Network management method and device, computer device and readable storage medium

Abstract

The application relates to a network management method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: in response to received service description information, extracting service requirement information in the service description information according to a pre-constructed service intention model; processing the service requirement information and network resource information through a first strategy function to generate a service service strategy; processing the service service strategy and network node state information based on a second strategy function to obtain an execution strategy of each network node; and the execution strategy of each network node is used for realizing network service management. The method realizes flexibility and expansibility of the network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of multi-modal network, and particularly relates to a network management method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] With the rapid development of the Internet, the demand for network service capabilities is also constantly upgrading. Especially, the current network needs to cope with various business scenarios, the integration of various communication modes and computing requirements, which poses a high challenge to the network system.

[0003] In the current traditional method, the network system can only carry one or several types of services, and its management strategy mainly depends on static network configuration requirements and rules to set network service capabilities.

[0004] However, the pre-set network configuration requirements and rules in the traditional method cannot adapt to changes in network conditions and business requirements, limiting the flexibility and scalability of the network. SUMMARY

[0005] Therefore, it is necessary to provide a network management method, device, computer equipment, computer readable storage medium and computer program product to solve the above technical problems.

[0006] In a first aspect, the present application provides a network management method, comprising:

[0007] In response to the received service description information, extracting the service requirement information in the service description information according to the pre-constructed service intention model;

[0008] processing the service requirement information and the network resource information through a first strategy function to generate a service service strategy;

[0009] processing the service service strategy and the network node state information based on a second strategy function to obtain the execution strategy of each network node; the execution strategy of each network node is used to realize network service management.

[0010] In one embodiment, before the step of responding to the received service description information and extracting the service requirement information in the service description information according to the pre-constructed service intention model, the method further comprises:

[0011] receiving initial service description information;

[0012] performing abnormality check and format preprocessing on the initial service description information to obtain service description information.

[0013] In one of the embodiments, before extracting the business requirement information from the business description information according to the pre-constructed business intention model, the method further comprises:

[0014] obtaining the business description information in natural language form;

[0015] processing the business description information based on a preset mapping function and a preset modeling framework to construct a business intention model; the business intention model comprises a first intention model and a second intention model.

[0016] In one of the embodiments, processing the business requirement information and the network resource information through the first policy function to generate a business service policy comprises:

[0017] inputting the business requirement information and the network resource information into a pre-trained first policy function; the first policy function is updated based on a policy gradient update algorithm and a reward function corresponding to execution policy feedback information;

[0018] processing the business requirement information and the network resource information through the first policy function to generate a business service policy.

[0019] In one of the embodiments, processing the business service policy and the network node state information based on the second policy function to obtain an execution policy of each network node comprises:

[0020] processing the business service policy and the network node state information through a second policy function to obtain business route planning information; the business route planning information contains transmission paths between network nodes corresponding to the business service;

[0021] inputting the transmission paths and the network node state information into the second policy function to generate an execution policy of each network node.

[0022] In one of the embodiments, the method further comprises:

[0023] indicating that the execution policy is input into a front-end compiler of each network node, processing through the front-end compiler to generate a hardware language representation;

[0024] transmitting the hardware language representation to a back-end compiler, performing code compilation and assembly processing through the back-end compiler to execute the execution policy.

[0025] In a second aspect, the application further provides a network management device, comprising:

[0026] extracting service requirement information in the service description information according to a pre-constructed service intention model in response to the received service description information;

[0027] generating a service service policy by processing the service requirement information and network resource information through a first policy function;

[0028] processing the service service policy and network node state information based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management.

[0029] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0030] extracting service requirement information in the service description information according to a pre-constructed service intention model in response to the received service description information;

[0031] generating a service service policy by processing the service requirement information and network resource information through a first policy function;

[0032] processing the service service policy and network node state information based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management.

[0033] In a fourth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0034] extracting service requirement information in the service description information according to a pre-constructed service intention model in response to the received service description information;

[0035] generating a service service policy by processing the service requirement information and network resource information through a first policy function;

[0036] processing the service service policy and network node state information based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management.

[0037] In a fifth aspect, the present application further provides a computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the following steps:

[0038] In response to the received service description information, service requirement information in the service description information is extracted according to a pre-constructed service intention model;

[0039] The service requirement information and network resource information are processed by a first policy function to generate a service service policy;

[0040] The service service policy and network node state information are processed based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management.

[0041] The network management method, device, computer equipment, computer readable storage medium and computer program product, in response to the received service description information, extract the service requirement information in the service description information according to the pre-constructed service intention model; the service requirement information and network resource information are processed by a first policy function to generate a service service policy; the service service policy and network node state information are processed based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management. By using the method, in response to the service description information input by the user, the service service policy is generated according to the service intention model and the first decision function, and the execution policy of each network node is determined based on the service service policy and the current network node state information, the execution policy of each network node is dynamically adjusted, the changes of network conditions and service requirements in the current network environment are adapted, and the flexibility and expansibility of the network are realized. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0043] Figure 1 An application environment diagram of a network management method in an embodiment;

[0044] Figure 2 A flowchart of a network management method in an embodiment;

[0045] Figure 3 A flowchart of a step of obtaining service description information in an embodiment;

[0046] Figure 4 A flowchart of a step of constructing a service intention model in an embodiment;

[0047] Figure 5 This is a schematic diagram of the business intent model in one embodiment;

[0048] Figure 6 This is a flowchart illustrating the steps for generating a business service strategy in one embodiment.

[0049] Figure 7 This is a flowchart illustrating the process of generating execution strategies for each network node in one embodiment.

[0050] Figure 8 This is a flowchart illustrating the execution of the execution strategy by each network node in one embodiment;

[0051] Figure 9 This is a structural block diagram of a network management device in one embodiment;

[0052] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0054] The network management method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, namely the multimodal network system 100, there are five layers: application layer 110, service layer 120, control layer 130, resource layer 140, and access layer 150. The application layer 110 directly provides diverse network services to end users. Through interaction between various application processes, it receives and processes network application service information, ensuring seamless access to the required network resources and applications for users. The service layer 120 effectively connects user business needs with network service capabilities. Its main functions include business intent modeling, business intent parsing, resource awareness, and service policy generation, achieving highly personalized and optimized business network services. The control layer 130 transforms service policies into executable network policies based on network state information, enabling intelligent service demand management and dynamic adjustment of network state. This layer ensures the continuity and security of network services to adapt to constantly changing network needs and environments. The resource layer 140 includes the network topology, protocols used, and related hardware and software. This layer provides flexible and configurable network components and services, thereby enhancing the network's adaptability, stability, and scalability. Access Layer 150: Responsible for connecting various user devices (such as computers, drones, vehicles, etc.) to the network, providing users with the necessary interfaces to access the network.

[0055] In one exemplary embodiment, as shown in Figure 2 A network management method is provided, which is applied to a multi-modal network system as an example for illustration, including the following steps 202 to 206. Among them: Figure 1

[0056] Step 202, in response to the received service description information, extracting the service requirement information in the service description information according to the pre-constructed service intent model.

[0057] In implementation, the application layer in the multi-modal network system receives the service description information from the user, which can be text information in natural language form, parameter setting information, information in voice form, etc. The form of the service description information input by the user is not limited in the embodiments of the present disclosure. After the application layer receives the service description information, the service description information is processed, and the processed service description information is transmitted to the service layer of the multi-modal network system. The service layer inputs the service description information into the pre-constructed service intent model, extracts the service requirement information in the service description information through the mapping relationship contained in the service intent model, which can include basic business parameters and business service expectation information, etc. The specific content contained in the service requirement information is not limited in the embodiments of the present disclosure.

[0058] Step 204, processing the service requirement information and the network resource information through the first strategy function to generate the business service strategy.

[0059] In implementation, the first strategy function is defined in advance in the multi-modal network system, which can be a reinforcement learning strategy function. The reinforcement learning strategy function can be composed of multi-layer perception, recurrent neural network, attention mechanism network, etc. The algorithm type contained in the reinforcement learning strategy function is not limited in the embodiments of the present disclosure. In the dynamic response process of the multi-modal network system to the service description information of the user, the service layer can train the pre-defined first strategy function based on the reward function and the network feedback information, etc. to obtain the trained first strategy function. In this way, when the service layer receives the service requirement information transmitted by the application layer, the service layer can process the service requirement information and the network resource information extracted by the service layer through the first strategy function to generate the business service strategy.

[0060] Optionally, the business service strategy contains strategies of different dimensions, for example, it can include network service modal strategy, network resource allocation strategy, security guarantee strategy, reliability guarantee strategy, etc. The type and dimension of the business service strategy are not limited in the embodiments of the present disclosure.

[0061] ​At step 206, the service service policy and the network node state information are processed based on a second policy function to obtain an execution policy of each network node.

[0062] The execution policy of each network node is used to implement network service management.

[0063] In implementation, a second policy function is also predefined in the multi-modal network system. The second policy function can also be a reinforcement learning policy function. The second policy function is a hierarchical deep reinforcement learning algorithm, which can specifically be composed of a graph neural network and an attention mechanism. The algorithm type of the second policy function is not limited in the embodiments of the present disclosure. In the multi-modal network system, the service layer can still train the second policy function in advance to obtain a trained second policy function. In this way, the multi-modal network system inputs the service service policy and the network node state information into the second policy function, obtains routing planning information based on the trained second policy function, and further processes the routing planning information and the network node state information to obtain an execution policy of each network node. The execution policy of each network node is used to implement network management of each network node in the multi-modal network system, that is, each network node provides a service by running the execution policy, and the service providing process meets the current network condition and service demand.

[0064] The training process of the second policy function is similar to that of the first policy function. The training process of the first policy function will be described in detail in the following embodiments, and will not be repeated here.

[0065] In the above network management method, in response to the received service description information, the service demand information in the service description information is extracted according to the pre-constructed service intention model; the service demand information and the network resource information are processed by the first policy function to generate a service service policy; the service service policy and the network node state information are processed based on a second policy function to obtain an execution policy of each network node; and the execution policy of each network node is used to implement network service management. By using the method, in response to the service description information input by the user, the service service policy is generated according to the service intention model and the first decision function, and the execution policy of each network node is determined based on the service service policy and the current network node state information, the execution policy of each network node is dynamically adjusted, the changes of network conditions and service demands in the current network environment are adapted, and the flexibility and expansibility of the network are realized.

[0066] In one exemplary embodiment, as shown in FIG. 2, after step 202, the method further includes: Figure 3

[0067] At step 302, initial service description information is received. ​

[0068] In implementation, the system architecture of the multi-modal network system mainly includes five layers, and each layer includes a plurality of functional modules, each of which is used to provide different functions. In this way, when the multi-modal network system provides various types of service services, the user can input initial service description information to the multi-modal network system to dynamically determine the execution strategy of each network node in the current multi-modal network system. The application layer of the multi-modal network system includes a service information access module and a service information preprocessing module. In this way, when the user inputs the initial service description information, the service information access module receives the initial service description information.

[0069] The initial service description information includes service basic parameter information and user's original intention information. The service basic parameter information may include, but is not limited to, source node, target node, service mode, service level, and the like. The original intention information is a service service expectation, which includes forwarding performance expectation, reliability expectation, security expectation, cost benefit expectation, and the like.

[0070] Step 304, performing exception check and format preprocessing on the initial service description information to obtain service description information.

[0071] In implementation, after the service information access module receives the initial service description information, the initial service description information is transmitted to the service information preprocessing module. The initial service description information is checked and preprocessed by the service information preprocessing module to obtain preprocessed service description information. Specifically, the service information preprocessing module performs information error check, information missing check, and information format check, etc. on the initial service description information, deletes or corrects the errors in the initial service description information, and obtains the preprocessed service description information.

[0072] In this embodiment, the initial service description information input by the user is received by the application layer of the multi-modal network system, and the initial service description information is preprocessed to improve the data quality of the service description information, so as to realize accurate analysis and processing of the service description information.

[0073] In one embodiment, as shown in Figure 4 Before step 202, the method further includes:

[0074] Step 401, obtaining service description information in natural language form.

[0075] In implementation, the user can input the original intention data in natural language form, and obtain service description information in combination with the basic parameter information of the current multi-modal network system. For example, the service description information is: "now link A and B nodes with high-speed link". In this way, the application layer of the multi-modal network system transmits the service description information to the service layer, which can include a service intention modeling module, a network resource perception module, a service information extraction module, and a service policy generation module, and the like. Therefore, when the service layer receives the service description information, the service description information can be further processed.

[0076] At step 402, the service description information is processed based on a preset mapping function and a preset modeling framework to construct a service intention model.

[0077] The service intention model includes a first intention model and a second intention model.

[0078] In implementation, the service intention modeling module in the application layer processes the service description information based on a preset mapping function and a preset modeling framework (for example, Unified Modeling Language (UML) or Resource Description Framework (RDF)) to determine the related attributes of the service intention model, so that the service intention model can be constructed. The service intention model includes a general intention model (also referred to as a first intention model) and a specific intention model (also referred to as a second intention model). In this way, the modular method is used for intention modeling to determine the related attributes of the general intention model and the specific intention model, so that the general vocabulary model of the general service attribute can be combined with the specific domain model of the specialized vocabulary of the related field concept to accurately generate the service service policy.

[0079] In this embodiment, the service intention model corresponding to the service description information is constructed through the pre-defined mapping function and the preset modeling framework, and then information extraction is performed based on the service service policy to realize the automatic generation of the service service policy, realize the dynamic decision of the service service, and improve the flexibility of the service service decision.

[0080] In one of the optional embodiments, the service intention model is defined in advance, which can be a tree structure. Specifically, the service intention model is defined as wherein represents a general intention model, which is used to extract general intention (for example, service flow information, transmission demand, and the like), is used to represent a specific intention model, and the information extracted in the specific intention model is specific intention information, that is for a specific intent (security requirement, etc.). In this way, by information extraction of the business description information through the general intent model (first intent model) and the specific intent model (second intent model) in the business intent model, the business requirement information in the business description information can be obtained, such as Figure 5 As shown in the business intent model, the root node extracted contains business description information, and the original intent in the business description information is "now link A and B nodes with high-speed link". The business intent modeling module performs information recognition on the business description information, extracts the information content in the business description information through a preset mapping function and a preset modeling framework, for example, extracts the link in the business description information: the source node A and the destination node B at both ends of the communication link. The extracted information transmission condition is: information transmission speed 10 mb / s (10 megabits per second). The extracted information transmission time is: 20:05, and thus the business requirement information in the business description information is obtained.

[0081] In one of the embodiments, as shown in Figure 6 The specific processing process of step 204 includes:

[0082] Step 602, input the business requirement information and the network resource information into the first strategy function pre-trained.

[0083] The first strategy function is updated based on the policy gradient update algorithm and the reward function corresponding to the execution strategy feedback information.

[0084] In the implementation, the service layer pre-defines various information formats, specifically, the service layer defines the business requirement information as , wherein the business requirement information contains business basic parameters and business service expectations, wherein the business basic parameters (including source node, destination node, business mode, business level, etc.), and the business service expectations (including forwarding performance expectations, reliability expectations, security expectations, cost benefit expectations, etc.). The network resource information is defined as (including storage, computing, and forwarding resource information in the network resource pool).

[0085] Thus, in practical applications, after the service layer's business information extraction module extracts the business requirement information from the business description information, it transmits this information to the service layer's service policy generation module. Simultaneously, the resource pool in the network controller aggregates real-time network information. The service layer's network resource awareness module extracts this real-time information from the network controller to obtain network resource information. The business requirement information and network resource information are in the predefined data format described above. Then, the network resource awareness module transmits the network resource information to the service policy generation module. The service policy generation module uses a deep reinforcement learning algorithm, defining the input state vector at time t as... Then, the state vector The input is fed into a pre-trained first policy function. This first policy function can be expressed as: . For the corresponding number At any given moment, in response to the state Business service strategy The action output.

[0086] Optionally, the specific algorithm training process for the first policy function includes: defining... The network feedback information obtained after execution is The network feedback information includes the degree to which service expectations are met, the number of network congestion events, etc. This indicates that the business service strategy is executed at time t. And define the reward function based on the feedback information from the final business service strategy execution. To maximize the cumulative expected reward For the goal, among which, Let be the discount factor at time t. Update the first policy function using the policy gradient update method. , To update the step size, ,in From the first The cumulative reward starts from a certain time. The training of the first policy function is considered complete when the number of policy gradient updates of the first policy function reaches a preset threshold.

[0087] Step 604: Process the business requirement information and network resource information through the first strategy function to generate a business service strategy.

[0088] In implementation, the multimodal network system defines the service strategy as follows: (including network service modalities, network resource allocation strategies, security guarantee strategies, reliability guarantee strategies, etc.). The service strategy generation module in the service layer of the multi-modal network system processes the business demand information and the network resource information through a first policy function of reinforcement learning to generate customized business service strategies, .

[0089] In this embodiment, the first policy function of reinforcement learning is used to process the business demand information and the network resource information to generate a business service strategy that is adapted to the current network environment and the business service strategy. The flexible customization of the business service strategy ensures that the business service strategy meets the current network conditions and the business demand.

[0090] In one of the embodiments, as shown in Figure 7 , the specific processing process of step 206 includes:

[0091] Step 702: Processing the business service strategy and the network node state information through the second policy function to obtain the business routing planning information.

[0092] The business routing planning information includes the transmission paths between the network nodes corresponding to the business service.

[0093] In implementation, the network node state information is defined in the multi-modal network system (wherein represents the number of nodes in the network, and any is the network node identifier), and further, the state information of each network node can be subdivided into (including node function chain information, node resource information, etc.). The business routing planning information is defined as , wherein to refers to the network nodes on the routing path. The network node execution strategy is defined as , including node resource matching, function chain adaptation, etc.). In actual application, the service strategy generation module in the control layer adopts a hierarchical deep reinforcement learning algorithm, the first layer being a routing strategy output network and the second layer being a node execution strategy output network. Thus, the routing strategy output network (i.e., the routing planning generation module) in the control layer is pre-trained with the second policy function (i.e., the second policy sub-function (1), which can be composed of a graph neural network and an attention mechanism network) . Wherein is the routing output for the state at the moment. That is, the routing planning generation module in the control layer receives the business service strategy from the service layer and output network node state information according to the network state information perception module . The service service policy and the service network node state information are taken as input vectors , and input into the second policy function (1). The service service policy and the service network node state information are processed by the second policy function (the second policy function (1)) to obtain service routing planning information, that is .

[0094] Step 704, input the transmission path and the network node state information into the second policy function to generate the execution policy of each network node.

[0095] In implementation, the second policy function (the second policy function (2)) is also pre-trained in the execution policy output network of the control layer. The second policy function (2) (composed of a graph neural network and an attention mechanism network) is used to generate the execution policy of each network node. is the execution policy output of each routing node at the time , according to the routing output policy , the network state and the service policy . In this way, after obtaining the service routing planning information, the service routing planning information contains the transmission path between each network node. Then, the multi-modal network system inputs the transmission path and the network node state information into the second policy function (the second policy function (2)) to generate the execution policy of each network node. .

[0096] In this embodiment, the service service policy and the network node state information are processed by each functional module in the control layer of the multi-modal network system to obtain the execution policy of each network node, so as to realize flexible configuration and management of each network node, improve the flexibility of network management, and meet the service processing requirements.

[0097] In one of the embodiments, as shown in Figure 8 , the method further includes:

[0098] Step 802, instructing to input the execution policy into the front-end compiler of each network node, processing by the front-end compiler to generate a hardware language representation.

[0099] In implementation, after the execution policy of each network node is determined in the control layer of the multi-modal network system, the execution policy of each network node is transmitted to the resource layer of the multi-modal network system, and the resource layer receives the service flow of the access layer of the multi-modal network system. In this way, each network node contained in the resource layer will execute the corresponding execution policy to process the service flow, that is, the execution policy is input into the front-end compiler to generate a preliminary hardware language representation.

[0100] Step 804, the hardware language representation is transmitted to a backend compiler, code compilation and assembly processing are performed by the backend compiler, and the execution strategy is executed.

[0101] In implementation, after obtaining the hardware language representation, the front-end compiler in each network node transmits the hardware language representation to the backend compiler, then the backend compiler performs code compilation and assembly processing on the received intermediate identification, executes the assembled execution strategy, processes the service flow of the access layer, and provides corresponding services.

[0102] In the embodiment, the execution strategy of each network node is determined and executed in each network node to process the received service flow, the management of the network node is realized, the current network condition and service demand are met, and the service quality is ensured.

[0103] It should be understood that, although each step in the flowchart involved in each embodiment described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0104] Based on the same inventive concept, the embodiments of the present application also provide a network management device for implementing the network management method described above. The implementation scheme of the problem solving provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more network management device embodiments provided below can refer to the limitations of the network management method described above, and will not be repeated here.

[0105] In one exemplary embodiment, as shown in Figure 9 A network management device 900 is provided, which includes an extraction module 910, a generation module 920, and a processing module 930, wherein:

[0106] The extraction module 910 is configured to extract service demand information in the service description information according to a pre-constructed service intention model in response to the received service description information.

[0107] The generation module 920 is configured to process the service demand information and the network resource information by a first strategy function to generate a service service strategy.

[0108] The processing module 930 is configured to process the service service policy and the network node state information based on a second policy function to obtain an execution policy of each network node; and the execution policy of each network node is used to implement network service management.

[0109] In one of the embodiments, the apparatus 900 further includes:

[0110] The receiving module is configured to receive initial service description information.

[0111] The preprocessing module is configured to perform abnormality checking and format preprocessing on the initial service description information to obtain service description information.

[0112] In one of the embodiments, the apparatus further includes:

[0113] The obtaining module is configured to obtain service description information in a natural language form.

[0114] The construction module is configured to process the service description information based on a preset mapping function and a preset modeling framework to construct a service intent model; the service intent model includes a first intent model and a second intent model.

[0115] In one of the embodiments, the generation module 920 is specifically configured to input the service requirement information and the network resource information into a pre-trained first policy function; the first policy function is updated based on a policy gradient update algorithm and a reward function corresponding to the execution policy feedback information to obtain

[0116] The service requirement information and the network resource information are processed by the first policy function to generate a service service policy.

[0117] In one of the embodiments, the processing module 930 is specifically configured to process the service service policy and the network node state information by a second policy function to obtain service routing planning information; the service routing planning information includes transmission paths between network nodes corresponding to the service service.

[0118] The transmission paths and the network node state information are input into the second policy function to generate the execution policy of each network node.

[0119] In one of the embodiments, the apparatus 900 further includes:

[0120] The indication module is configured to indicate that the execution policy is input into a front-end compiler of each network node, and the execution policy is processed by the front-end compiler to generate a hardware language representation.

[0121] The execution module is configured to transmit the hardware language representation to a back-end compiler, and the execution module is configured to perform code compilation and assembly processing by the back-end compiler to execute the execution policy.

[0122] Each module in the network management apparatus can be implemented by software, hardware, and combinations thereof, in whole or in part. Each module can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be invoked and executed by a processor to perform operations corresponding to each module.

[0123] In an exemplary embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in FIG. 1. Figure 10 The computer device includes a processor, a memory, an input / output interface, and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store service-related data and network-related data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a network management method.

[0124] In an exemplary embodiment, a computer device, which can be a terminal, is provided, and an internal structure diagram of the computer device can be as shown in FIG. 2. Figure 10The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected through a system bus. The communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be implemented through WIFI, mobile cellular network, Near Field Communication (NFC), or other technologies. The computer program is executed by the processor to implement a network management method. The display unit of the computer device is configured to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or can be a key, a trackball, or a touchpad arranged on the shell of the computer device, or can be an external keyboard, a touchpad, a mouse, or the like.

[0125] Those skilled in the art can understand that, Figure 10 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0126] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0127] In response to the received service description information, extracting service requirement information in the service description information according to a pre-constructed service intention model;

[0128] Processing the service requirement information and the network resource information through a first policy function to generate a service service policy;

[0129] Processing the service service policy and the network node state information based on a second policy function to obtain an execution policy of each network node; the execution policy of each network node is used to implement network service management.

[0130] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0131] receiving initial service description information;

[0132] performing exception check and format preprocessing on the initial service description information to obtain service description information.

[0133] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0134] obtaining service description information in natural language form;

[0135] processing the service description information based on a preset mapping function and a preset modeling framework to construct a service intent model; the service intent model includes a first intent model and a second intent model.

[0136] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0137] inputting the service requirement information and the network resource information into a pre-trained first strategy function; the first strategy function is updated based on a policy gradient update algorithm and a reward function corresponding to the execution strategy feedback information to obtain

[0138] processing the service requirement information and the network resource information through the first strategy function to generate a service service strategy.

[0139] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0140] processing the service service strategy and the network node state information through the second strategy function to obtain service routing planning information; the service routing planning information includes transmission paths between network nodes corresponding to the service service;

[0141] inputting the transmission path and the network node state information into the second strategy function to generate an execution strategy for each network node.

[0142] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0143] indicating that the execution strategy is input to a front-end compiler of each network node, and processing through the front-end compiler to generate a hardware language representation;

[0144] transmitting the hardware language representation to a back-end compiler, and performing code compilation and assembly processing through the back-end compiler to execute the execution strategy.

[0145] In an embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

[0146] In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

[0147] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0148] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, the processes of the above-mentioned embodiment methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., and is not limited thereto. The processor involved in the embodiments provided in the present application can be a general processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., and is not limited thereto.

[0149] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, it should be considered as the scope of the present application.

[0150] The above embodiments only express several implementation ways of the present application, and the description is specific and detailed, but it should not be understood as a limitation to the patent scope of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A network management method characterized by comprising: The method comprises: in response to the received service description information, extracting service requirement information in the service description information according to a pre-constructed service intention model; processing the service requirement information and network resource information through a first strategy function to generate a service service strategy; processing the service service strategy and network node state information based on a second strategy function to obtain an execution strategy of each network node; the execution strategy of each network node is used to implement network service management; the processing of the service requirement information and network resource information through the first strategy function to generate the service service strategy comprises: inputting the service requirement information and network resource information into a pre-trained first strategy function; the first strategy function is updated based on a policy gradient update algorithm and a reward function corresponding to execution strategy feedback information; processing the service requirement information and network resource information through the first strategy function to generate the service service strategy; the processing of the service service strategy and network node state information based on the second strategy function to obtain the execution strategy of each network node comprises: processing the service service strategy and network node state information through a second strategy function to obtain service routing planning information; the service routing planning information contains transmission paths between network nodes corresponding to the service service; inputting the transmission paths and network node state information into the second strategy function to generate the execution strategy of each network node.

2. The method of claim 1, wherein, Before the response to the received service description information, the method further comprises: receiving initial service description information; performing abnormality check and format preprocessing on the initial service description information to obtain service description information.

3. The method of claim 1, wherein, Before the extraction of the service requirement information in the service description information according to the pre-constructed service intention model, the method further comprises: obtaining service description information in natural language form; processing the service description information based on a preset mapping function and a preset modeling framework to construct a service intention model; the service intention model comprises a first intention model and a second intention model.

4. The method of claim 1, wherein, The method further comprises: indicating inputting the execution strategy into a front-end compiler of each network node to process through the front-end compiler to generate a hardware language representation; transmitting the hardware language representation to a back-end compiler to perform code compilation and assembly processing through the back-end compiler to execute the execution strategy.

5. A network management apparatus characterized by comprising: The device comprises: an extraction module configured to extract service requirement information in the service description information according to a pre-constructed service intention model in response to received service description information; a generation module configured to process the service requirement information and network resource information through a first strategy function to generate a service service strategy; a processing module configured to process the service service strategy and network node state information based on a second strategy function to obtain an execution strategy of each network node; the execution strategy of each network node is used to implement network service management; The generating module is configured to input the service demand information and the network resource information into a first policy function which is obtained by updating a reward function corresponding to policy feedback information based on a policy gradient update algorithm; and generate a service policy by processing the service demand information and the network resource information through the first policy function. The processing module is configured to process the service policy and network node state information through a second policy function to obtain service routing planning information, wherein the service routing planning information includes transmission paths between network nodes corresponding to the service policy; and input the transmission paths and the network node state information into the second policy function to generate an execution policy of each network node.

6. The apparatus of claim 5, wherein, The apparatus further includes: a receiving module configured to receive initial service description information; a preprocessing module configured to perform abnormality checking and format preprocessing on the initial service description information to obtain service description information.

7. The apparatus of claim 5, wherein, The apparatus includes: an obtaining module configured to obtain service description information in a natural language form; a constructing module configured to process the service description information based on a preset mapping function and a preset modeling framework to construct a service intent model, wherein the service intent model includes a first intent model and a second intent model. 8.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-7. The processor, when executing the computer program, implements the steps of the method in any one of claims 1 to 4.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method in any one of claims 1 to 4.

10. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method in any one of claims 1 to 4.

Images