping module is configured to perform data shaping on the network resource call information based on the backend.

[0093] The generation sub-module is configured to generate a call request for the network resource management service according to the data-shaped network resource call information.

[0094] In some embodiments, the network resource management device further comprises:

[0095] The log generation module is configured to generate a call log for the network resource management service according to the network resource management requirement information and the corresponding call request.

[0096] The log storage module is configured to store the call log in a storage module of a cloud computing platform.

[0097] The features of the embodiments of the network resource management device can be understood with reference to the related descriptions of the embodiments of the network resource management method, which will not be repeated here.

[0098] The embodiments of the present application also provide an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above network resource management method embodiments.

[0099] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to perform the steps in any of the above network resource management method embodiments when running.

[0100] In an example embodiment, the computer readable storage medium described above can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0101] Embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program, when executed by a processor, implements the steps in any of the network resource management method embodiments described above.

[0102] Embodiments of the present application also provide another computer program product, which comprises a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps in any of the network resource management method embodiments described above.

[0103] The skilled in the art can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in the above description in a general manner. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0104] The above describes in detail a network resource management method, device, equipment and medium provided by the present application. The principle and implementation mode of the present application are described by applying specific examples in this paper, and the above example description is only used to help understand the method of the present application and its core idea. It should be pointed out that, for those skilled in the art, without departing from the principle of the present application, some improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A network resource management method, characterized by, The method is applied to a cloud computing platform, and comprises the following steps: When receiving network resource management demand information of a user, submitting the network resource management demand information to a pre-constructed language model to generate corresponding network resource calling information; wherein the language model is a model customized and trained based on basic information of the cloud computing platform and corresponding resource management interface information; Generating a calling request for a network resource management service according to the network resource calling information; Sending the calling request to the network resource management service so that the network resource management service responds to the network resource management demand of the user.

2. The network resource management method according to claim 1, characterized by, The customized training process of the language model comprises the following steps: Obtaining training data; wherein the training data at least contains cloud computing platform basic information, resource management interface information, resource management interface input parameter definition and corresponding return value information, interface calling response processing information, interface calling exception processing information and response format information; Determining the writing format and training input mode of the training data; Customizing and training the language model according to the training data, the writing format and the training input mode.

3. The network resource management method of claim 2, wherein, Obtaining training data comprises the following steps: Obtaining cloud platform environment address information, cloud platform login information and cloud platform default system configuration information; Obtaining creation network interface information, modification network interface information, deletion network interface information, query network interface information, creation router interface information, modification router interface information, deletion router interface information, query router interface information, creation security group interface information, modification security group interface information, deletion security group interface information and query security group interface information.

4. The network resource management method of claim 1, wherein, After sending the calling request to the network resource management service, the method further comprises the following steps: Obtaining response information of the network resource management service to the calling request; Inputting the response information into the language model to determine the processing result of the calling request.

5. The network resource management method according to claim 4, characterized by, Inputting the response information into the language model to determine the processing result of the calling request comprises the following steps: Based on the interface calling response processing information input during the customized training, identifying a preset code in the response information; wherein the preset code is a code set in advance to represent the success or failure of interface calling; Determining whether the network interface calling based on the calling request is successful according to the preset code; If yes, outputting prompt information representing the success of the operation; If no, analyzing the failure cause in the response information according to the interface calling exception processing information input during the customized training; Outputting corresponding modification suggestions according to the failure cause so as to re-determine the network resource management demand information according to the modification suggestions.

6. The network resource management method according to any one of claims 1 to 5, characterized by, Generating a calling request for a network resource management service according to the network resource calling information comprises the following steps: Based on data shaping of the network resource calling information by a back end; Generating the calling request for the network resource management service according to the network resource calling information after data shaping.

7. The network resource management method of claim 1, wherein, After sending the calling request to the network resource management service, the method further comprises the following steps: generating a call log of the network resource management service this time according to the network resource management requirement information and the corresponding call request; storing the call log into a storage module of the cloud computing platform.

8. A network resource management apparatus characterized by comprising: The application is applied to a cloud computing platform, and the device comprises: a submission module configured to submit the network resource management requirement information to a pre-constructed language model to generate corresponding network resource call information when receiving the network resource management requirement information of a user, wherein the language model is a model customized and trained based on basic information of the cloud computing platform and corresponding resource management interface information; a generation module configured to generate a call request for a network resource management service according to the network resource call information; a sending module configured to send the call request to the network resource management service so that the network resource management service responds to the network resource management requirement of the user.

9. An electronic device, comprising: comprise: a memory configured to store a computer program; a processor configured to implement the steps of the network resource management method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein the computer program is executed by the processor to implement the steps of the network resource management method according to any one of claims 1 to 7.

Images