Service request processing method, network element and related equipment

Abstract

The invention provides a service request processing method, a network element and related equipment, and relates to the technical field of communication. The method comprises the following steps: receiving an AI service creation request sent by a converged service management network element; querying an AI service required by the AI service creation request, and determining computing power resources required for running the AI service; a computing power resource creation request is sent to the computing power strategy control network element, and the computing power resource creation request carries computing power resources; receiving a computing power resource creation response sent by the computing power strategy control network element; and sending an AI service creation response to the converged service management network element. Through the above technical means, the problem that cooperative management of computing power and AI service cannot be realized in a communication network in related technologies is solved.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a service request processing method, network element and related equipment. Background Technology

[0002] Currently, the core network's AI capabilities are mainly realized through the Network Data Analysis Element (NWDAF). The data collection and coordination element (DCCF) and the management data analysis element (MFAF) work together to complete data collection and transmission. Combined with the analysis and data storage element (ADRF), models and data are stored. This supports intelligent services such as multi-access service load prediction and is mainly used for internal analysis and optimization of the communication network.

[0003] In reality, current communication networks only transmit data for AI services; AI service requests must be handled by external internet service providers. The communication networks themselves lack intelligent service capabilities and cannot coordinate the allocation of computing power and AI resources. Future communication networks will not only need to leverage AI to improve operational efficiency and resource utilization, but also serve as service platforms to provide personalized AI services. Therefore, it is urgent to build a converged computing and network architecture that integrates edge, cloud, and endpoint technologies, with the core network as the scheduling hub to uniformly manage distributed computing power and AI services. Summary of the Invention

[0004] This disclosure provides a service request processing method, network element and related equipment to realize the integrated deployment and on-demand scheduling of AI services and computing power, and promote the intelligent evolution of communication networks.

[0005] According to one aspect of this disclosure, a service request processing method is provided, applied to an artificial intelligence (AI) policy control network element, comprising: receiving an AI service creation request sent by a converged service management network element; querying the AI ​​service required by the AI ​​service creation request and determining the computing power resources required to run the AI ​​service; sending a computing power resource creation request to a computing power policy control network element, wherein the computing power resource creation request carries computing power resources; receiving a computing power resource creation response sent by the computing power policy control network element; and sending an AI service creation response to the converged service management network element.

[0006] In one embodiment of this disclosure, before receiving the computing resource creation response sent by the computing power policy control network element, the method further includes: receiving an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation request carries an identifier and name of an AI service; and receiving an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation response carries a download address, identifier, and name of the AI ​​service.

[0007] In one embodiment of this disclosure, the method further includes: receiving an AI service registration request sent by an AI service, wherein the AI ​​service registration request carries a configuration file of the AI ​​service; storing the configuration file and determining that the AI ​​service is available; and sending an AI service registration response to the AI ​​service.

[0008] In one embodiment of this disclosure, the configuration file includes at least one of the following: basic service information, service model information, computing resources, and input / output information; the basic service information includes at least one of the following: the identifier, name, version number, provider, and download address of the AI ​​service; the service model information includes at least one of the following: the name and performance metrics of the model, and the dataset corresponding to the performance metrics; the computing resources include at least one of the following: CPU information, GPU information, and memory information required to run the AI ​​service.

[0009] In one embodiment of this disclosure, the method further includes: receiving an AI service update request sent by an AI service, wherein the AI ​​service update request carries the latest configuration file of the AI ​​service; storing and updating the configuration file; and sending an AI service update response to the AI ​​service.

[0010] In one embodiment of this disclosure, the method further includes: receiving an AI service deregistration request sent by an AI service, wherein the AI ​​service deregistration request; determining that the state of the AI ​​service is unavailable, deleting the configuration file; and sending an AI service deregistration response to the AI ​​service.

[0011] According to another aspect of this disclosure, a service request processing method is provided, applied to a converged service management network element, comprising: receiving a converged service creation request sent by a session management network element; determining that the converged service creation request belongs to an AI service; sending an AI service creation request to an AI policy control network element; receiving an AI service creation response fed back by the AI ​​policy control network element; and feeding back a converged service creation response to the session management network element.

[0012] In one embodiment of this disclosure, determining that a fusion service creation request belongs to an artificial intelligence (AI) service includes: parsing the fusion service creation request to obtain an identifier; determining the service type of the fusion service creation request based on the identifier; and determining that the fusion service creation request belongs to an AI service based on the service type.

[0013] In one embodiment of this disclosure, the AI ​​service creation request carries the Quality of Service (QoS) requirement information of the AI ​​service, which includes an identifier, latency requirement information, and accuracy requirement information.

[0014] According to another aspect of this disclosure, an AI policy control network element is provided, comprising: a first receiving unit configured to receive an AI service creation request sent by a converged service management network element; a querying unit configured to query the AI ​​service required by the AI ​​service creation request and determine the computing power resources required to run the AI ​​service; a first sending unit configured to send a computing power resource creation request to the computing power policy control network element, wherein the computing power resource creation request carries computing power resources; a second receiving unit configured to receive a computing power resource creation response sent by the computing power policy control network element; and a second sending unit configured to send an AI service creation response to the converged service management network element.

[0015] According to another aspect of this disclosure, a converged service management network element is provided, characterized in that it includes: a third receiving unit configured to receive a converged service creation request sent by a session management network element; a determining unit configured to determine that the converged service creation request belongs to an AI service; a third sending unit configured to send an AI service creation request to an AI policy control network element; a fourth receiving unit configured to receive an AI service creation response fed back by the AI ​​policy control network element; and a fourth sending unit configured to feed back the converged service creation response to the session management network element.

[0016] According to another aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any of the methods described above by executing the executable instructions.

[0017] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements any of the methods described above.

[0018] According to another aspect of this disclosure, a computer program product is provided, including computer instructions stored in a computer-readable storage medium, which, when executed by a processor, implement operation instructions for any of the methods described above.

[0019] In the embodiments of this disclosure, the AI ​​service required for the AI ​​service creation request is queried, the computing power resources required to run the AI ​​service are determined, a computing power resource creation request carrying computing power resources is sent to the computing power policy control network element, and a computing power resource creation response sent by the computing power policy control network element is received. This solves the problem in related technologies that the collaborative management of computing power and AI services cannot be realized within the communication network, thereby realizing the integrated deployment and on-demand scheduling of AI services and computing power, and promoting the intelligent evolution of the communication network.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0022] Figure 1 This diagram illustrates a service request processing system according to an embodiment of the present disclosure.

[0023] Figure 2 A flowchart of a service request processing method according to an embodiment of this disclosure is shown.

[0024] Figure 3 A flowchart illustrating an AI resource creation method according to an embodiment of this disclosure is shown.

[0025] Figure 4 A flowchart of an AI service registration method according to an embodiment of this disclosure is shown.

[0026] Figure 5 A flowchart of an AI service update method according to an embodiment of this disclosure is shown.

[0027] Figure 6 A flowchart of an AI service deregistration method is shown in an embodiment of this disclosure.

[0028] Figure 7 A flowchart illustrating another service request processing method in an embodiment of this disclosure is shown.

[0029] Figure 8 A flowchart of another service request processing method in an embodiment of this disclosure is shown.

[0030] Figure 9 A schematic diagram of an AI strategy control network element is shown in an embodiment of this disclosure.

[0031] Figure 10 This diagram illustrates a converged service management network element according to an embodiment of the present disclosure.

[0032] Figure 11 A schematic diagram of an electronic device provided in an embodiment of this disclosure is shown. Detailed Implementation

[0033] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0034] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0035] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0036] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0037] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0038] It should be noted that, unless otherwise specified, the embodiments of this disclosure and the technical features thereof can be combined with each other.

[0039] To facilitate understanding, the following is an explanation of several terms used in this disclosure: UE (User Equipment): refers to devices used to access mobile networks and use various services, including smartphones, tablets, and IoT devices. These devices can initiate and receive communications and utilize network resources to perform specific tasks.

[0040] Core Network (CN): A crucial component of a mobile communication system responsible for data transmission, routing, and quality of service assurance. It connects the radio access network to external networks (such as the Internet) and supports the transmission of voice, data, and multimedia services.

[0041] SMF (Session Management Function): A key component of the core network, its main responsibilities include establishing, modifying, and releasing PDU sessions. This function ensures that user equipment can smoothly access the network and enjoy various services.

[0042] AMF (Access and Mobility Management Function): In the core network, it is responsible for user registration, connection management, and mobility control. It ensures that user equipment can seamlessly switch between different locations while maintaining stable service connections.

[0043] QoS (Quality of Service) describes the level of performance guarantees a network provides for specific traffic or applications, including but not limited to requirements regarding latency, jitter, and packet loss rate. A good QoS mechanism can ensure that critical services receive priority processing, thereby improving user experience.

[0044] PDU (Protocol Data Unit): A data structure defined at the network layer that contains data passed down from the upper layer and necessary control information, used to enable data transmission in the network.

[0045] CPCF (Computing Policy Control Function): With the allocation and management of computing resources as its core task, CPCF is responsible for selecting appropriate computing nodes and scheduling resources based on the AI ​​service demand information provided by AIPCF, so as to ensure that AI services can obtain the necessary CPU, GPU and memory resources as needed.

[0046] ISMF (Integrated Service Management Function): Its main responsibility is to process service requests from various parts of the network and identify those involving AI services. Through collaboration with other functional entities, such as AIPCF, it enables the creation, updating, and deletion of AI services.

[0047] AIPCF (AI Policy Control Function): Focuses on the registration, discovery, matching, and quality of service assurance of AI services, supporting AI service requests initiated by internal or external entities. Its core functions include maintaining AI service configuration files and service status, and collaborating with CPCF to complete the integrated deployment of AI models and computing resources.

[0048] MFAF (Management Data Analytics Function): A component within a network management system, primarily used for analyzing operational and maintenance data. By collaborating with data analytics functions such as NWDAF, it can analyze network operations across different domains, supporting decision-making.

[0049] ADRF (Analytics Data Repository Function): This function stores datasets, trained AI models, and analysis results required for network data analysis. It supports persistent data storage and version management, ensuring the continuity and traceability of data analysis work.

[0050] MTLF (Multi-access Traffic Load Forecasting) is a typical data analysis capability primarily used to predict traffic load under different access methods in communication networks. Through horizontal (cross-regional) and vertical (cross-time) federated learning mechanisms, it can improve prediction accuracy while protecting user privacy.

[0051] CPU (Central Processing Unit): One of the core hardware components of a computer system, responsible for interpreting and executing program instructions, performing arithmetic and logical operations, and coordinating the work of other hardware components.

[0052] GPU (Graphics Processing Unit): A microprocessor designed specifically for processing images and videos. In recent years, it has also been widely used in fields such as machine learning and scientific computing, and is favored for its powerful parallel computing capabilities.

[0053] Figure 1 The diagram illustrates a service request processing system according to an embodiment of the present disclosure. The service request processing system includes: UE101, AMF102, SMF103, ISMF104, AIPCF105, and CPCF106.

[0054] UE101 initiates an AI service request.

[0055] AMF102 is responsible for selecting the SMF that supports AI tasks.

[0056] SMF103 is responsible for handling PDU session establishment requests and forwarding AI service-related requests.

[0057] ISMF104 is responsible for parsing fusion service requests and determining whether they are AI services.

[0058] AIPCF105 is responsible for managing the registration, discovery, scheduling, and QoS assurance of AI services. AIPCF105 connects multiple AI service instances, including: AI Service A, AI Service B, and AI Service C.

[0059] CPCF106 is responsible for managing the allocation and deployment of computing resources. CPCF106 connects multiple computing nodes, including: computing node 1, computing node 2, and computing node 3.

[0060] AIPCF serves as the central hub for AI service capabilities, while CPCF serves as the central hub for computing resources, together supporting the scheduling and execution of end-to-end AI services.

[0061] The CPCF106 can be equipped with an application that performs the following actions: receiving an AI service creation request from the converged service management network element; querying the AI ​​service required by the AI ​​service creation request and determining the computing resources required to run the AI ​​service; sending a computing resource creation request to the computing power policy control network element, wherein the computing resource creation request carries computing resources; receiving a computing resource creation response from the computing power policy control network element; and sending an AI service creation response to the converged service management network element.

[0062] The AIPCF105 can be equipped with an application that performs the following actions: receiving a converged service creation request from the session management network element; determining that the converged service creation request belongs to the AI ​​service; sending an AI service creation request to the AI ​​policy control network element; receiving an AI service creation response from the AI ​​policy control network element; and sending a converged service creation response back to the session management network element.

[0063] Figure 2 This diagram illustrates a flowchart of a service request processing method according to an embodiment of the present disclosure. The method is applied to a converged service management network element. The method is as follows: Figure 2 As shown, it includes the following steps: S201, Receive AI service creation request sent by converged service management network element.

[0064] As an example, an AI service creation request is a message sent by the Converged Service Management Element (ISMF) to the AI ​​Policy Control Element (AIPCF) to request the activation of a specific AI service, such as vehicle navigation and image recognition.

[0065] In one embodiment of this disclosure, the AI ​​service creation request carries the Quality of Service (QoS) requirement information of the AI ​​service, which includes an identifier, latency requirement information, and accuracy requirement information.

[0066] As an example, QoS requirement information is used to describe the performance guarantee requirements of AI services.

[0067] As an example, identifiers are used to identify AI services.

[0068] As an example, latency requirement information is used to specify the upper limit of AI server-side end-to-end processing latency.

[0069] Exemplary accuracy requirement information is used to specify the accuracy metrics that an AI service model needs to achieve for inference or training.

[0070] In one exemplary embodiment, a new Container is added, such as a message with the 10010100 message. The last two bits identify whether the message belongs to an AI inference task or an AI training task, the middle three bits identify latency requirements, and the high three bits identify accuracy requirements.

[0071] S202, query the AI ​​services required for the AI ​​service creation request, and determine the computing resources required to run the AI ​​services.

[0072] As an example, computing resources include at least one of the following: CPU information, GPU information, and memory information required to run AI services.

[0073] As an example, computing resources are a description of the computing hardware resources required to run AI services.

[0074] As an example, CPU information is a parameter that describes the general computing power required to run AI services. CPU information includes the number of CPU cores, clock speed requirements, or computing power level.

[0075] As an example, GPU information is a parameter describing the parallel accelerated computing power required to run AI services. GPU information includes GPU model type, memory capacity requirements, or floating-point performance metrics.

[0076] As an example, memory information is a parameter describing the main storage space required to run AI services, including memory capacity and access bandwidth requirements.

[0077] S203, send a computing resource creation request to the computing power policy control network element, wherein the computing resource creation request carries computing resources.

[0078] As an example, a computing resource creation request is a message sent by the AIPCF to the computing policy control network element CPCF to request the allocation of computing resources.

[0079] S204, Receive the computing resource creation response sent by the computing power policy control network element.

[0080] As an example, the computing resource creation response is a message returned by CPCF to AIPCF confirming that computing resources have been allocated. The computing resource creation response includes the actual allocated CPU resources, GPU resources, memory resources, computing node identifier, and service deployment status.

[0081] S205, send an AI service creation response to the converged service management network element.

[0082] As an example, the AI ​​service creation response is a message returned by AIPCF to ISMF confirming that the AI ​​service can be scheduled. The AI ​​service creation response includes the service matching result, confirmed QoS information, the download address of the AI ​​service, and its identifier. The service matching result includes whether an AI service was matched or not, and the confirmed QoS information includes the identifier, the latency requirements that can be met, and the accuracy requirements that can be achieved.

[0083] In this embodiment, the AIPCF receives an AI service creation request, queries the required AI service, determines the computing resources needed to run the AI ​​service, sends a computing resource creation request carrying the computing resource requirements to the CPCF, receives the computing resource creation response from the CPCF, and sends an AI service creation response to the ISMF. This addresses the problem in related technologies where collaborative management of computing power and AI services cannot be achieved within the communication network, thereby enabling integrated deployment and on-demand scheduling of AI services and computing power, and promoting the intelligent evolution of communication networks.

[0084] Figure 3 This invention discloses a flowchart of an AI resource creation method according to an embodiment of the present disclosure. The method is as follows: Figure 3 As shown, it includes the following steps: S301, receive an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation request carries the identifier and name of the AI ​​service; S302 receives an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation response carries the download address, identifier and name of the AI ​​service.

[0085] As an example, an AI resource creation request is a message sent by the CPCF to the AIPCF to obtain deployment information for a specific AI service. The AIPCF determines the corresponding AI service deployment information based on the identifier and name of the AI ​​service carried in the AI ​​resource creation request.

[0086] As an example, the AI ​​resource creation response is a message returned by AIPCF to CPCF that provides information on AI service deployment, specifically including the download address, identifier, and name of the AI ​​service.

[0087] As an example, the download address of an AI service is a network location identifier used to obtain the AI ​​service model file or executable program, and the download address of the AI ​​service can use the Uniform Resource Locator (URL) format.

[0088] As an example, an identifier is a string used to uniquely identify an instance of an AI service.

[0089] As an example, the name is an identifier used to describe the functionality and purpose of the AI ​​service. The name is defined by the provider during registration, such as "image classification service" or "speech recognition service".

[0090] In this embodiment, the CPCF determines computing nodes that meet the required computing resources. If no AI service is deployed on the computing node, an AI resource creation request is sent. The AIPCF receives the AI ​​resource creation request and returns an AI resource creation response containing the service download address to the CPCF. Through the above technical means, the CPCF can dynamically acquire and deploy the required AI services on the target computing node, realizing on-demand collaborative supply of AI models and computing resources.

[0091] Figure 4 This disclosure illustrates a flowchart of an AI service registration method according to an embodiment of the present disclosure. The method is as follows: Figure 4 As shown, it includes the following steps: S401, Receive AI service registration request sent by AI service, wherein the AI ​​service registration request carries the configuration file of AI service; S402, store the configuration file and determine that the AI ​​service is available; S403, Send AI service registration response to AI service.

[0092] As an example, an AI service registration request is a message sent by an AI service to the AIPCF to declare its service capabilities.

[0093] As an example, the AI ​​service registration response is a message returned by AIPCF to the AI ​​service to confirm successful registration.

[0094] As an example, the configuration file is a structured collection of metadata describing the capabilities and requirements of AI services.

[0095] In one embodiment of this disclosure, the configuration file includes at least one of the following: basic service information, service model information, computing resources, and input / output information; the basic service information includes at least one of the following: the identifier, name, version number, provider, and download address of the AI ​​service; the service model information includes at least one of the following: the name and performance metrics of the model, and the dataset corresponding to the performance metrics; the computing resources include at least one of the following: CPU information, GPU information, and memory information required to run the AI ​​service.

[0096] As an example, basic service information consists of attributes that identify the identity and origin of an AI service.

[0097] As an example, the version number is a string that identifies the iterative evolution status of the AI ​​service.

[0098] As an example, the provider is an entity that claims ownership or development rights for the AI ​​service.

[0099] Exemplary service model information is data that describes the algorithms used in an AI service and their performance.

[0100] As an example, the model's name is an identifier for the artificial intelligence algorithm or neural network architecture used by the AI ​​service.

[0101] As an example, performance metrics are quantitative parameters that measure the ability of an AI service to perform on a specific task. Performance metrics include accuracy, precision, recall, or inference throughput, etc.

[0102] As an example, the dataset corresponding to the performance metrics is a standardized set of test data used to evaluate the performance metrics of the AI ​​service. The dataset corresponding to the performance metrics is used to ensure the verifiability and comparability of performance claims.

[0103] As an example, the input information defines input-related properties, such as maxlength=512 indicating the maximum length of the input token.

[0104] As an example, the output information defines the relevant characteristics of the output, such as the number of categories in a classification problem.

[0105] In this embodiment, AIPCF receives an AI service registration request sent by an AI service, parses and stores the configuration file carried therein, marks the status of the AI ​​service as available, and returns an AI service registration response to the AI ​​service. Through the above technical means, the network can perceive and manage the inherent AI service capabilities, providing basic data support for subsequent service discovery, matching, and scheduling.

[0106] In one exemplary embodiment, the configuration file is shown in the following table:

[0107] Table 1

[0108] Figure 5 This diagram illustrates a flowchart of an AI service update method according to an embodiment of the present disclosure. The method is as follows: Figure 5 As shown, it includes the following steps: S501, receive an AI service update request sent by the AI ​​service, wherein the AI ​​service update request carries the latest configuration file of the AI ​​service; S502, store and update configuration files; S503, send an AI service update response to the AI ​​service.

[0109] As an example, an AI service update request is a message sent by an AI service to the AIPCF to update the description of its registered service capabilities.

[0110] As an example, the AI ​​service update response is a message returned by AIPCF to the AI ​​service to confirm that the configuration file update was successful.

[0111] In this embodiment, AIPCF receives an AI service update request from the AI ​​service, retrieves the latest configuration file carried within, updates the corresponding stored configuration file, and returns an AI service update response to the AI ​​service. Through these technical means, the AI ​​service can dynamically adjust its capability description or resource requirements during operation, ensuring the real-time nature and accuracy of service information in the network.

[0112] Figure 6 This diagram illustrates a flowchart of an AI service deregistration method according to an embodiment of the present disclosure. The method is as follows: Figure 6 As shown, it includes the following steps: S601, Receive AI service cancellation request sent by AI service, wherein AI service cancellation request; S602, the AI ​​service status is determined to be unavailable, and the configuration file is deleted; S603, send an AI service deregistration response to the AI ​​service.

[0113] As an example, an AI service deregistration request is a message sent by an AI service to the AIPCF to declare that it is about to go offline or cease service.

[0114] As an example, the AI ​​service deregistration response is a message returned by AIPCF to the AI ​​service to confirm that the deregistration operation has been completed.

[0115] In this embodiment, AIPCF receives an AI service deregistration request from an AI service, determines the status of the corresponding AI service as unavailable based on the request, deletes its stored configuration file, and then sends an AI service deregistration response to the AI ​​service. Through these technical means, expired AI service registration information is promptly cleared, preventing the network scheduling system from misusing unavailable services and ensuring the accuracy of service discovery and invocation.

[0116] In an optional embodiment, the method further includes: upon receiving an AI service deregistration request, AIPCF does not immediately delete the configuration file, but instead marks the AI ​​service status as "pending cleanup" and starts a delayed deletion timer; if no update request for recovery is received before the timer expires, the configuration file deletion operation is performed; simultaneously, AIPCF sends a computing power release notification to the computing power policy control network element, instructing it to reclaim the computing resources occupied by the AI ​​service. Through the above technical means, while ensuring the consistency of service status, a recovery window is provided for abnormal interruption scenarios, and underlying computing power resources are released in a coordinated manner.

[0117] In one alternative embodiment, after determining that the AI ​​service is unavailable, there is no need to delete the configuration file; instead, an AI service deregistration response is sent directly to the AI ​​service.

[0118] Figure 7 This invention discloses a flowchart of another service request processing method in an embodiment of the present disclosure. This method is applied to a converged service management network element. The method is as follows: Figure 7 As shown, it includes the following steps: S701 receives a converged service creation request sent by the session management network element; S702, It is determined that the fusion service creation request belongs to the AI ​​service; S703 sends an AI service creation request to the AI ​​policy control network element; S704 receives AI service creation response feedback from AI policy control network element; S705 sends a response to the session management network element regarding the creation of converged services.

[0119] As an example, a converged service creation request is a message sent by a session management network element to a converged service management network element to request the initiation of an intelligent service. The converged service creation request includes the service identifier, QoS requirement information, and session association information.

[0120] As an example, a converged service creation response is a message returned by the converged service management network element to the session management network element to confirm that the converged service (such as an AI service) has been successfully scheduled. The converged service creation response includes the service scheduling result, the confirmed QoS information, and the AI ​​service identifier.

[0121] In this embodiment, the converged service management network element receives a converged service creation request sent by the session management network element, parses the request and determines that it belongs to an AI service, then sends an AI service creation request to the AI ​​policy control network element, receives an AI service creation response from the AI ​​policy control network element, and sends a converged service creation response back to the session management network element. Through the above technical means, the converged service management network element can act as a central hub for identifying and forwarding AI service requests, establishing a collaborative link between session management functions and AI service control functions.

[0122] In one embodiment of this disclosure, determining that a fusion service creation request belongs to an artificial intelligence (AI) service includes: parsing the fusion service creation request to obtain an identifier; determining the service type of the fusion service creation request based on the identifier; and determining that the fusion service creation request belongs to an AI service based on the service type.

[0123] As an example, the service type is an attribute field used to classify service function categories. The service type is declared by the service provider in the configuration file and stored in AIPCF in association with an identifier.

[0124] In this embodiment, the converged service management network element receives a converged service creation request sent by the session management network element, parses the request to obtain an identifier, queries the corresponding service type based on the identifier, determines that the converged service creation request belongs to an AI service based on the service type, and then sends an AI service creation request to the AI ​​policy control network element. It then receives an AI service creation response from the AI ​​policy control network element and sends a converged service creation response back to the session management network element. Through the above technical means, accurate service type identification based on identifiers is achieved, avoiding reliance on hard-coded or protocol extension fields, and improving the flexibility and maintainability of service classification.

[0125] In one optional embodiment, determining that the fusion service creation request belongs to the artificial intelligence (AI) service includes: parsing the fusion service creation request to obtain an identifier and service type; and determining that the fusion service creation request belongs to the AI ​​service based on the identifier and service type.

[0126] Figure 8 This invention discloses a flowchart of another service request processing method in an embodiment of the present disclosure, the method being as follows: Figure 8 As shown, it includes the following steps: S801: The UE sends an AI service session establishment request (the AI ​​service session establishment request is equivalent to the converged service creation request).

[0127] S802: The AMF selects the SMF that supports the AI ​​task based on the AI ​​service identifier.

[0128] S803: AMF forwards the AI ​​service session establishment request to SMF.

[0129] S804: SMF receives and determines that the AI ​​service session establishment request belongs to the AI ​​service.

[0130] S805: SMF sends a fusion service creation request.

[0131] S806: ISMF parses the fusion service creation request and determines whether the fusion service creation request belongs to the AI ​​service by service type and service ID.

[0132] S807: If it is an AI service, then ISMF sends an AI service creation request.

[0133] S808: Based on the AI ​​service creation request, AIPCF compares the AI ​​services registered with AIPCF to determine whether there is an available AI service; if so, AIPCF determines the computing resources required for the service.

[0134] S809: AIPCF sends a computing resource creation request to CPCF, carrying the computing resources.

[0135] S810: CPCF resolves the required computing resources and selects computing nodes that can provide those resources.

[0136] S811: CPCF sends an AI resource creation request to AIPCF, carrying information such as the serviceId and servicename of the required AI service.

[0137] S812: AIPCF sends an AI resource creation response to CPCF, carrying information such as the service download address, serviceId, and servicename.

[0138] S813: CPCF obtains the corresponding AI service and sends an AI resource creation response to AIPCF.

[0139] S814: AIPCF sends an AI service creation response to ISMF, carrying acknowledgment QoS information.

[0140] S815: ISMF sends a fusion service creation response to SMF.

[0141] S816: Send a service response to the UE.

[0142] Based on the same inventive concept, this disclosure also provides an AI policy control network element and a converged service management network element, as shown in the following embodiments. Since the principles by which the AI ​​policy control network element and the converged service management network element solve problems are similar to those in the above method embodiments, the implementation of the AI ​​policy control network element and the converged service management network element can refer to the implementation of the above method embodiments, and repeated details will not be elaborated further.

[0143] Figure 9 This invention discloses an AI policy control network element in an embodiment of the present invention, such as... Figure 9 As shown, the converged service management network element may include: The first receiving unit 901 is configured to receive AI service creation requests sent by the converged service management network element; Query unit 902 is configured to query the AI ​​service required for the AI ​​service creation request and determine the computing resources required to run the AI ​​service. The first sending unit 903 is configured to send a computing resource creation request to the computing power policy control network element, wherein the computing resource creation request carries computing resources; The second receiving unit 904 is configured to receive the computing resource creation response sent by the computing power policy control network element; The second sending unit 905 is configured to send an AI service creation response to the converged service management network element.

[0144] In some embodiments, the second receiving unit 904 is further configured to receive an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation request carries the identifier and name of the AI ​​service; and to receive an AI resource creation response sent by the computing power policy control network element, wherein the AI ​​resource creation response carries the download address, identifier, and name of the AI ​​service.

[0145] In some embodiments, the second receiving unit 904 is further configured to receive an AI service registration request sent by the AI ​​service, wherein the AI ​​service registration request carries a configuration file of the AI ​​service; store the configuration file and determine that the status of the AI ​​service is available; and send an AI service registration response to the AI ​​service.

[0146] In one embodiment of this disclosure, the configuration file includes at least one of the following: basic service information, service model information, computing resources, and input / output information; the basic service information includes at least one of the following: the identifier, name, version number, provider, and download address of the AI ​​service; the service model information includes at least one of the following: the name and performance metrics of the model, and the dataset corresponding to the performance metrics; the computing resources include at least one of the following: CPU information, GPU information, and memory information required to run the AI ​​service.

[0147] In some embodiments, the second receiving unit 904 is further configured to receive an AI service update request sent by the AI ​​service, wherein the AI ​​service update request carries the latest configuration file of the AI ​​service; store and update the configuration file; and send an AI service update response to the AI ​​service.

[0148] In some embodiments, the second receiving unit 904 is further configured to receive an AI service deregistration request sent by the AI ​​service, wherein the AI ​​service deregistration request; determines that the status of the AI ​​service is unavailable, deletes the configuration file; and sends an AI service deregistration response to the AI ​​service.

[0149] Figure 10 This disclosure illustrates a converged service management network element, such as... Figure 10 As shown, the converged session management network element may include: The third receiving unit 1001 is configured to receive converged service creation requests sent by the session management network element; Unit 1002 is configured to determine whether a fusion service creation request belongs to an AI service. The third sending unit 1003 is configured to send an AI service creation request to the AI ​​policy control network element; The fourth receiving unit 1004 is configured to receive the AI ​​service creation response from the AI ​​policy control network element; The fourth sending unit 1005 is configured to send a converged service creation response back to the session management network element.

[0150] In some embodiments, the determining unit 1002 is further configured to parse the fusion service creation request to obtain an identifier; determine the service type of the fusion service creation request based on the identifier; and determine that the fusion service creation request belongs to an AI service based on the service type.

[0151] In one embodiment of this disclosure, the AI ​​service creation request carries the Quality of Service (QoS) requirement information of the AI ​​service, which includes an identifier, latency requirement information, and accuracy requirement information.

[0152] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0153] The following reference Figure 11 To describe an electronic device 1100 according to such an embodiment of the present disclosure. Figure 11 The electronic device 1100 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0154] like Figure 11As shown, the electronic device 1100 is presented in the form of a general-purpose computing device. The components of the electronic device 1100 may include, but are not limited to: at least one processor 1110, at least one memory 1120, and a bus 1130 connecting different system components (including memory 1120 and processor 1110).

[0155] The memory stores program code that can be executed by the processor 1110, causing the processor 1110 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processor 1110 can perform the following steps of the above method embodiments: receiving an AI service creation request sent by a converged service management network element; querying the AI ​​service required by the AI ​​service creation request and determining the computing power resources required to run the AI ​​service; sending a computing power resource creation request to a computing power policy control network element, wherein the computing power resource creation request carries computing power resources; receiving a computing power resource creation response sent by the computing power policy control network element; and sending an AI service creation response to the converged service management network element.

[0156] For example, processor 1110 may execute the following steps in the above method embodiment: receiving a converged service creation request sent by the session management network element; determining that the converged service creation request belongs to an AI service; sending an AI service creation request to the AI ​​policy control network element; receiving an AI service creation response from the AI ​​policy control network element; and sending a converged service creation response back to the session management network element.

[0157] The memory 1120 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 11201 and / or cache memory 11202, and may further include read-only memory (ROM) 11203.

[0158] The memory 1120 may also include a program / utility 11204 having a set (at least one) of program modules 11205, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0159] Bus 1130 can represent one or more of several types of bus structures, including a memory bus or memory controller, peripheral bus, graphics acceleration port, processor, or a local bus using any of the various bus structures.

[0160] Electronic device 1100 can also communicate with one or more external devices 1140 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more service request processing devices that enable user interaction with electronic device 1100, and / or with any device that enables electronic device 1100 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 1150. Furthermore, electronic device 1100 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 1160. As shown, network adapter 1160 communicates with other modules of electronic device 1100 via bus 1130. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 1100, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0161] In the disclosed exemplary embodiments, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium.

[0162] In some possible implementations, various aspects of this disclosure may also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the foregoing “Detailed Description” section of this specification according to various exemplary embodiments of this disclosure.

[0163] More specific examples of computer-readable storage media in this disclosure may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0164] In this disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting a program for use by or in connection with an instruction execution system, apparatus, or device.

[0165] Optionally, the program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0166] In practical implementation, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on a terminal device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0167] This disclosure provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform a service request processing method provided in various optional embodiments of this disclosure.

[0168] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0169] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0170] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0171] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope of this disclosure is indicated by the appended claims.

Claims

1. A business request processing method, characterized in that, Applications to AI policy control network elements include: Receive AI service creation requests sent by the converged service management network element; Query the AI ​​services required for the AI ​​service creation request, and determine the computing resources required to run the AI ​​service; Send a computing resource creation request to the computing power policy control network element, wherein the computing resource creation request carries the computing resource; Receive the computing resource creation response sent by the computing power strategy control network element; Send an AI service creation response to the converged service management network element.

2. The method according to claim 1, characterized in that, Before receiving the computing resource creation response sent by the computing power policy control network element, the method further includes: Receive an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation request carries the identifier and name of the AI ​​service; The system receives an AI resource creation request sent by the computing power policy control network element, wherein the AI ​​resource creation response carries the download address, identifier, and name of the AI ​​service.

3. The method according to claim 1, characterized in that, The method further includes: Receive an AI service registration request sent by the AI ​​service, wherein the AI ​​service registration request carries the configuration file of the AI ​​service; Store the configuration file and determine that the AI ​​service is available; Send an AI service registration response to the AI ​​service.

4. The method according to claim 3, characterized in that, The configuration file includes at least one of the following: basic service information, service model information, computing resources, and input / output information; The basic information of the service includes at least one of the following: the identifier, name, version number, provider, and download address of the AI ​​service; The service model information includes at least one of the following: the name and performance metrics of the model, and the dataset corresponding to the performance metrics; The computing resources include at least one of the following: CPU information, GPU information, and memory information required to run the AI ​​service.

5. The method according to claim 3, characterized in that, The method further includes: Receive an AI service update request sent by the AI ​​service, wherein the AI ​​service update request carries the latest configuration file of the AI ​​service; Store and update the configuration file; Send an AI service update response to the AI ​​service.

6. The method according to claim 3, characterized in that, The method further includes: Receive the AI ​​service cancellation request sent by the AI ​​service, wherein the AI ​​service cancellation request; If the AI ​​service is determined to be unavailable, delete the configuration file. Send an AI service deregistration response to the AI ​​service.

7. A business request processing method, characterized in that, Applied to converged service management network elements, including: Receive converged service creation request sent by the session management network element; It has been determined that the fusion service creation request belongs to the AI ​​service; Send an AI service creation request to the AI ​​policy control network element; Receive the AI ​​service creation response from the AI ​​policy control network element; The session management network element sends a response to the creation of the converged service.

8. The method according to claim 7, characterized in that, The determination that the fusion service creation request belongs to artificial intelligence (AI) service includes: Parse the fusion service creation request to obtain the identifier; The service type of the fusion service creation request is determined based on the identifier; Based on the service type, it is determined that the fusion service creation request belongs to the AI ​​service.

9. The method according to claim 7, characterized in that, The AI ​​service creation request carries the Quality of Service (QoS) requirement information of the AI ​​service, which includes an identifier, latency requirement information, and accuracy requirement information.

10. An AI strategy control network element, characterized in that, include: The first receiving unit is configured to receive AI service creation requests sent by the converged service management network element; The query unit is configured to query the AI ​​service required for the AI ​​service creation request and determine the computing resources required to run the AI ​​service. The first sending unit is configured to send a computing resource creation request to the computing power policy control network element, wherein the computing resource creation request carries the computing resource; The second receiving unit is configured to receive the computing resource creation response sent by the computing power policy control network element; The second sending unit is configured to send an AI service creation response to the converged service management network element.

11. A converged service management network element, characterized in that, include: The third receiving unit is configured to receive converged service creation requests sent by the session management network element; The determining unit is configured to determine that the fusion service creation request belongs to the AI ​​service; The third sending unit is configured to send an AI service creation request to the AI ​​policy control network element; The fourth receiving unit is configured to receive the AI ​​service creation response fed back by the AI ​​policy control network element; The fourth sending unit is configured to send a fusion service creation response back to the session management network element.

12. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1-6 or 7-9 by executing the executable instructions.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-6 or 7-9.

14. A computer program product comprising computer instructions stored in a computer-readable storage medium, wherein the computer instructions, when executed by a processor, implement the operation instructions of the method according to any one of claims 1-6 or 7-9.

Images