Computation migration method and related apparatus

Abstract

Disclosed in the embodiments of the present application are a computation migration method and a related apparatus. The method comprises: a first access network device receiving a first request from a second access network device, wherein the first request indicates that a terminal device requests to switch from the second access network device to the first access network device, and the first request comprises contextual information of a computation task that the terminal device requests to execute; and in response to the first request, the first access network device sending a second request to a first computation node, wherein the second request is used for indicating that an access network device related to the computation task is changed to the first access network device. Therefore, it is ensured that a computation task that needs to be executed by a terminal device can be successfully executed in a scenario where switching of the terminal device is performed, thereby ensuring that the computation task that the terminal device requests to execute is not interrupted, and thus improving the user experience.

Description

A computing migration method and related apparatus

[0001] The present application claims priority from the Chinese patent application No. CN202411005104.1 filed on July 24, 2024, and entitled "A computing migration method and related apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of communication technology, and in particular to a computing migration method and related apparatus. BACKGROUND

[0003] Cloud computing platforms provide computing services, and terminal devices can send data related to a computing task to a cloud computing platform, and the cloud computing platform executes the computing task. However, there is a problem of long latency when the cloud computing platform executes the computing task. Therefore, edge computing is developed on the basis of cloud computing. Edge computing refers to dispersing computing tasks to edge computing nodes (referred to as computing nodes) for execution. Since the computing nodes are closer to users (such as terminal devices), edge computing has lower latency.

[0004] Since the access network devices are closer to the terminal devices, providing computing services for the terminal devices by the access network devices has the advantage of lower latency, and therefore how the access network devices provide computing services for the terminal devices has become a current research hotspot.

[0005] Terminal device switching between different access network devices is a common communication scenario, for example, a second access network device provides communication services for a terminal device, and after the terminal device switches to a first access network device, the first access network device provides communication services for the terminal device. However, how to migrate computing tasks between different access network devices has become a current problem to be solved. SUMMARY

[0006] Embodiments of the present application propose a computing migration method and related apparatus, which ensure that computing tasks that a terminal device needs to execute can be successfully executed in the scenario of terminal device switching, guarantee that computing tasks requested by the terminal device to execute are not interrupted, and improve user experience.

[0007] In a first aspect, an embodiment of the present application provides a computing migration method, characterized by the method being applied to a first access network device. For example, the method is executed by the first access network device. The first access network device can be a device or apparatus with a chip, or a device or apparatus integrated with a circuit, or a chip, a chip system, a module, a control unit, a circuit, or a processor in the foregoing devices or apparatus, or a centralized unit (CU) and / or a distributed unit (DU). It should be noted that, in the present application, the first access network device can refer to the access network device itself, or a chip, a functional module, an integrated circuit, or a processor in the first access network device that completes the method provided by the present application, or a CU and / or a DU included in the first access network device. For example, the method executed by the first access network device is executed by the CU of the first access network device. In the first aspect and possible implementation manners thereof, the method is described by taking the method executed by the first access network device as an example.

[0008] The method comprises: receiving, by the first access network device, a first request from a second access network device, the first request indicating that a terminal device requests to switch from the second access network device to the first access network device, and the first request comprising context information of a computing task requested to be executed by the terminal device; and in response to the first request, sending, by the first access network device, a second request to a first computing node, the second request being used to indicate that an access network device related to the computing task is changed to the first access network device.

[0009] In the embodiments of the present application, the access network device that provides computing services and communication services for the terminal device before switching is referred to as the second access network device, and the access network device that provides computing services and communication services for the terminal device after switching is referred to as the first access network device.

[0010] Optionally, the second request is used to request to establish a transmission path between the first access network device and the first computing node, and the transmission path is used to transmit related data of the computing task. Correspondingly, the first computing node establishes the transmission path with the first access network device in response to the second request.

[0011] In another possible implementation manner, the second request is used to indicate that a transmission path between the second access network device and the first computing node related to the terminal device is switched to the first access network device. Correspondingly, the first computing node changes the transmission path between the second access network device and the first computing node related to the terminal device in response to the second request. Before the change, two ends of the transmission path are the second access network device and the first computing node respectively; and after the change, two ends of the transmission path are the first access network device and the first computing node respectively.

[0012] In another possible implementation manner, the second request is used to inform the terminal device to switch (or move) to the first access network device.

[0013] In another possible implementation manner, the second request is used to inform a location update of the terminal device.

[0014] It should be noted that the computing task in the embodiments of the present application includes but is not limited to: a video decoding task, an image processing task, an image recognition task, a speech recognition task, a machine translation task, a gesture recognition task, an unmanned driving task, a route planning task, a neural network model inference task, or a neural network model training task, and the like.

[0015] In the above technical solution, in the scenario that the terminal device switches from the second access network device to the first access network device, the first access network device sends the second request to the first computing node according to the first request sent by the second access network device, the first request includes context information of a computing task requested to be executed by the terminal device, and the second request is used to request the first computing node to execute the computing task. So as to ensure that the computing task requested to be executed by the terminal device can be smoothly migrated to the first computing node in the process of moving of the terminal device, so that the terminal device is not affected, and the user experience is improved.

[0016] With reference to the first aspect, in a possible implementation manner of the first aspect, the second request includes any one or more of the following information: context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device.

[0017] The routing information is used to indicate a transmission path of the computing task related data or a computing session. For example, the transmission path is used to carry the data related to the computing task, downlink packets of the computing task, or computing results of the computing task. The transmission path can be understood as a point-to-point connection between the base station and the computing node, and the transmission path is used to transmit user equipment (UE) related data (or terminal device related data). Alternatively, the routing information is used to identify a point-to-point connection or a transmission path between the terminal device and the computing node. Alternatively, the routing information is used to identify a session for computing data transmission between the terminal device and the computing node. The transmission path includes but is not limited to: a connection, a path, a tunnel, or a link.

[0018] Exemplarily, the routing information includes identification information of the transmission path. At this time, the identification information is used to identify the transmission path between the terminal device and the second computing node.

[0019] For example, when a protocol stack between the base station and the computing node is based on a general packet radio service tunneling protocol-user plane (GTP-U), the identification information of the transmission path can be a tunnel endpoint identifier (TEID).

[0020] The address information of the terminal device can include an internet protocol (IP) address of the terminal device and / or a port number of the terminal device. The address information of the terminal device can be address information allocated to the terminal device by an operator or a network side, or address information allocated to the terminal device by an access network device, which is not limited in the present application.

[0021] For another example, the identification information of the terminal device is an international mobile equipment identity (IMEI) of the terminal device or a media access control (MAC) address of the terminal device.

[0022] With reference to the first aspect, in a possible implementation manner of the first aspect, in response to the first request, the second request is sent to the first computing node, including: in response to the first request, determining the first computing node; sending the second request to the first computing node; receiving a second response from the first computing node, the second response indicating that the first computing node allows the execution of the computing task requested by the terminal device to be executed; and in response to the second response, establishing a transmission path related to the terminal device with the first computing node.

[0023] In the above technical solution, the first access network device can also determine the first computing node by itself, which improves the implementation flexibility of the solution.

[0024] With reference to the first aspect, in a possible implementation manner of the first aspect, the method further includes: sending second indication information to the second access network device, the second indication information being used to indicate that a second computing node migrates a computing task to the first computing node, the second computing node being an original computing node for executing the computing task; and receiving data related to the computing task from the second access network device.

[0025] The first access network device determines the first computing node by itself, and therefore, the first access network device has established a transmission path with the first computing node. The first access network device needs to send the second indication information to the second access network device, which forwards the second indication information to the second computing node. The second computing node sends data related to the computing task to the first computing node according to the indication, so as to realize the migration of the computing task.

[0026] With reference to the first aspect, in a possible implementation manner of the first aspect, in response to the first request, the second request is sent to the first computing node, including: sending a third request to the core network device, the third request carrying context information of the computing task requested by the terminal device to execute, so that the core network device sends the second request to the first computing node, the second computing node being an original computing node for executing the computing task; receiving a third response from the core network device, the third response carrying second indication information, the second indication information being used to indicate that the second computing node migrates the computing task to the first computing node, the second computing node being the original computing node for executing the computing task; and in response to the third response, establishing a transmission path related to the terminal device with the first computing node.

[0027] In the above technical solution, the core network device can also determine the first computing node, which improves the implementation flexibility of the solution.

[0028] With reference to the first aspect, in a possible implementation manner of the first aspect, the second indication information includes:

[0029] routing information, context information of the computing task, and / or related information of the first computing node, the routing information indicating a transmission path carrying the computing task.

[0030] With reference to the first aspect, in a possible implementation manner of the first aspect, the context information of the computing task includes any one or more of the following information: identification information of the computing task, identification information of an application related to the computing task, routing information, identification information of a computing node supporting execution of the computing task, or first indication information indicating that the computing task needs to be switched to the first access network device for continuous execution.

[0031] For example, the identification information of the computing task includes a task identifier (task ID). The identification information of the computing task can also indicate the type of the computing task, such as “picture coding”, “video coding”, “AI model training”, “path planning”, or “automatic driving”. The target access network device directly or indirectly determines other information of the computing task according to the identification information of the computing task, for example, the target access network device determines the QoS parameter of the computing task corresponding to the task ID according to the task ID, so as to save communication overhead.

[0032] For example, the application is an application related to a computing task, such as a video application, a navigation application, an automatic driving application, or a communication application.

[0033] For example, the identification information of the computing node supporting the execution of the computing task indicates which computing nodes support the execution of the computing task requested by the terminal device. For example, the second access network device is associated with a computing node 1 and a computing node 2, and the second access network device determines that the computing node 1 supports the execution of the computing service requested by the terminal device. The second request can carry the identification information of the computing node 1. In this way, the first access network device can detect whether the first access network device and the computing node 1 have an association relationship, and further determine whether the computing node 1 can provide the computing service to the terminal device. For example, the identification information of the computing node can be a tunnel endpoint identifier (TEID) of the computing node, an IP address of the computing node, or an identifier of the computing node.

[0034] In the above technical solution, the context information of the computing task can include various information to ensure that the computing task can normally run after migration.

[0035] In combination with the first aspect, in a possible implementation manner of the first aspect, the context information of the computing task further includes any one or more of the following: identification information of a neural network model related to the computing task, type information of the neural network model, parameter information of the neural network model, KV Cache information of the neural network model, information of one or more intermediate layers included in the neural network model, initial input feature information of the neural network model, or output feature information of one or more intermediate layers of the neural network model.

[0036] In the above technical solution, when the computing task is a neural network model related computing task (for example, a training task), the context information of the computing task can further include neural network model related information. In this way, after the terminal device switches to the first access network device, the first access network device can continue to execute the computing task according to the context information, thereby ensuring the smooth execution of the computing task. When the computing task is a training task of a neural network model, the training effect of the neural network model can be improved.

[0037] In a second aspect, an embodiment of the present application provides a method for computing migration, which is applied to a first computing node. For example, the method is executed by the first computing node, which can be a device or apparatus with a chip, or a device or apparatus integrated with a circuit, or a chip, a chip system, a module, a control unit, a circuit, or a processor in the foregoing devices or apparatus. It should be noted that, in the present application, when referring to the first computing node, it can refer to the computing node itself, or a chip, a functional module, an integrated circuit, or a processor in the first computing node that completes the method provided by the present application. In the second aspect and possible implementation manners thereof, the method is taken as an example executed by the first computing node.

[0038] The method comprises: receiving a second request, the second request being used to indicate that a terminal device requesting to execute a computing task is switched from a second access network device to a first access network device; and in response to the second request, executing the computing task.

[0039] It should be noted that the computing task in the embodiments of the present application includes but is not limited to a video decoding task, an image processing task, an image recognition task, a speech recognition task, a machine translation task, a gesture recognition task, an unmanned driving task, a route planning task, a neural network model inference task, or a neural network model training task, and the like.

[0040] Optionally, the second request is used to request to establish a transmission path between the first access network device and the first computing node, the transmission path being used to transmit related data of the computing task. Correspondingly, the first computing node establishes the transmission path with the first access network device in response to the second request.

[0041] In another possible implementation manner, the second request is used to indicate that a transmission path between the second access network device and the first computing node with respect to the terminal device is switched to the first access network device. Correspondingly, the first computing node changes the transmission path between the second access network device and the first computing node with respect to the terminal device in response to the second request. Before the change, two ends of the transmission path are the second access network device and the first computing node respectively; after the change, two ends of the transmission path are the first access network device and the first computing node respectively.

[0042] In another possible implementation manner, the second request is used to notify the terminal device to switch (or move) to the first access network device.

[0043] In another possible implementation manner, the second request is used to notify a location update of the terminal device.

[0044] In the technical solution, in a scenario where the terminal device switches from the second access network device to the first access network device, the first access network device sends a second request to the first computing node according to a first request sent by the second access network device, the first request including context information of a computing task requested to be executed by the terminal device, and the second request being used to indicate that the terminal device requesting to execute the computing task switches from the second access network device to the first access network device. The first computing node establishes a transmission path with the first access network device about the terminal device according to the second request. This ensures that the computing task requested to be executed by the terminal device can be smoothly migrated to the first computing node through the transmission path in the process of migration of the terminal device, and the first computing node can continue to execute the computing task, so that the computing task is not interrupted in the process of migration of the terminal device, and user experience is improved.

[0045] In a possible implementation of the second aspect, the second request includes any one or more of the following information: the context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device.

[0046] The routing information is used to indicate a transmission path of computing task related data or a computing session. For example, the transmission path is used to carry data related to the computing task, downlink packets of the computing task, or computing results of the computing task. The transmission path can be understood as a point-to-point connection between the base station and the computing node, and the transmission path is used to transmit UE related data. Alternatively, the routing information is used to identify a point-to-point connection or a transmission path between the terminal device and the computing node. Alternatively, the routing information is used to identify a session for computing data transmission between the terminal device and the computing node. The transmission path includes but is not limited to a connection, a path, a tunnel, or a link.

[0047] For example, the routing information includes identification information of the transmission path. At this time, the identification information is used to identify the transmission path between the terminal device and the second computing node.

[0048] For example, when a protocol stack between the base station and the computing node is based on a packet radio service user plane tunneling protocol (GTP-U), the identification information of the transmission path can be a tunnel endpoint identification (TEID).

[0049] The address information of the terminal device can include an Internet Protocol (IP) address of the terminal device and / or a port number of the terminal device. The address information of the terminal device can be address information allocated to the terminal device by an operator or a network side, or address information allocated to the terminal device by an access network device, and the application does not limit this.

[0050] For example, the identification information of the terminal device is an international mobile equipment identity (IMEI) of the terminal device or a media access control (MAC) address of the terminal device.

[0051] For example, the routing information includes level identification information (UE level ID) of the terminal device. The level identification information of the terminal device is used to indicate a tunnel between the access network device and the computing node with respect to the terminal device. For example, different levels of tunnels can be used for different terminal devices.

[0052] The address information of the first access network device can also be user plane address information of the first access network device.

[0053] The address information of the terminal device can include an internet protocol (IP) address of the terminal device and / or a port number of the terminal device. The address information of the terminal device can be address information allocated to the terminal device by an operator or a network side, or address information allocated to the terminal device by the access network device, which is not limited in the present application.

[0054] With reference to the second aspect, in a possible implementation manner of the second aspect, the transmission path with respect to the terminal device is established with the first access network device according to the second request, including: sending a second response to the first access network device in response to the second request, the second response indicating that the first computing node allows the computing task requested by the terminal device to be executed; and establishing a transmission path related to the terminal device with the first computing node.

[0055] With reference to the second aspect, in a possible implementation manner of the second aspect, the data related to the computing task is obtained, including: receiving the data related to the computing task from the first access network device; or receiving the data related to the computing task from a second computing node, the second computing node being an original computing node for executing the computing task.

[0056] In the above technical solution, the first computing node can obtain the data related to the computing task from multiple channels, thereby improving the implementation flexibility of the solution.

[0057] With reference to the second aspect, in a possible implementation manner of the second aspect, the second request is received, including: receiving the second request from the first access network device; or receiving the second request from a core network device, wherein the core network device sends the second request to the first computing node according to a third request sent by the first access network device, and the third request carries context information of the computing task requested by the terminal device to be executed.

[0058] In the above technical solution, the first computing node can execute the computing task in response to the request of the first access network device, and the first computing node can also execute the computing task in response to the request of the core network device, thereby improving the implementation flexibility of the solution.

[0059] With reference to the second aspect, in a possible implementation form of the second aspect, the context information of the computing task comprises any one or more of the following: identification information of the computing task, identification information of an application, the application being related to the computing task, routing information, identification information of a computing node supporting execution of the computing task, or first indication information, the first indication information indicating that the computing task needs to be switched to the first access network device for continuous execution.

[0060] For example, the identification information of the computing task comprises a task identifier (task ID). The identification information of the computing task can also indicate a type of the computing task, such as “picture coding”, “video coding”, “AI model training”, “path planning”, or “automatic driving”, etc. The target access network device directly or indirectly determines other information of the computing task according to the identification information of the computing task, for example, the target access network device determines the computing task application information and / or the QoS parameter of the computing task according to the task ID, so as to save communication overhead.

[0061] For example, the identification information of the application is the application related to the computing task, such as a video application, a navigation application, an automatic driving application, or a communication application, etc. For example, the identification information of the application can be an application identifier (APP ID), a fully qualified domain name (FQDN) of the application, or an application name, etc.

[0062] For example, the identification information of the computing node supporting execution of the computing task indicates which computing nodes support execution of the computing task requested by the terminal device. For example, the second access network device is associated with a computing node 1 and a computing node 2, and the second access network device determines that the computing node 1 supports execution of the computing service requested by the terminal device, and the identification information of the computing node 1 can be carried in the second request. So that the first access network device detects whether the first access network device and the computing node 1 have an association relationship, and further determines whether the computing node 1 can be used to provide the computing service to the terminal device. For example, the identification information of the computing node can be a tunnel endpoint identifier (TEID) of the computing node, an IP address of the computing node, or an identifier of the computing node.

[0063] With reference to the second aspect, in a possible implementation form of the second aspect, the context information of the computing task further includes any one or more of the following: identification information of a neural network model related to the computing task, type information of the neural network model, parameter information of the neural network model, KV Cache information of the neural network model, information of one or more intermediate layers included in the neural network model, initial input feature information of the neural network model, or output feature information of one or more intermediate layers of the neural network model.

[0064] In the above technical solution, when the computing task is a neural network model related computing task (for example, a training task), the context information of the computing task can further include neural network model related information. After the terminal device switches to the first access network device, the first access network device can continue to perform the computing task according to the context information, thereby ensuring smooth execution of the computing task. When the computing task is a training task of a neural network model, the training effect of the neural network model can be improved.

[0065] In the third aspect, an embodiment of the present application provides a computing migration method. The method is applied to a second access network device. For example, the method is executed by the second access network device. The second access network device can be a device or apparatus with a chip, or a device or apparatus integrated with a circuit, or a chip, a chip system, a module, a control unit, a circuit, or a processor in the foregoing devices or apparatus, or a centralized unit (CU) and / or a distributed unit (DU). It should be noted that, in the present application, the second access network device can refer to the access network device itself, or a chip, a functional module, an integrated circuit, or a processor in the second access network device that completes the method provided by the present application, or a CU and / or a DU included in the second access network device. The method executed by the second access network device is executed by a CU of the second access network device. In the third aspect and possible implementation forms thereof, the method executed by the second access network device is taken as an example for description. The method includes the following steps.

[0066] receiving a second message from the first access network device, the second message carrying address information of the first access network device;

[0067] According to the second message, forwarding data of a computing task requested by the terminal device to execute to the first access network device.

[0068] In a possible implementation, in response to the first message, the first access network device feeds back address information of the first access network device to the second access network device. The first access network device sends a second message to the second access network device, where the second message includes the address information of the first access network device. The first message is carried in a response to a handover request, and after the first access network device determines that the terminal device needs to be handed over to the first access network device, and the terminal device requests the first access network device to perform a computing task, the first access network device sends a handover request to the second access network device (an anchor access network device) to obtain related information of the computing task. The first message is carried in a response to the handover request.

[0069] In another possible implementation, the second message is user plane data, and the user plane data includes the address information of the first access network device.

[0070] Specifically, the first access network device is a serving base station currently serving the terminal device, and the second access network device is an anchor base station. The first access network device can be the first serving base station to which the terminal device is handed over after leaving the anchor base station (the second access network device), or the first access network device can be the xth serving base station to which the terminal device is handed over after leaving the anchor base station (the second access network device), where x is a positive integer greater than 1.

[0071] When the terminal device is handed over to a new serving base station (for example, the first access network device), the serving base station triggers the anchor base station (for example, the second access network device) to report address information of the serving base station. In this way, the anchor base station forwards data related to the computing task of the terminal device to the serving base station (the first access network device), and the serving base station further forwards the data to the terminal device, thereby ensuring that the computing task of the terminal device is not interrupted.

[0072] In the foregoing technical solution, during the process in which the terminal device is handed over from the second access network device to the first access network device, the first access network device can request the first computing node to continue to perform the computing task of the terminal device, thereby ensuring that the computing task requested by the terminal device to be performed is not interrupted during the handover process, ensuring service continuity of the terminal device, and improving user experience.

[0073] With reference to the third aspect, in a possible implementation of the third aspect, the method further includes:

[0074] sending the address information of the first access network device to the first computing node.

[0075] With reference to the third aspect, in a possible implementation form of the third aspect, the method further includes:

[0076] sending, to a computing node management function, address information of the first access network device, the computing node management function being configured to manage the first computing node.

[0077] In the above technical solution, the second access network device can inform the first computing node of the address information of the first access network device in multiple ways, thereby improving the implementation flexibility of the solution.

[0078] With reference to the third aspect, in a possible implementation form of the third aspect, the method further includes:

[0079] sending, to the first access network device, a first message, the first message including any one or more of the following information: routing information, identification information of the first computing node, and / or address information of the terminal device, the routing information indicating a transmission path carrying the computing task.

[0080] With reference to the third aspect, in a possible implementation form of the third aspect, the second access network device serves as an anchor base station of the terminal device, and the first access network device serves as a serving base station of the terminal device.

[0081] The fourth aspect provides a communication apparatus, which is applicable to a first access network device. For example, the communication apparatus can be the first access network device itself, or can be a component (for example, a processor, a circuit, a chip, or a chip system) inside the first access network device. The communication apparatus can include corresponding modules, units, or means for implementing the method in the first aspect and any possible implementation form thereof. The modules, units, or means can be implemented by hardware, software, or by executing corresponding software by hardware. The hardware or software includes one or more modules or units corresponding to the above functions. The apparatus includes a transceiver module and a processing module. The constituent modules of the communication apparatus can also be used to perform the steps performed in each possible implementation form of the first aspect and achieve the corresponding technical effects. For details, refer to the first aspect, which will not be described here again.

[0082] In a fifth aspect, the fifth aspect provides a computing node, which is suitable for the first computing node. For example, the first computing node can be the first computing node itself, or can be a component (for example, a processor, a circuit, a chip, or a chip system, etc.) inside the first computing node. The computing node can include a module, a unit, or a means corresponding to the method in the above-mentioned second aspect and any possible implementation manner thereof, which can be implemented by hardware, software, or by hardware executing corresponding software. The hardware or software includes one or more modules or units corresponding to the above functions. The computing node includes a transceiver module and a processing module, and the constituent modules of the computing node can also be used to perform the steps performed in each possible implementation manner of the second aspect and achieve the corresponding technical effects, which can be referred to the second aspect for details and will not be described here.

[0083] In a sixth aspect, the sixth aspect provides a communication device, which is suitable for the first access network device. For example, the communication device can be the first access network device itself, or can be a component (for example, a processor, a circuit, a chip, or a chip system, etc.) inside the first access network device. The communication device can include a module, a unit, or a means corresponding to the method in the above-mentioned third aspect and any possible implementation manner thereof, which can be implemented by hardware, software, or by hardware executing corresponding software. The hardware or software includes one or more modules or units corresponding to the above functions. The device includes a transceiver module and a processing module, and the constituent modules of the communication device can also be used to perform the steps performed in each possible implementation manner of the third aspect and achieve the corresponding technical effects, which can be referred to the third aspect for details and will not be described here.

[0084] In a seventh aspect, the seventh aspect provides a communication device, which includes at least one processor, the at least one processor is coupled with a memory; the memory is used to store programs or instructions; the at least one processor is used to execute the programs or instructions, so that the device implements the method in any one of the possible implementation manners of any one of the preceding first aspects. Optionally, the communication device can include the memory.

[0085] In an eighth aspect, the eighth aspect provides a communication device, which includes at least one logic circuit and an input and output interface; the logic circuit is used to execute the method in any one of the possible implementation manners of any one of the preceding first aspects.

[0086] In a ninth aspect, the ninth aspect of the present application provides a computing node, comprising at least one processor coupled with a memory; the memory is configured to store a program or an instruction; the at least one processor is configured to execute the program or the instruction, so that the computing node implements the method in any possible implementation manner of any one of the preceding second aspects. Optionally, the computing node can comprise the memory.

[0087] In a tenth aspect, the tenth aspect of the present application provides a computing node, comprising at least one logic circuit and an input / output interface; the logic circuit is configured to execute the method in any possible implementation manner of any one of the preceding second aspects.

[0088] In an eleventh aspect, the eleventh aspect of the present application provides a communication apparatus, comprising at least one processor coupled with a memory; the memory is configured to store a program or an instruction; the at least one processor is configured to execute the program or the instruction, so that the apparatus implements the method in any possible implementation manner of any one of the preceding third aspects. Optionally, the communication apparatus can comprise the memory.

[0089] In a twelfth aspect, the twelfth aspect of the present application provides a communication apparatus, comprising at least one logic circuit and an input / output interface; the logic circuit is configured to execute the method in any possible implementation manner of any one of the preceding third aspects.

[0090] In a thirteenth aspect, the thirteenth aspect of the present application provides a communication system, comprising the first access network device and / or the first computing node.

[0091] With reference to the thirteenth aspect, in a possible implementation manner of the thirteenth aspect, the communication system, the communication system comprises the communication apparatus of the seventh aspect and the computing node of the ninth aspect.

[0092] In a fourteenth aspect, the fourteenth aspect of the present application provides a communication system, comprising the second access network device.

[0093] With reference to the fourteenth aspect, in a possible implementation manner of the fourteenth aspect, the communication system, the communication system comprises the communication apparatus of the eleventh aspect.

[0094] In a fifteenth aspect, the fifteenth aspect of the present application provides a computer readable storage medium, the storage medium is configured to store one or more computer execution instructions, when the computer execution instructions are executed by a processor, the processor executes the method in any possible implementation manner of any one of the preceding first aspect, second aspect and / or third aspect.

[0095] In a sixteenth aspect, the sixteenth aspect provides a computer program product (or computer program), when a computer program in the computer program product is executed by the processor, the processor executes the method in any possible implementation manner of any one of the first aspect, the second aspect and / or the third aspect.

[0096] In a seventeenth aspect, the seventeenth aspect provides a chip or chip system, the chip or chip system comprising at least one processor for supporting a communication apparatus to implement the method in any possible implementation manner of any one of the first aspect, the second aspect and / or the third aspect. For example, the chip can be a baseband chip, a modem chip, a system on chip (SoC) chip containing a modem core, a system in package (SIP) chip, or a communication module, etc.

[0097] In a possible design, the chip or chip system can further comprise a memory, the memory being configured to store necessary program instructions and data of the communication apparatus. The chip system can be composed of a chip, or can contain a chip and other discrete devices. Optionally, the chip system further comprises an interface circuit, the interface circuit being configured to provide program instructions and / or data for the at least one processor.

[0098] The technical effects brought by any design manner in the fourth aspect to the seventeenth aspect can be referred to the technical effects brought by different design manners in the first aspect to the third aspect, and will not be described herein again. BRIEF DESCRIPTION OF DRAWINGS

[0099] FIGS. 1a to 1e are schematic diagrams of AI processing processes involved in the present application;

[0100] FIG. 2 is a schematic diagram of a communication system in an embodiment of the present application;

[0101] FIG. 3a is a schematic diagram of another communication system in an embodiment of the present application;

[0102] FIG. 3b is a schematic diagram of another communication system in an embodiment of the present application;

[0103] FIG. 4 is a schematic diagram of another communication system in an embodiment of the present application;

[0104] FIG. 5 is a schematic diagram of another communication system in an embodiment of the present application;

[0105] FIG. 6 is a schematic diagram of an application architecture involving a RIC module under an ORAN architecture;

[0106] FIG. 7 is a schematic diagram of a structure of an access network device in an embodiment of the present application;

[0107] FIG. 8 is a schematic diagram of a communication scenario involved in an embodiment of the present application;

[0108] FIG. 9 is a schematic diagram of an embodiment of a method for computing handover in an embodiment of the present application;

[0109] FIG. 10 is a schematic diagram of another communication system structure involved in an embodiment of the present application;

[0110] FIG. 11 is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0111] FIG. 12a is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0112] FIG. 12b is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0113] FIG. 13 is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0114] FIG. 14 is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0115] FIG. 15 is a schematic diagram of another embodiment of a method for computing migration in an embodiment of the present application;

[0116] FIG. 16 is a schematic diagram of a structure of a communication apparatus in an embodiment of the present application;

[0117] FIG. 17 is a schematic diagram of another structure of a communication apparatus in an embodiment of the present application;

[0118] FIG. 18 is a schematic diagram of another structure of a communication apparatus in an embodiment of the present application;

[0119] FIG. 19 is a schematic diagram of a structure of a computing node provided in an embodiment of the present application;

[0120] FIG. 20 is a schematic diagram of a structure of a computing node 2000 provided in an embodiment of the present application;

[0121] FIG. 21 is a schematic diagram of a structure of a computing node cluster provided in an embodiment of the present application;

[0122] FIG. 22 is a schematic diagram of another structure of a computing node cluster provided in an embodiment of the present application;

[0123] FIG. 23 is a schematic diagram of a structure of a computer readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0124] The terms "first", "second", "third", "fourth", and the like in the description and in the claims of the present application, and above-described drawings (if there are) are used to distinguish similar objects, and do not necessarily have to be used to describe a particular order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0125] It should be understood that the term "and / or" herein only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the following three cases: A exists alone, A and B exist together, B exists alone, where A and B can be single or multiple. In addition, the character " / " herein generally represents an "or" relationship between the associated objects. In addition, "at least one of the following" or similar expressions herein are used to represent any combination of the listed items; for example, at least one of A, B and (or) C can represent the following six cases: A exists alone, B exists alone, C exists alone, A and B exist together, B and C exist together, A and C exist together, A, B and C exist together, where A, B and C can be single or multiple.

[0126] First, some of the terms in the embodiments of the present application are explained and described, so as to facilitate the understanding of those skilled in the art.

[0127] (1) Configuration and pre-configuration: in the present application, configuration and pre-configuration will be used together. Configuration means that the network device sends some parameter configuration information or parameter values to the terminal device through messages or signaling, so that the terminal device determines the communication parameters or resource at the time of transmission according to these values or information. Pre-configuration is similar to configuration, which can be parameter information or parameter values agreed by the network device and the terminal device in advance, or parameter information or parameter values adopted by the network device or the terminal device according to the standard protocol, or parameter information or parameter values pre-stored in the network device or the terminal device. The present application does not limit this.

[0128] Further, these values and parameters can be changed or updated.

[0129] (2) The terms "system" and "network" in the embodiments of the present application can be used interchangeably. "At least one" refers to one or more, and "multiple" refers to two or more. The "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the cases of A alone, A and B together, and B alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after it. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or multiple items. For example, "at least one of A, B and C" includes A, B, C, AB, AC, BC or ABC. In addition, unless otherwise specified, the ordinal numbers "first", "second", etc. mentioned in the embodiments of the present application are used to distinguish a plurality of objects, and are not used to limit the order, time sequence, priority or importance of the plurality of objects.

[0130] (3) In the embodiments of the present application, "sending" and "receiving" represent the direction of signal transmission. For example, "sending information to XX" can be understood as that the destination of the information is XX, which can include direct sending through the air interface, or indirect sending through the air interface by other units or modules. "Receiving information from YY" can be understood as that the source of the information is YY, which can include direct receiving from YY through the air interface, or indirect receiving from YY through the air interface by other units or modules. "Sending" can also be understood as "output" of a chip interface, and "receiving" can also be understood as "input" of a chip interface.

[0131] In other words, sending and receiving can be between devices, such as between network devices and terminal devices, or within devices, such as between components, modules, chips, software modules or hardware modules within a device through a bus, wire or interface.

[0132] It can be understood that the information may be processed as necessary between the source and the destination of the information transmission, such as encoding and modulation, but the destination can understand the valid information from the source. Similar expressions in the present application can be similarly understood, and will not be repeated here.

[0133] (4) Artificial intelligence (AI).

[0134] AI can enable machines to have human intelligence, for example, enabling machines to apply computer software and hardware to simulate certain intelligent behaviors of humans. To achieve artificial intelligence, a machine learning method can be employed. In the machine learning method, a machine learns (or trains) a neural network model using training data. The neural network model can also be referred to as an AI model, an AI large model, a large language model (LLM), or a model. The model represents a mapping between an input and an output. The learned model can be used for inference (or prediction), i.e., the model can be used to predict an output corresponding to a given input. The output can also be referred to as an inference result (or a prediction result).

[0135] Machine learning can include supervised learning, unsupervised learning, and reinforcement learning. The unsupervised learning can also be referred to as non-supervised learning.

[0136] Supervised learning learns a mapping relationship from sample values to sample labels using a machine learning algorithm according to the sample values and the sample labels that have been collected, and uses an AI model to express the learned mapping relationship. The process of training a machine learning model is a process of learning such a mapping relationship. In the training process, the sample values are input into the model to obtain predicted values of the model, and the model parameters are optimized by calculating the error between the predicted values of the model and the sample labels (ideal values). After the mapping relationship is learned, the learned mapping can be used to predict new sample labels. The mapping relationship learned by supervised learning can include linear mapping or non-linear mapping. According to the type of label, the learned task can be divided into classification tasks and regression tasks.

[0137] Unsupervised learning uses algorithms to discover the internal patterns of samples according to the collected sample values. In unsupervised learning, there is a class of algorithms that use the samples themselves as a supervision signal, i.e., the model learns a mapping relationship from the samples to the samples, which is called self-supervised learning. In the training process, the model parameters are optimized by calculating the error between the predicted values of the model and the samples themselves. Self-supervised learning can be used for signal compression and decompression recovery applications, and common algorithms include autoencoders and generative adversarial networks.

[0138] Reinforcement learning, unlike supervised learning, is a type of algorithm that learns problem-solving strategies through interaction with its environment. Unlike supervised and unsupervised learning, reinforcement learning problems do not have explicit "correct" action labels. The algorithm needs to interact with the environment, obtain reward signals from the environment, and then adjust its decision actions to obtain a larger reward signal value. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user based on the total system throughput feedback from the wireless network, aiming to achieve a higher system throughput. The goal of reinforcement learning is to find the decision action that maximizes the cumulative reward over a relatively long period. Training in reinforcement learning is achieved through iterative interaction with the environment.

[0139] Neural networks (NNs) are a specific model in machine learning techniques. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Traditional communication systems rely on extensive expert knowledge to design communication modules, while deep learning communication systems based on neural networks can automatically discover hidden pattern structures from large datasets, establish mapping relationships between data, and achieve performance superior to traditional modeling methods.

[0140] The idea behind neural networks comes from the neuronal structure of the brain. For example, each neuron performs a weighted summation of its input values ​​and outputs the result through an activation function.

[0141] Figure 1a shows a schematic diagram of a neuron structure. Assume the input to the neuron is x = [x0, x1, ..., x...]. n The weights corresponding to each input are w = [w, w1, ..., w2]. n ], where n is a positive integer, w i and x i It can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. i As x i The weights are used to assign weights to x. i Weighting is applied. The bias for the weighted summation of the input values ​​is, for example, b. Activation functions can take many forms. Assuming a neuron's activation function is y = f(z) = max(0, z), then the neuron's output is: For example, if the activation function of a neuron is y = f(z) = z, then the output of that neuron is: Here, b can be any possible type, such as a decimal, an integer (e.g., 0, a positive integer, or a negative integer), or a complex number. The activation functions of different neurons in a neural network can be the same or different.

[0142] In addition, a neural network generally includes multiple layers, and each layer can include one or more neurons. By increasing the depth and / or width of the neural network, the expressiveness of the neural network can be improved, providing a more powerful information extraction and abstract modeling capability for complex systems. The depth of the neural network can refer to the number of layers included in the neural network, and the number of neurons included in each layer can be referred to as the width of the layer. In an implementation, the neural network includes an input layer and an output layer. The input layer of the neural network processes the received input information through neurons, and transmits the processing result to the output layer to obtain the output result of the neural network. In another implementation, the neural network includes an input layer, a hidden layer, and an output layer. The input layer of the neural network processes the received input information through neurons, and transmits the processing result to the intermediate hidden layer. The hidden layer calculates the received processing result to obtain a calculation result, and transmits the calculation result to the output layer or the next adjacent hidden layer. Finally, the output layer obtains the output result of the neural network. The neural network can include one hidden layer, or include multiple sequentially connected hidden layers, without limitation.

[0143] The neural network is, for example, a deep neural network (DNN). According to the construction manner of the network, the DNN can include a feedforward neural network (FNN), a convolutional neural network (CNN), and a recurrent neural network (RNN).

[0144] FIG. 1b is a schematic diagram of an FNN network. The FNN network is characterized in that the neurons in adjacent layers are completely connected two by two. This feature makes the FNN usually require a large amount of storage space, resulting in a high computational complexity.

[0145] The CNN is a neural network specially designed to process data with a similar grid structure. For example, time series data (e.g., time axis discrete sampling) and image data (e.g., two-dimensional discrete sampling) can be considered as data with a similar grid structure. The CNN does not use all the input information for operation at one time, but uses a fixed-size window to extract part of the information for convolution operation, which greatly reduces the calculation amount of model parameters. In addition, according to the different types of information extracted by the window (such as people and objects in the same image), each window can use different convolution kernel operations, which enables the CNN to better extract the features of the input data.

[0146] RNN is a kind of DNN network using feedback time series information. The input of RNN includes the new input value at the current time and the output value of itself at the previous time. RNN is suitable for obtaining sequence features with correlation in time, and is particularly suitable for applications such as speech recognition and channel coding and decoding.

[0147] In the above model training process of machine learning, a loss function can be defined. The loss function describes the gap or difference between the output value of the model and the ideal target value. The loss function can be embodied in various forms, and the specific form of the loss function is not limited. The model training process can be regarded as the following process: by adjusting part or all of the parameters of the model, the value of the loss function is less than the threshold value or meets the target requirement.

[0148] The model can also be referred to as an AI model, a rule, or other names. The AI model can be considered as a specific method to realize the AI function. The AI model represents the mapping relationship or function between the input and the output of the model. The AI function can include one or more of the following: data collection, model training (or model learning), model information publishing, model inference (or model reasoning, reasoning, or prediction, etc.), model monitoring or model verification, or reasoning result publishing, etc. The AI function can also be referred to as an AI (related) operation, or an AI related function.

[0149] The implementation process of the fully connected neural network will be described below with reference to the accompanying drawings. The fully connected neural network is also called multilayer perceptron (MLP).

[0150] As shown in FIG. 1c, an MLP includes an input layer (left side), an output layer (right side), and multiple hidden layers (middle). Each layer of the MLP includes a plurality of nodes, which are called neurons. The neurons of adjacent two layers are connected to each other.

[0151] Optionally, considering the neurons of adjacent two layers, the output h of the neuron of the next layer is the weighted sum of all the neurons x of the previous layer connected to it and is subjected to an activation function, which can be expressed as: h = f (wx + b).

[0152] Where w is a weight matrix, b is a bias vector, and f is an activation function.

[0153] Further optionally, the output of the neural network can be recursively expressed as: y = f n (w n f n-1 (…)+b n ).

[0154] wherein n is an index of the neural network layer, n is greater than or equal to 1, and n is less than or equal to N, wherein N is the total number of layers of the neural network.

[0155] In other words, the neural network can be understood as a mapping from a set of input data to a set of output data. Usually, the neural network is randomly initialized, and the process of obtaining this mapping from the random w and b using the existing data is called training of the neural network.

[0156] Optionally, the specific manner of training is to evaluate the output result of the neural network by using a loss function.

[0157] As shown in FIG. 1d, the error can be back-propagated, and the neural network parameters (including w and b) can be iteratively optimized by the gradient descent method until the loss function reaches the minimum value, i.e. the "better point (e.g. optimal point)" in FIG. 1d. It can be understood that the neural network parameters corresponding to the "better point (e.g. optimal point)" in FIG. 1d can be used as the neural network parameters in the trained AI model information.

[0158] Further optionally, the process of gradient descent can be expressed as:

[0159] wherein θ is the parameter to be optimized (including w and b), L is the loss function, η is the learning rate, and controls the step size of gradient descent, represents the derivation operation, represents the derivative of L with respect to θ.

[0160] Further optionally, the process of back-propagation utilizes the chain rule of partial derivative.

[0161] As shown in FIG. 1e, the gradient of the parameters of the previous layer can be recursively calculated from the gradient of the parameters of the next layer, which can be expressed as:

[0162] wherein w ij is the weight of node j connected to node i, and s i is the input weighted sum on node i.

[0163] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments.

[0164] Some possible communication systems to which the present application is applicable will be described below. It should be understood that the present application is still applicable to other communication systems, and the present application is not limited specifically.

[0165] FIG. 2 is a schematic diagram of a communication system according to an embodiment of the present application. Referring to FIG. 2, the communication system includes a terminal device 101, an access network device 102, an access network device 103, a computing node 104, and a computing node 105. The computing node management function 106 can be deployed on the side of the core network, or the computing node management function 106 can be deployed on the side of the access network.

[0166] The terminal device 101 communicates with the access network device (such as the access network device 102 or the access network device 103 in FIG. 2) through a Uu interface. The access network device 102 and the access network device 103 are access network devices in a new radio (NR) communication system. In the communication system, the access network devices communicate with each other through an Xn interface.

[0167] The access network device (such as the access network device 102 or the access network device 103 in FIG. 2) is a device deployed in a wireless access network to provide wireless communication functions for terminal devices.

[0168] The computing node (including the computing node 104) can be connected to the access network device without a user plane function (UPF).

[0169] The computing node 104 or the computing node 105 refers to a node for providing computing resources (or providing computing services) and is responsible for the execution of computing tasks. In the embodiments of the present application, the computing tasks include but are not limited to image rendering tasks, video decoding tasks, image processing tasks, image recognition tasks, speech recognition tasks, machine translation tasks, gesture recognition tasks, unmanned driving tasks, route planning tasks, neural network model inference tasks, or neural network model training tasks, etc.

[0170] In the embodiments of the present application, one access network device can have an association relationship with one or more computing nodes, and the access network device obtains computing resources through the plurality of computing nodes. Alternatively, a plurality of access network devices can have a management relationship with one computing node, and the computing resources provided by the computing node can be allocated to the plurality of access network devices for use.

[0171] For ease of understanding, an example is described with reference to FIG. 3a, which is another schematic diagram of a structure of a communication system in embodiments of the present application. The access network device 1 has an association relationship with the computing node 1, the computing node 2 and the computing node 3. The association relationship between the access network device and the computing node means that the access network device and the computing node have a physical connection, the access network device can use the computing resource provided by the computing node, and the access network device can request the computing node to perform a computing task. The access network device 1 can use the computing resource provided by the computing node 1, the computing node 2 and the computing node 3, and the computing task required to be performed by the access network device 1 can be performed by the computing node 1, the computing node 2 and / or the computing node 3. The access network device 2 has an association relationship with the computing node 3. The access network device 2 can use the computing resource provided by the computing node 3, and the computing task required to be performed by the access network device 2 can be performed by the computing node 3.

[0172] It should be noted that the names of the computing node 104 and the computing node 105 can change as the communication system evolves, and as long as other functional network elements with similar functions to the computing node 104 and the computing node 105 have other names, they can be understood as the computing node 104 and the computing node 105 of the present application. For example, the computing node can also be referred to as a task execution function (TEF), a far-edge intelligent node (FeIN), a far-edge computing node (FeCN), a task execution node, a computing execution function (CEF), a computing power node, a computing power device, a computing power apparatus, a computing power board, a computing execution node, an edge computing node, a computing function node, a computing task node, a computing task execution node, a computing function, a computing task function, or a computing task execution function, etc., and the specific application is not limited.

[0173] The computing node management function 106 is configured to manage the connection of the computing. For example, for the execution of the computing task, the transmission of data between the terminal and the computing node is involved. The computing node management function 106 is configured to manage the connection between the terminal and the computing node. For example, in an implementation, a protocol data unit (PDU) session or a computing dedicated session can be established between the terminal and the computing node, and then the computing node management function 106 is configured to manage the session, such as the establishment, modification and / or release of the session. For another example, the computing node management function 106 is configured to select the computing node for the computing task. For another example, the computing node management function 106 is configured to select the computing node for the access network device according to the computing task required to be performed.

[0174] In a possible implementation, the computing node management function 106 is separately arranged as a network element of the core network.

[0175] In another possible implementation, the computing node management function 106 is deployed in an access and mobility management function (AMF) or a session management function (SMF), and the AMF or the SMF provides the related function of the computing node management function 106. In other words, the computing node management function 106 is arranged together with the AMF or the SMF.

[0176] In another possible implementation, the computing node management function 106 is deployed in the computing node 104 or the computing node 105. In other words, the computing node 104 or the computing node 105 itself provides the management function, and the computing node management function 106 is the control plane of the computing node 104 or the computing node 105.

[0177] In another possible implementation, the computing node management function 106 is deployed in the access network device, and the access network device implements the management of the computing node as a function.

[0178] For ease of understanding, an example is described with reference to FIG. 3b, which is another structural schematic diagram of a communication system in the embodiment of the present application. The connection between the access network device and the computing node 1 or the computing node 2 is used to transmit user plane data. The computing node management function is responsible for managing the connection between the access network device and the computing node, and can also be used to select the computing node for a computing task. For example, a computing task requested by a terminal device can be executed by the computing node 1 but cannot be executed by the computing node 2. In this case, the computing node management function establishes the connection between the terminal device and the computing node 1, so that after the access network receives the computing task submitted by the terminal device, the computing task can be submitted to the computing node 1 for execution.

[0179] It should be noted that the name of the computing node management function 106 can change as the communication system evolves, and as long as other function network elements with similar functions of the computing node management function 106 have other names, they can be understood as the computing node management function 106 of the present application. For example, the computing node management function 106 can also be a task management function (TMF), a computing management function (CMF), a task management node, a computing management function, a computing task management function, a computing task management node, a computing power management function, or a computing power management node, and the like, and the embodiments of the present application do not limit the same.

[0180] It should be noted that the communication system shown in FIG. 2 can further include more or less computing nodes, which is not limited herein.

[0181] It should be noted that the name of the AMF in the communication system shown in FIG. 2 is only an example. The name of the AMF can change as the communication system evolves. As long as other functional network elements with similar functions to the AMF have other names, they can be understood as the AMF of the present application. For example, the AMF can also be referred to as a mobile management network element, or a mobile management function, etc. The specific application is not limited. The name of the UPF can change as the communication system evolves. As long as other functional network elements with similar functions to the UPF have other names, they can be understood as the UPF of the present application. For example, the UPF can also be referred to as a user plane network element, or a user plane management network element, etc. The specific application is not limited.

[0182] Referring to FIG. 4, FIG. 4 is a schematic diagram of another communication system in an embodiment of the present application. In the communication system shown in FIG. 4, the access network device and the computing node 1 to the computing node M establish a transmission link (or establish a transmission path, or establish a connection, or establish an association relationship), and M is an integer greater than or equal to 1. In other words, the computing node is an external device relative to the access network device, and the computing node and the access network device are independent of each other. For example, the computing node is an external service card or computing card, which establishes a connection with the access network device through an interface.

[0183] Referring to FIG. 5, FIG. 5 is a schematic diagram of another communication system in an embodiment of the present application. In the communication system shown in FIG. 5, the access network device and the computing node 1 to the computing node M establish a transmission link (or establish a transmission path, or establish a connection, or establish an association relationship), and M is an integer greater than or equal to 1. In other words, the computing node is an internal device relative to the access network device, and the access network device is internally provided with the computing node, and the computing node is deployed in the access network device. For example, the computing unit (such as a processor or a computing card) built-in the access network device is the computing node.

[0184] In other words, the relative position relationship between the computing node and the access network device is not limited in the embodiments of the present application. The computing node can be an internal device of the access network device or an external device of the access network device.

[0185] It should be noted that when the access network device adopts a separation architecture of a centralized unit (CU) and a distributed unit (DU), the computing node can be deployed or integrated on the CU or the DU, and the specific application is not limited.

[0186] The technical solutions of the present application can be applied to a cellular communication system related to the 3rd generation partnership project (3GPP). For example, a fourth generation (4G) communication system, a fifth generation (5G) communication system, a future communication system after the fifth generation communication system. For example, the fourth generation communication system can include a long term evolution (LTE) communication system, an LTE frequency division duplex (FDD) system, or an LTE time division duplex (TDD) system. The fifth generation communication system can include a new radio (NR) communication system. The technical solutions of the present application can also be applied to a wireless fidelity (WiFi) system, a communication system supporting multiple wireless technology fusion, a device-to-device (D2D) system, an Internet of Things communication system, an industrial Internet communication system, a vehicle to everything (V2X) communication system, or a satellite communication system, etc.

[0187] The terminal device, the access network device and the computing node related to the present application are introduced as follows.

[0188] A terminal device, also referred to as a user equipment (UE), a mobile station (MS), a mobile terminal (MT), a customer premise equipment (CPE), etc. A terminal device is a device that includes a wireless communication function (providing voice / data connectivity to a user). For example, a handheld device with wireless connectivity, a vehicle-mounted device, a machine type communication (MTC) terminal, etc. Currently, a terminal device can include a mobile phone, a tablet computer, a notebook computer, a palm computer, a mobile internet device (MID), a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical surgery, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, or a wireless terminal in smart home, etc. For example, a wireless terminal in self driving can be a drone, a helicopter, or an airplane, etc. For example, a wireless terminal in vehicle-to-everything (V2X) can be a vehicle-mounted device, a whole vehicle device, a vehicle-mounted module, a vehicle, or a ship, etc. A wireless terminal in industrial control can be a camera, a robot, or a mechanical arm, etc. A wireless terminal in smart home can be a television, an air conditioner, a sweeping machine, a sound box, or a set-top box, etc. A terminal device can also be a transport vehicle with a wireless communication function, a communication module, a road side unit (RSU) with a terminal device function.

[0189] It should be noted that a terminal device can be a device or apparatus with a chip, or a device or apparatus integrated with a circuit, or a chip, a module or a control unit in the above-mentioned devices or apparatus, and the specific application is not limited. For example, the chip can be a chip responsible for communication function in the terminal device, for example, a Modem chip, also known as a baseband chip, or a system on chip (SoC) chip or a system in package (SIP) chip containing a modem core.

[0190] An access network device is a device deployed in a wireless access network to provide wireless communication functions for terminal devices. The access network device can also be referred to as a radio access network (RAN) entity, an access node, a network node, a network device, or a communication apparatus, etc.

[0191] Specifically, the access network device can be an access network device for a 3rd generation partnership project (3GPP) related cellular system. For example, a fourth-generation (4G) mobile communication system, a 5G mobile communication system, or a future mobile communication system. The access network device can also be an access network device in an open RAN (O-RAN or ORAN) or a cloud radio access network (CRAN). Alternatively, the access network device can also be an access network device in a communication system obtained by fusing two or more of the above communication systems.

[0192] The access network device includes, but is not limited to, an evolved Node B (eNB), a radio network controller (RNC), a Node B (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (for example, a home evolved NodeB, or home Node B, HNB), a baseband unit (BBU), an access point (AP) in a wireless fidelity (WiFi) system, a macro base station, a micro base station, a wireless relay node, a donor node, a wireless controller in a CRAN scenario, a wireless backhaul node, a transmission point (TP), or a transmission and receiving point (TRP). The access network device can also be an access network device in a 5G mobile communication system. For example, a next generation NodeB (gNB), a TRP, a TP, or one or a group (including multiple antenna panels) of antenna panels of a base station in a new radio (NR) system, or a base station in a 5G mobile communication system. Alternatively, the access network device can also be a network node constituting a gNB or a transmission point. For example, a centralized unit (CU), a distributed unit (DU), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and the DU can be separately arranged, or can be included in the same network element. For example, a BBU. The RU can be included in a radio frequency device or a radio frequency unit. For example, in a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH). Alternatively, the access network device can also be a server, a wearable device, a vehicle or a vehicle-mounted device, etc. For example, in a V2X technology, the access network device can be a road side unit (RSU).

[0193] It should be noted that the CU (or CU-CP and CU-UP), DU or RU can also have different names in different systems, but those skilled in the art can understand their meanings. For example, in the ORAN system, the CU can also be referred to as an open centralized unit (O-CU) or an open CU, the DU can also be referred to as an open distributed unit (O-DU), the centralized unit control plane (CU-CP) can also be referred to as an open centralized unit control plane (O-CU-CP) or an open CU-CP, the centralized unit user plane (CU-UP) can also be referred to as an open centralized unit user plane (O-CU-UP) or an open CU-UP, and the RU can also be referred to as an open radio unit (O-RU). The specific application is not limited. Any one of the CU, CU-CP, CU-UP, DU and RU in the present application can be realized by a software module, a hardware module, or a combination of a software module and a hardware module.

[0194] Optionally, for network elements in the ORAN system, each network element can implement the protocol layer functions shown in Table 1 as follows.

[0195] Table 1

[0196] It should be noted that the access network device can be one network element or multiple network elements shown in Table 1 above, and the specific application is not limited.

[0197] It should be noted that Table 1 above is only an example. In actual applications, the protocol layer functions supported by each network element are not limited. For example, each network element can support more protocol layer functions, or the protocol layer functions supported by each network element can be configured in combination with the actual situation. The specific application is not limited.

[0198] The architecture of the CU and the DU of the access network device will be introduced below. The access network device includes at least one CU and at least one DU. Optionally, the access network device also includes at least one RU.

[0199] The following is an example of an access network device including a CU and a DU. The CU has part of the function of the core network, and the CU can include a CU-CP and a CU-UP. The CU and the DU can be configured according to the protocol layer function of the wireless network they implement. For example, the CU is configured to implement the function of the PDCP layer and the protocol layer above (for example, the function of the RRC layer and / or the SDAP layer). The DU is configured to implement the function of the protocol layer below the PDCP layer (for example, the RLC layer, the MAC layer, and / or the PHY layer). For another example, the CU is configured to implement the function of the PDCP layer and the protocol layer above (for example, the RRC layer and / or the SDAP layer), and the DU is configured to implement the function of the protocol layer below the PDCP layer (for example, the RLC layer, the MAC layer, and / or the PHY layer, etc.).

[0200] When the CU includes a CU-CP and a CU-UP, the CU-CP is used to implement the control plane function of the CU, and the CU-UP is used to implement the user plane function of the CU. For example, when the CU is configured to implement the function of the PDCP layer, the RRC layer and the SDAP layer, the CU-CP is used to implement the function of the RRC layer and the control plane function of the PDCP layer, and the CU-UP is used to implement the function of the SDAP layer and the user plane function of the PDCP layer.

[0201] The CU-CP can interact with a network element in the core network for implementing the control plane function. The network element in the core network for implementing the control plane function can be an access and mobility function network element, such as an AMF in a 5G mobile communication system. The AMF is used to be responsible for mobility management in the mobile network, such as location update of the terminal device, registration network of the terminal device, handover of the terminal device, etc. The CU-UP can interact with a network element in the core network for implementing the user plane function. The network element in the core network for implementing the user plane function, for example, the user plane function in a 5G mobile communication system, is used to be responsible for the forwarding and receiving of data in the terminal device.

[0202] The above configuration of the CU and the DU is only an example, and the CU and the DU can be configured to have functions as needed. For example, the CU or the DU can be configured to have more protocol layer functions, or the CU or the DU can be configured to have part of the processing function of the protocol layer. For example, part of the function of the RLC layer and the function of the protocol layer above the RLC layer are arranged in the CU, and the remaining function of the RLC layer and the function of the protocol layer below the RLC layer are arranged in the DU. For another example, the function of the CU or the DU can be divided according to the service type or other system requirements. For example, according to the delay, the function that needs to meet the requirement of shorter delay is arranged in the DU, and the function that does not need to meet the requirement of shorter delay is arranged in the CU.

[0203] The DU and the RU can cooperate to jointly implement the functions of the PHY layer. One DU can be connected with one or more RUs. The functions of the DU and the RU can be configured in various manners according to design. For example, the DU is configured to implement baseband functions, and the RU is configured to implement radio frequency functions. For another example, the DU is configured to implement high-layer functions in the PHY layer, and the RU is configured to implement low-layer functions in the PHY layer or implement the low-layer functions and radio frequency functions. The high-layer functions in the PHY layer can include a part of functions of the PHY layer that are closer to the MAC layer, and the low-layer functions in the PHY layer can include another part of functions of the PHY layer that are closer to the radio frequency side.

[0204] It should be noted that the access network device can be the device or apparatus shown above, or a component (for example, a chip), a module, or a unit in the device or apparatus shown above, and the specific application does not make any limitation.

[0205] Optionally, the RIC module is also involved in the ORAN architecture. As shown in FIG. 6, FIG. 6 is a schematic diagram of an application architecture involving the RIC module in the ORAN architecture. As shown in FIG. 6, the communication system includes the RIC module, which is specifically divided into a near-real time RIC (near-RT RIC) module and a non-real time RIC (Non-RT RIC) module. The near-real time RIC module is used for model training and inference. For example, the near-real time RIC module is used for training an artificial intelligence (AI) model and performing inference using the AI model. The near-real time RIC module can also obtain information for training or inferring the AI model from the access network node (for example, the CU, the CU-CP, the CU-UP, the DU, and / or the RU) and from the terminal device. Optionally, the near-real time RIC module can deliver the result of training or inferring the AI model to the access network node and / or the terminal device. Optionally, the CU, the DU, and the RU can interact the result. For example, the near-real time RIC delivers the result to the DU, and the DU sends the result to the RU, thereby realizing near-real time intelligent management of the access network.

[0206] The non-real-time RIC module is used for AI model training and inference. For example, the AI model is used for training and inference. The non-real-time RIC module can obtain data from the access network node (for example, CU, CU-CP, CU-UP, DU, and / or RU) and / or terminal device, and perform AI model training or inference on the data as training data or inference data to obtain corresponding results. Then, the non-real-time RIC module delivers the results to the access network node and / or terminal device. Optionally, the CU, DU, and RU can interact the results. For example, the non-real-time RIC module delivers the results to the DU, and the DU sends the results to the RU.

[0207] It should be understood that the near-real-time RIC module and the non-real-time RIC module can be deployed independently as independent network elements, or can be part of other network devices. For example, the near-real-time RIC module is deployed in the access network node. For example, the near-real-time RIC module can be deployed in the CU and the DU, and the non-real-time RIC module is deployed in the operations, administration and maintenance (OAM), cloud server, core network device or other network device.

[0208] The structure of the access network device including the CU and the DU is introduced below. FIG. 7 is a structure diagram of an access network device according to an embodiment of the present application. Taking the access network device as an example, the gNB, please refer to FIG. 7, in the 5G communication system, the gNBs are connected through the Xn interface, and the gNBs are connected with the 5th generation mobile communication technology core network (5GC) through the NG interface. As shown in FIG. 7, the gNB1 and the gNB2 are connected through the Xn interface. The gNB1 is connected with the 5GC through the NG interface 1, and the gNB2 is connected with the 5GC through the NG interface 2.

[0209] The gNB can include the CU and the DU. That is, the functions of the base station are divided, and part of the functions of the base station are deployed in a gNB-CU, and the remaining functions are deployed in a gNB-DU. Multiple gNB-DUs share one gNB-CU, which can save cost and is easy to expand the network. For example, as shown in FIG. 7, the gNB1 includes the gNB-CU1, the gNB-DU1 and the gNB-DU2. The gNB-CU1 is connected with the gNB-DU1 through the F1 interface 1, and connected with the gNB-DU2 through the F1 interface 2. The structure of the gNB2 is similar to that of the gNB1, which will not be described one by one here.

[0210] The above FIG. 7 is only an example, one gNB-CU can connect one or more gNB-DUs, which is not limited in the present application.

[0211] The split of the gNB-CU and the gNB-DU can be according to the protocol stack. For example, the RRC layer, the SDAP layer, and the PDCP layer correspond to the protocol stacks deployed in the gNB-CU, respectively. The radio link control (RLC) layer, the MAC layer, and the PHY layer correspond to the protocol stacks deployed in the gNB-DU, respectively. The gNB-CU and the gNB-DU are connected through the F1 interface. The above example is only for introducing the gNB-CU and the gNB-DU, and the protocol stacks deployed in the gNB-CU and the gNB-DU are not limited in the present application.

[0212] In the present application, in the solution in which the access network device adopts the CU and the DU structure, the gNB-CU is referred to as the CU, and the gNB-DU is referred to as the DU.

[0213] It should be noted that the CU and the DU are an implementation manner in which the access network device is divided into network units, and the functions contained in the CU and the DU can be divided according to evolution or requirements. The present application does not limit the functions contained in the CU and the DU, respectively. The names corresponding to the CU and the DU are also not limited in the present application.

[0214] The computing node:

[0215] The computing node can be a part of the access network device (for example, a processor, a circuit, a chip, or a chip system, etc.), or can be executed by a logical module or software (for example, a computing module) inside the access network device. The computing node can also be an algorithm board or a service board connected to the access network device. The computing node can also be a computing device, a terminal device, an Internet of Things device, a smart home device, an industrial control device, a vehicle device, or a drone device, etc. which can establish a connection with the access network device.

[0216] The computing node can be directly connected to the access network device, or can be connected to the access network device through other links or a bearer network, and the embodiments of the present application do not limit this. The access network device can discover the computing node in various ways, for example: the computing node actively registers with the access network device, the computing node registers with a computing node management function, the computing node management function configures the computing node to the access network device, or the core network device configures the computing node to the access network device, and the embodiments of the present application do not limit this.

[0217] Next, refer to FIG. 8, which is a schematic diagram of a communication scenario according to an embodiment of the present application. The communication scenario shown in FIG. 8 includes at least one first access network device and N second access network devices, where N is a positive integer greater than or equal to 1. The second access network device refers to another access network device that is adjacent or adjacent to the first access network device in a geographical area. Taking the communication scenario shown in FIG. 8 as an example, a migration calculation method proposed in the embodiments of the present application is introduced. Refer to FIG. 9, which is a schematic diagram of an embodiment of a calculation handover method according to an embodiment of the present application. The migration calculation method proposed in the embodiments of the present application includes the following steps:

[0218] S0, the first computing node executes the computing task requested by the terminal device to execute.

[0219] In the embodiments of the present application, the access network device that provides computing services and communication services for the terminal device before handover is referred to as a second access network device, and the access network device that provides computing services and communication services for the terminal device after handover is referred to as a first access network device.

[0220] In step S0, the second access network device provides computing services and communication services for the terminal device. Specifically, the computing task requested by the terminal device to execute is executed by the first computing node. The first computing node refers to a computing node associated with the second access network device.

[0221] S1, the second access network device sends a first request to the first access network device, carrying context information of the computing task requested by the terminal device to execute.

[0222] In a possible implementation, the first request can be a handover request.

[0223] Specifically, the context information of the computing task carried by the first request includes any one or more of the following information: identification information of the computing task, identification information of the application, routing information, identification information of a computing node supporting execution of the computing task, or first indication information. The context information of the computing task is described in detail below.

[0224] The identification information of the computing task can be a task identifier (task ID). The identification information of the computing task can also indicate the type of the computing task, such as "picture coding", "video coding", "AI model training", "path planning", or "autonomous driving", etc. The first access network device directly or indirectly determines other information of the computing task according to the identification information of the computing task, for example, the first access network device determines the computing task application information and / or the QoS parameter of the computing task corresponding to the task ID according to the task ID, so as to save communication overhead.

[0225] Exemplarily, the task ID can be a set of serial numbers or a set of strings. The task ID can be preconfigured by a core network device or a computing node management function or a network manager, for example, an OAM, or pre-defined in a protocol, or negotiated by the computing nodes, or defined by the computing nodes or the access network device, and the embodiments of the present application do not limit this.

[0226] identifier (APP ID) of the application, or a fully qualified domain name (FQDN) of the application or an application name, etc.

[0227] The routing information is used to indicate a transmission path or a computing session of the computing task related data. For example, the transmission path is used to carry the computing task related data, downlink packets of the computing task, or a computing result of the computing task, etc. The transmission path can be understood as a point-to-point connection between the base station and the computing node, and the transmission path is used to transmit UE related data. Alternatively, the routing information is used to identify a point-to-point connection or a transmission path between the terminal device and the computing node. Alternatively, the routing information is used to identify a session for computing data transmission between the terminal device and the computing node. The transmission path includes but is not limited to a connection, a path, a tunnel or a link.

[0228] Exemplarily, the routing information includes identification information of the transmission path. At this time, the identification information is used to identify the transmission path between the terminal device and the second computing node.

[0229] For example, when a protocol stack between the base station and the computing node is based on a general packet radio service tunneling protocol-user plane (GTP-U), the identification information of the transmission path can be a tunnel endpoint identifier (TEID).

[0230] For another example, the identification information of the transmission path can be identification information of the terminal device, for example, including an international mobile equipment identity (IMEI) of the terminal device or a media access control (MAC) address of the terminal device, etc.

[0231] Exemplarily, the routing information comprises address information of the terminal device, which can comprise an Internet protocol (IP) address of the terminal device and / or a port number of the terminal device. The address information of the terminal device can be address information allocated to the terminal device by an operator or a network side, or address information allocated to the terminal device by the access network device, which is not limited in the present application.

[0232] Exemplarily, when the protocol stack between the access network device and the computing node is as shown in Table 2, the routing information can comprise a UE level ID. Table 2 is as follows:

[0233] Table 2

[0234] In yet another example, when the protocol stack between the access network device and the computing node is as shown in Table 3, the routing information can comprise an IP address of the UE. Table 3 is as follows:

[0235] Table 3

[0236] The identification information of the computing node supporting execution of the computing task indicates which computing nodes support execution of the computing task requested by the terminal device to be executed, or the identification information of the computing node indicates the computing node currently executing the computing task. For example, the second access network device is associated with computing node 1 and computing node 2, and the second access network device determines that computing node 1 supports execution of the computing service requested by the terminal device, and the identification information of computing node 1 can be carried in the first request. So that the first access network device detects whether the first access network device and the computing node 1 have an association relationship, and further determines whether the computing node 1 can provide the computing service to the terminal device. Exemplarily, the identification information of the computing node can be a tunnel endpoint identifier (TEID) of the computing node, can be an IP address of the computing node, or can be an identifier of the computing node.

[0237] The first indication information indicates that the computing task needs to be switched to the first access network device for continuous execution. In one example, during execution of the computing task by the second access network device, the terminal device needs to be switched to the first access network device, and at this time, the computing task has not been completed. Therefore, the first indication information needs to be carried in the second request sent by the second access network device to the first access network device, and the first indication information is used to instruct the first access network device to continue execution of the computing task.

[0238] Optionally, the context information of the computing task can further include: relevant information of the terminal device, such as an identity of the terminal device or address information of the terminal device; and a quality of service (QoS) parameter of the computing task, the QoS parameter indicating a QoS requirement of computing data related to the computing task.

[0239] Optionally, when the computing task is related to a neural network model, the context information of the computing task can further include any one or more of the following information: computing state information of the neural network model related to the computing task, type information of the neural network model, parameter information of the neural network model, KV Cache information of the neural network model, information of one or more intermediate layers included in the neural network model, initial input feature information of the neural network model, or output feature information of one or more intermediate layers of the neural network model. The following will be described respectively.

[0240] The type information of the neural network model and / or the parameter information of the neural network model include, but are not limited to: an architecture type of the neural network model, an activation function of the neural network model, a loss function of the neural network model, an optimizer of the neural network model, a layer type of the neural network model, or a regularization method of the neural network model.

[0241] Exemplarily, the architecture type of the neural network model: for example, a feedforward neural network (FNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM), a generative adversarial network (GAN), or a generative pre-trained (GPT) model, etc.; the activation function of the neural network model: for example, a rectified linear unit (ReLU), a Sigmoid function, or a hyperbolic tangent function (Tanh), etc.; the loss function of the neural network model: for example, a mean squared error (MSE), a cross-entropy, etc., used to evaluate the performance of the model during the training process; the optimizer of the neural network model: for example, a stochastic gradient descent (SGD), an adaptive moment estimation (Adam), or a root mean square propagation (RMSprop), etc.; the layer type of the neural network model: for example, a dense layer (Dense), a convolutional layer (Convolutional), a pooling layer (Pooling), or a recurrent layer (such as LSTM, GRU), etc.; and the regularization method of the neural network model: for example, L1, L2 regularization, Dropout, etc.

[0242] The parameter information of the neural network model includes but is not limited to: weights, biases, learning rate, momentum, decay rate, batch size, or number of epochs, etc. The weights refer to the weight value of each connection, which determines how the input signal affects the downstream nodes in the neural network model; the biases refer to the bias value of each layer or each node in the neural network model, which is used to adjust the output of the activation function; the learning rate refers to the rate used to adjust the weights in the optimization process; the momentum refers to the parameter used to accelerate convergence in the optimization process; the decay rate refers to the rate used to control the decay of the learning rate over time; the batch size refers to the number of samples used in each training iteration; and the number of epochs refers to the number of times the dataset is used to train the model.

[0243] The computing state information of the neural network model can be used to indicate that the terminal device needs to switch the related task of the neural network model to the first access network device for execution when switching. The computing state information of the neural network model can also indicate that the related task of the neural network model needs to be continued by a new access network device (or a new computing node). Optionally, the computing state information of the neural network model includes any one or more of the following information: intermediate features of the neural network model, gradient information of the neural network model, model configuration information of the neural network model, or training progress of the neural network model.

[0244] The intermediate features indicate intermediate activation values or feature maps generated by the neural network model during the training process. The gradient information indicates gradient values calculated by the neural network model during the back propagation process. The model configuration information indicates configuration parameters of the neural network model during the model training, such as the number of layers, the number of nodes in each layer, etc. The training progress of the neural network model indicates the batch, the number of iterations, or the completion degree of the current training of the neural network model.

[0245] The information of one or more intermediate layers included in the neural network model is used to indicate one or more intermediate layers included in the neural network model, which can also be referred to as a split position of the neural network model or a breakpoint position of the neural network model.

[0246] The KV Cache is a method for accelerating inference of a large model, characterized by a key (K) value (V) table. The context information of the computing task includes KV Cache information of the neural network model, which can improve the inference efficiency of the first access network device for the neural network model.

[0247] The initial input feature information of the neural network model is, for example, feature information initially input by the terminal device to the neural network model.

[0248] The output feature information of one or more intermediate layers of the neural network model is, for example, feature information output by one or more intermediate layers of the neural network model when the second access network device executes the training task of the neural network model.

[0249] The input or output feature information of the neural network model can also be the number of basic data units, such as the number of tokens, which refers to the number of basic data units input or output during the inference process of the neural network model.

[0250] S2, the first access network device sends a first response to the second access network device.

[0251] Step S2 is an optional step.

[0252] In step S2, in response to the first request, the first access network device performs admission control on the terminal device. If the terminal device is allowed to switch to the first access network device, the first access network device sends a first response to the second access network device, which can be a handover request ACK message.

[0253] S3, the first access network device sends a second request to the first computing node.

[0254] Specifically, the second request includes any one or more of the following information: context information of the computing task, or address information of the first access network device. The second request can also include identification information of the first access network device, or information of a serving cell of the terminal device. The context information of the computing task is described above in step S1 and will not be repeated here. The address information of the first access network device can also be user plane address information of the first access network device.

[0255] The second request is used to instruct the terminal device that requests to perform the computing task to switch (or move) to the first access network device, or to notify location update of the terminal device.

[0256] In one possible implementation, the second request is used to request to establish a transmission path between the first access network device and the first computing node, which is used to transmit data related to the computing task. Correspondingly, the first computing node establishes the transmission path with the first access network device in response to the second request.

[0257] In another possible implementation, the second request is used to instruct the transmission path between the second access network device and the first computing node about the terminal device to switch to the first access network device. Correspondingly, the first computing node changes the transmission path between the second access network device and the first computing node about the terminal device in response to the second request. Before the change, the two ends of the transmission path are the second access network device and the first computing node respectively; after the change, the two ends of the transmission path are the first access network device and the first computing node respectively.

[0258] In another possible implementation, the second request can also instruct the first computing node to send data of the computing task to the corresponding terminal device through the first access network device.

[0259] After step S1, in response to the first request, the first access network device sends a second request to the first computing node, the second request being used to request the first computing node to perform the computing task requested by the terminal device to perform. Alternatively, the second request is used to instruct the first computing node to send computing task related data to the terminal device. Alternatively, the second request is used to request to establish a connection (link, tunnel or transmission path) between the first computing node and the second access network device with respect to the computing task.

[0260] The second request can be an Nx switch path message, Nx (port) referring to a port between the computing node and the access network device, and the Nx port can be replaced by other ports, and the embodiments of the present application do not limit this.

[0261] S4, the first computing node sends a second response to the first access network device.

[0262] Step S4 is an optional step.

[0263] In step S4, in response to the second request, the first computing node sends a second response to the first access network device, the second response indicating that the first computing node allows to process the computing task requested by the first access network device to perform. For example, the second response is an Nx switch path ACK message.

[0264] Optionally, the second response can include address information of the first computing node.

[0265] Optionally, after step S4, the first computing node modifies the destination address of the computing task related downlink packet to the address information of the first access network device, so that the downlink packet can reach the first access network device.

[0266] Optionally, the first computing node can also carry identification information of the transmission path, for example, when the transmission path between the second access network device and the first computing node is based on GTP-U, the first computing node can also modify the TEID (Tunnel Endpoint Identifier) of the transmission path to the TEID of the first access network. So that after the first access network device receives the downlink packet, it can identify according to the TEID that the downlink packet is the computing task related packet sent to the corresponding terminal device.

[0267] In another possible implementation, the first computing node modifies the destination address to the address information of the UE, and sends the computing data to the second access network device, so that after the first access network device receives the downlink packet, it can identify according to the address that the downlink packet is the computing task related packet sent to which UE.

[0268] S5, the first access network device instructs the second access network device to release the context of the terminal device.

[0269] Step S5 is an optional step.

[0270] In the above technical solution, during the process of switching the terminal device from the second access network device to the first access network device, the first access network device can request the first computing node to continue to execute the computing task of the terminal device, so as to guarantee the continuity of the computing task requested by the terminal device during the switching process, guarantee the service continuity of the terminal device, and improve the user experience.

[0271] Next, another possible implementation manner of the computing migration method is introduced in combination with the foregoing embodiments. Please refer to FIG. 10, which is another communication system structure diagram according to the embodiments of the present application. The connection between the terminal device and the first access network device and the second access network device is a data radio bearer (DRB), and the second access network device serves as an anchor access network device. The connection between the first access network device, the second access network device and the computing node is a GTP-U tunnel. The computing node management function (optionally) can be connected with the computing node, the first access network device and the second access network device.

[0272] In combination with FIG. 10, please refer to FIG. 11, which is another flowchart of the computing migration method according to the embodiments of the present application. The computing migration method according to the embodiments of the present application comprises the following steps:

[0273] R1, the second access network device sends a first message to the first access network device, and the first message comprises the context information of the computing task, such as the routing information, the identification information of the first computing node, and / or the address information of the terminal device.

[0274] In one possible implementation manner, the first message is carried in the response of the handover request. Specifically, after determining that the terminal device can be switched to the first access network device, the second access network device sends a handover request to the first access network device (anchor access network device), and carries the information related to the computing task.

[0275] The routing information is used to indicate a transmission path or a computing session of the computing task related data. For example, the transmission path is used to carry the computing task related data, downlink packet of the computing task, or computing result of the computing task. The transmission path can be understood as a point-to-point connection between the base station and the computing node, and the transmission path is used to transmit the UE related data. Alternatively, the routing information is used to identify a point-to-point connection or a transmission path between the terminal device and the computing node. Alternatively, the routing information is used to identify a session for computing data transmission between the terminal device and the computing node. The transmission path includes but is not limited to connection, path, tunnel or link.

[0276] For example, the routing information includes identification information of the transmission path. At this time, the identification information is used to identify the transmission path between the terminal device and the second computing node.

[0277] For example, when the protocol stack between the base station and the computing node is based on a packet radio service user plane tunneling protocol (GTP-U), the identification information of the transmission path can be a tunnel endpoint identification (TEID).

[0278] The address information of the terminal device can include an Internet Protocol (IP) address of the terminal device and / or a port number of the terminal device. The address information of the terminal device can be address information allocated to the terminal device by an operator or a network side, or address information allocated to the terminal device by an access network device, which is not limited in the present application.

[0279] For another example, the identification information of the terminal device is an International Mobile Equipment Identity (IMEI) of the terminal device or a Media Access Control (MAC) address of the terminal device.

[0280] For example, the identification information of the first computing node can be a tunnel endpoint identifier (TEID) of the first computing node, an IP address of the first computing node, or an identifier of the first computing node, which is not limited in the embodiments of the present application.

[0281] It should be noted that step R1 is an optional step.

[0282] R2, the second access network device receives the second message, and the second message includes the address information of the first access network device.

[0283] In a possible implementation, in response to the first message, the first access network device feeds back the address information of the first access network device to the second access network device. The first access network device sends a second message to the second access network device, and the second message includes the address information of the first access network device.

[0284] In another possible implementation manner, the step R1 and / or the step R2 is implemented through a user plane, which means that the first message is user plane data and the second message is user plane data. The user plane data can further include address information of the first access network device.

[0285] Specifically, the first access network device is a serving base station currently providing services for the terminal device, and the second access network device is an anchor base station. The first access network device can be the first serving base station after the terminal device leaves the anchor base station (the second access network device), or the first access network device can be the xth serving base station after the terminal device leaves the anchor base station (the second access network device), where x is a positive integer greater than 1.

[0286] For example, when the terminal device is ready to access a new serving base station, the new serving base station or the current serving base station (for example, the first access network device) triggers the anchor base station (for example, the second access network device) to report the address information of the anchor base station. In order to enable the anchor base station to deliver data related to the computing task of the terminal device to the new serving base station of the UE, the serving base station can further forward the data to the terminal device, so as to ensure that the computing task of the terminal device is not interrupted.

[0287] For example, when the second message is user plane data, the step R1 can not be performed.

[0288] After the step R2, a plurality of possible implementation manners are included, and the specific implementation manners are as follows.

[0289] Manner one: The first computing node modifies the destination address of the data of the computing task.

[0290] Manner two: The second access network device modifies the destination address of the data of the computing task, or the second access network device updates a routing table, and the updated routing table indicates that the data of the computing task needs to be forwarded to the first access network device.

[0291] Next, the specific implementation manners are described in detail.

[0292] Manner one:

[0293] R3, the second access network device (or the anchor access network device) sends the computing task context to the computing node management function, which includes the address information of the first access network device and / or the routing information and / or the information of the computing task, as described above.

[0294] Exemplarily, the second access network device (or the anchor access network device) sends a terminal device location update message (UE location update) to the computing node management function, the terminal device location update message carrying the computing task context. For example, the computing task context includes the address information of the first access network device. The message indicates that the location of the terminal device has moved, and has moved to the first access network device.

[0295] Optionally, the message can also include the routing information. For the routing information, refer to the foregoing embodiments, which are not described herein again.

[0296] R4, the computing node management function sends a connection establishment request to the first computing node.

[0297] In step R4, after receiving the terminal device location update message, the computing node management function sends a connection establishment request to the first computing node, the connection establishment request including: the computing task information, the address information of the first access network device and / or the routing information. The connection establishment request is used to establish the connection (or transmission path) between the computing node and the first access network device.

[0298] R5, the first computing node sends the data of the computing task to the terminal device according to the routing information. For example, the first computing node modifies the destination address of the IP layer to the IP address of the first access network device.

[0299] Alternatively, in step R5, the first computing node adds the routing information in the packet header of the data of the computing task, so that the second access network device determines that the data needs to be sent to the corresponding terminal device according to the routing information.

[0300] For example, in step R5, the first computing node carries the address information of the terminal device (or the identification information of the terminal device, or the identification information of the connection) in the packet header of the data of the computing task, so that the second access network device determines that the data needs to be sent to the corresponding terminal device according to the address information of the terminal device (or the identification information of the terminal device).

[0301] The data of the computing task includes but is not limited to: the related message of the computing task, for example, the downlink message of the computing task, and for example, the computing result of the computing task.

[0302] R6, the first computing node sends the data of the computing task through the first access network device.

[0303] It should be noted that the computing data can first arrive at the second access network device, and since the destination address of the data of the computing task has been modified to the address information of the first access network device, the second access network device forwards the data to the first access network device after receiving the data, or the data can be distributed to the terminal device through the first access network device through other paths.

[0304] Option 2:

[0305] R7, the first computing node sends the data of the computing task to the second access network device.

[0306] R8, the second access network device sends to the first access network device, so that the first access network device sends the data of the computing task to the terminal device.

[0307] In another possible implementation, the second access network device adds the routing information in the packet header of the data of the computing task, for example, the second access network device modifies the destination address of the data packet of the computing task to the address information of the first access network device, and the data packet carries the data of the computing task. So that the data packet can be sent to the first access network device.

[0308] In another possible implementation, the second access network device adds the address information (or the identification information) of the terminal device or the identification information of the connection in the packet header of the data of the computing task, so that the second access network device determines that the data needs to be forwarded to the first access network device according to the address information (or the identification information) of the terminal device.

[0309] R9, the second access network device forwards the data of the computing task to the first access network device.

[0310] In the above technical solution, during the process of switching the terminal device from the second access network device to the first access network device, the first access network device can request the first computing node to continue to execute the computing task of the terminal device, so as to ensure that the computing task requested by the terminal device is not interrupted during the switching process, guarantee the service continuity of the terminal device, and improve the user experience. In addition, the second access network device as an anchor access network device can maintain the connection between the first access network device and the second access network device, so the second access network device can forward the data of the computing task to the first access network device, ensuring that the computing task is not interrupted.

[0311] In combination with the foregoing embodiments, another possible implementation of the computing migration method proposed in the embodiments of the present application will be introduced. In this implementation, the computing task is executed by a new computing node after the terminal device is switched. The computing node that originally executes the computing task is referred to as a second computing node, and the new computing node is referred to as a first computing node. The following will be introduced in combination with the accompanying drawings.

[0312] Please refer to FIG. 12a, which is a flow diagram of another embodiment of the method for calculating migration according to an embodiment of the present application. The method for calculating migration according to an embodiment of the present application comprises the following steps:

[0313] K4. The second access network device sends second indication information to the second computing node, and the second indication information is used to indicate the migration of the computing task.

[0314] The second indication information comprises at least one of the following: the related information of the first computing node, the context information of the computing task, routing information, or the related information of the terminal device. The related information of the first computing node comprises address information of the first computing node and / or identification information of the first computing node. The related information of the terminal device comprises address information of the terminal device and / or identification information of the terminal device.

[0315] The context information of the computing task is described in step K2, and thus is not described here.

[0316] The routing information is described in the foregoing embodiments, and thus is not described here.

[0317] K5. In response to the second indication information, the second computing node sends data of the computing task to the first computing node.

[0318] Before step K4 and step K5, the embodiment can further comprise the following steps:

[0319] K0. The second computing node executes the computing task requested by the terminal device.

[0320] K1. The second access network device sends a first request to the first access network device, and the first request carries context information of the computing task requested by the terminal device.

[0321] In step K1, the context information of the computing task is described in the foregoing embodiments, and thus is not described here.

[0322] K2. The first access network device determines the first computing node according to the first request.

[0323] In step K2, the first access network device determines the first computing node according to the first request, and the first computing node supports providing computing services to the terminal device.

[0324] It should be noted that the first computing node supporting providing the computing service to the terminal device can be replaced with: the first computing node is a computing node supporting executing the computing task requested by the terminal device to execute. Alternatively, the computing resource (or the computing service) provided by the first computing node can execute the computing task requested by the terminal device to execute. Alternatively, the first computing node can execute the computing task requested by the terminal device to execute. Alternatively, the first computing node can meet the requirement of the terminal device on the computing capability (or the computing resource, or the computing service, or the quality of service) when the terminal device requests to execute the computing task.

[0325] The following describes how the first access network device determines the first computing node.

[0326] In a possible implementation, the first access network device determines the first computing node according to the computing information of one or more computing nodes and the computing requirement information of the terminal device. The first computing node is a computing node of the one or more computing nodes that meets the computing requirement information. The computing requirement information indicates information of a computing task required to be executed by the terminal device, or the computing requirement information indicates a requirement of a computing task supported to be executed by the terminal device on a computing capability, and the first computing node meets the computing requirement of the terminal device.

[0327] In an example, the first access network device can obtain the computing requirement information locally, for example, the first access network device processes a computing task for the terminal device, and the computing task needs to be switched to the first computing node to continue to execute, and the first access network device determines the computing requirement information according to context information of the computing task.

[0328] In another example, the first access network device can obtain the computing requirement information from a computing task management function.

[0329] In another example, the terminal device can actively report the computing requirement information to the first access network device.

[0330] Secondly, the specific content included in the computing information of the computing node is introduced. The computing information of the computing node includes any one or more of the following information:

[0331] Information 1: computing resource information of the computing node;

[0332] Information 2: computing load information of the computing node;

[0333] Alternatively, information 3: computing task information supported by the computing node.

[0334] The above information 1 to information 3 are introduced respectively.

[0335] Information 1:

[0336] The computing resource information of the computing node indicates a computing resource related to the computing service. The computing resource information includes any one or more of the following information: hardware resource information, processing capability information, storage capability information, transmission resource information, or software resource information.

[0337] The hardware resource information of the computing node indicates a hardware resource of the computing node. For example, the hardware resource includes, but is not limited to, a central processing unit (CPU) of the computing node, an application processor (AP) of the computing node, a graphics processing unit (GPU) of the computing node, an image signal processor (ISP) of the computing node, a video codec of the computing node, a digital signal processor (DSP) of the computing node, and / or a neural-network processing unit (NPU) of the computing node, etc. For another example, the hardware resource information indicates a model of a hardware device of the computing node.

[0338] The processing capability information can indicate a floating point operation capability of the computing node. The processing capability information can also indicate an average processing delay of the computing node in processing a computing task. The average processing delay can be a processing delay of the computing node in processing a computing task, an average processing delay of the computing node in processing all computing tasks, or an average processing delay of the computing node in processing a certain type of computing task. For example, an average processing delay in processing an image processing task, an average processing delay in processing an image recognition task, an average processing delay in processing a gesture recognition task, or an average processing delay in processing a neural network model training task.

[0339] The storage capability information can indicate a size of a storage space available to the computing node. For multiple computing nodes, each computing node can be configured with a dedicated storage space, or multiple computing nodes can be configured with a shared storage space. Therefore, the storage capability information can indicate a size of a dedicated storage space of the computing node, or a size of a shared storage space of the computing node. For example, the storage capability information indicates a total capacity of a storage space of the computing node, and / or an available amount of a storage space of the computing node.

[0340] The transmission resource information indicates a total transmission bandwidth of the computing node, and / or an available transmission bandwidth of the computing node.

[0341] The software resource information can indicate various types of software, computing services, computing tasks, application services, artificial intelligence (AI) algorithms or neural network models, etc. that the computing node supports to run (or process or execute). Examples: when the computing service supported by the computing node is gesture recognition, the software resource supported by the computing node can include recurrent neural networks (RNN); when the computing service supported by the computing node is image processing, the software resource supported by the computing node can include convolutional neural networks (CNN) and generative adversarial networks (GAN); when the computing service supported by the computing node is video decoding, the software resource supported by the computing node can include video encoding algorithms such as high efficiency video coding (H.265-HEVC) algorithm; when the computing service supported by the computing node is route planning, the software resource supported by the computing node can include the shortest path (Dijkstra) algorithm, etc. The software can also include an operating system supported by the computing node to run.

[0342] Information II:

[0343] The computing load information of the computing node indicates the load status of the computing resource provided by the computing node. The computing load information includes any one or more of the following information: processing resource load information, storage resource load information, or transmission resource load information. For example, the computing load information indicates the memory occupancy rate, transmission bandwidth occupancy rate, and / or processing resource occupancy rate of the computing node, etc.

[0344] Optionally, the computing load information can also indicate the load status of the computing resource provided by one or more cells managed by the computing node.

[0345] In an example, the computing information of the computing node is shown in Table 4.

[0346] Table 4

[0347] In another example, the computing load information of the computing node indicates the number of bearable computing tasks or the number of bearable users provided by the computing node. The number of bearable computing tasks or the number of bearable users represents the computing capability of the computing node (or the computing node associated with the computing node).

[0348] Information III:

[0349] The computing task information supported by the computing node is used to indicate a computing task supported by the computing node, a service supported by the computing node, an application supported by the computing node, and / or a computing service provided by the computing node. The computing task supported by the computing node refers to a computing task that can be processed by the computing node. The service supported by the computing node refers to a service or a computing task related to the service that can be processed by the computing node.

[0350] For example, the computing task information supported by the computing node includes an application type supported by the computing node, such as a name, a domain name, or an identifier of the application, and the like.

[0351] Further, the supported service type includes but is not limited to a video decoding service, an image rendering service, an image processing service, an image recognition service, a voice recognition service, a machine translation service, a gesture recognition service, an unmanned driving service, a route planning service, a neural network model inference service, or a neural network model training service, and the like.

[0352] In another example, for an AI large model service, the supported model or type is, for example, a generative pre-training (GPT) model, and the like.

[0353] For another example, the computing task information supported by the computing node further includes quality of service (QoS) information of a computing task supported by the computing node, which is used to indicate a service quality that can be achieved when the computing node executes the computing task. For example, a processing delay when the computing node executes the computing task. The QoS information can also be related to a specific computing service. For example, when the computing service is an image rendering service, the QoS information can further include a resolution of an image, a code rate of the image, or a definition of the image, and the like. For another example, when the computing service is a neural network model inference service, the QoS information can further include a prediction accuracy, and the like.

[0354] Optionally, the computing task information supported by the computing node can further include identification information of the computing task. For example, a task identification (Task ID).

[0355] The identification information of the computing task can be identification information agreed in advance. For example, the identification information is configured to the computing node or an access network device in advance, and the corresponding computing task can be identified according to the identification information between different access network devices or different computing nodes. For example, the identification information of a computing task related to a video application is set as “video”, the identification information of a computing task related to an extended reality (XR) application is set as “XR”, and the identification information of a computing task related to an AI application is set as “AI”.

[0356] The identification information of the computing task can also be identification information assigned by the computing node management function for each computing task. For example, the identification information of a computing task related to a video application is set as "Task ID#1", the identification information of a computing task related to an extended reality (XR) application is set as "Task ID#2", and the identification information of a computing task related to an AI application is set as "Task ID#3".

[0357] The identification information of the computing task can also be identification information defined by the access network device or the computing node. In another example, the computing information of the computing node is as shown in Table 5.

[0358] Table 5

[0359] K3, the first access network device sends second indication information to the second access network device.

[0360] In step K3, the second indication information, please refer to the foregoing embodiments, which will not be repeated here.

[0361] In the above technical solution, during the process of switching the terminal device from the second access network device to the first access network device, the first access network device can request the first computing node to execute the computing task of the terminal device, and the computing task is executed by the second computing node before the switching. The above method ensures that the computing task requested by the terminal device is not interrupted during the switching process, ensures the service continuity of the terminal device, and improves the user experience.

[0362] In combination with the foregoing embodiments, please refer to FIG. 12b, which is another embodiment flowchart of the computing migration method in the embodiments of the present application. The computing migration method proposed in the embodiments of the present application comprises:

[0363] D0, the second computing node executes the computing task requested by the terminal device.

[0364] D1, the second access network device sends a first request to the first access network device, and the first request carries context information of the computing task requested by the terminal device.

[0365] D2, according to the first request, the first access network device determines the first computing node.

[0366] In step D2, the first access network device determines the first computing node for executing the computing task according to the first request. After step D2, steps D3 and D4 are executed, and it should be noted that the execution order of steps D3 and D4 is not limited in the embodiments of the present application.

[0367] Step D2 is similar to the aforementioned Step K2, and is not described herein.

[0368] D3. After determining the first computing node, the first access network device sends a first response to the second access network device.

[0369] D4. The first access network device sends a second request to the first computing node, where the second request is used to instruct the terminal device to perform the computing task to switch to the first access network device.

[0370] Step D4 is similar to the aforementioned Step S3, and is not described herein.

[0371] D5. The first computing node sends a second response to the first access network device.

[0372] Step D4 is similar to the aforementioned Step S4, and is not described herein.

[0373] It should be noted that Steps D4 to D5 are optional steps.

[0374] D6. In response to the second response, the first access network device sends second indication information to the second access network device.

[0375] D7. The second access network device forwards the second indication information to the second computing node.

[0376] D8. In response to the second indication information, the second computing node forwards computing task related data to the second access network device. Then, the second access network device forwards the computing task related data to the first access network device, and the first access network device reports the data to the first computing node.

[0377] D9. The first access network device instructs the second access network device to release the context of the terminal device.

[0378] In another example, in a handover scenario, the second computing node can directly forward the computing task related data to the first computing node. Please refer to FIG. 13, which is a flow diagram of another embodiment of a computing migration method. The computing migration method includes:

[0379] H0. The second computing node performs the computing task requested by the terminal device.

[0380] H1. The second access network device sends a first request to the first access network device, carrying the context information of the computing task requested by the terminal device.

[0381] H3. The first access network device determines the first computing node according to the first request.

[0382] H4. The first access network device sends a first response to the second access network device.

[0383] H5. The first access network device sends a second request to the first computing node, the second request being used to instruct the terminal device that requests to perform the computing task to hand over to the first access network device.

[0384] H6. The first computing node sends a second response to the first access network device.

[0385] H7. The first access network device sends second indication information to the second access network device.

[0386] Steps H0 to H7 are similar to the aforementioned steps D0 to D7, and thus are not described herein.

[0387] H8. After receiving the second indication information, the second access network device forwards the second indication information to the second computing node.

[0388] H9. In response to the second indication information, the second computing node forwards data related to the computing task directly to the first computing node.

[0389] H10. The first access network device instructs the second access network device to release a context of the terminal device.

[0390] After step H7, step H10 is performed.

[0391] In yet another example, the core network device (or the computing node management function) can determine the first computing node. Please refer to FIG. 14, which is a flow diagram of yet another embodiment of a computing migration method according to an embodiment of the present application. The computing migration method proposed by the embodiment of the present application includes the following steps:

[0392] F0. The second computing node performs the computing task requested by the terminal device.

[0393] F1. The second access network device sends a first request to the first access network device, carrying context information of the computing task requested by the terminal device.

[0394] F2. The first access network device sends a first response to the second access network device.

[0395] Steps F0 to F2 are similar to the aforementioned steps D0 to D2, and thus are not described herein.

[0396] F3. In response to the first request, the second access network device sends a third request to the core network device, the third request carrying the context information of the computing task requested by the terminal device.

[0397] The third request is similar to the first request, and the third request can include any one or more of the following information: identification information of the computing task, identification information of the application, routing information, identification information of a computing node supporting execution of the computing task, or first indication information. For details, please refer to the foregoing step S1, which will not be repeated here.

[0398] F4. According to the third request, the core network device (or the computing node management function) determines the first computing node.

[0399] In step F4, the specific method for the core network device (or the computing node management function) to determine the first computing node is similar to the foregoing step D2, which will not be repeated here.

[0400] F5. The core network device (or the computing node management function) sends a second request to the first computing node, and the second request is used to instruct the terminal device requesting execution of the computing task to switch to the first access network device.

[0401] F6. The first computing node sends a second response to the core network device (or the computing node management function).

[0402] Steps F5 to F6 are similar to the foregoing steps S3 to S4, which will not be repeated here.

[0403] F7. In response to the third request, the core network device (or the computing node management function) sends a third response to the first access network device, and the third response carries second indication information.

[0404] F8. In response to the third request, the core network device (or the computing node management function) sends the second indication information to the second computing node.

[0405] F9. The second computing node sends data related to the computing task to the first computing node.

[0406] F10. The first access network device instructs the second access network device to release the context of the terminal device.

[0407] Steps F8 to H10 are similar to the foregoing steps D7 to D9, which will not be repeated here.

[0408] In another example, please refer to FIG. 15, which is a flow diagram of another embodiment of a computing migration method according to an embodiment of the present application. The computing migration method according to an embodiment of the present application includes the following steps:

[0409] G0. The second computing node executes the computing task requested by the terminal device.

[0410] G1. The second access network device sends a first request to the first access network device, carrying context information of the computing task requested by the terminal device.

[0411] G2, the first access network device sends a first response to the second access network device.

[0412] G3, the first access network device sends a third request to a core network device (or a computing node management function), carrying context information of the computing task requested to be executed by the terminal device.

[0413] G4, according to the third request, the core network device (or the computing node management function) determines a first computing node.

[0414] G5, the core network device (or the computing node management function) sends a second request to the first computing node, the second request being used to instruct the terminal device requesting to execute the computing task to switch to the first access network device.

[0415] G6, the first computing node sends a second response to the core network device (or the computing node management function).

[0416] G7, the core network device (or the computing node management function) sends a third response to the first access network device, carrying second indication information.

[0417] G8, the core network device (or the computing node management function) sends the second indication information to a second computing node.

[0418] Steps G0-G8 are similar to the foregoing steps F0-F8, and will not be described here.

[0419] G9, in response to the second indication information, the second computing node sends data related to the computing task to the first computing node.

[0420] G10, the first access network device instructs the second access network device to release the context of the terminal device.

[0421] In the foregoing technical solution, during the process of switching the terminal device from the second access network device to the first access network device, the first access network device can request the first computing node to continue to execute the computing task of the terminal device, guaranteeing that the computing task requested to be executed by the terminal device does not interrupt during the switching process, guaranteeing the service continuity of the terminal device, and improving the user experience.

[0422] Next, a communication apparatus related to an embodiment of the present application is introduced. The communication apparatus can be used for the first access network device and / or the second access network device in the foregoing embodiments.

[0423] FIG. 16 is a structural schematic diagram of a communication apparatus according to an embodiment of the present application. Please refer to FIG. 16, the communication apparatus 1600 includes a transceiver module 1601 and a processing module 1602.

[0424] The communication apparatus 1600 comprises an access network device, which can be the first access network device, and / or the second access network device. Alternatively, the communication apparatus 1600 comprises components (e.g., chips), modules or units in the access network device, which can be the first access network device, and / or the second access network device.

[0425] The communication apparatus 1600 can be configured to perform all or part of the steps executed by the first access network device in the embodiments shown in FIG. 9 to FIG. 15. For details, reference can be made to related description in the foregoing embodiment shown in FIG. 9.

[0426] The communication apparatus 1600 can be configured to perform all or part of the steps executed by the second access network device in the embodiments shown in FIG. 9 to FIG. 15. For details, reference can be made to related description in the foregoing embodiment shown in FIG. 9.

[0427] The processing module 1602 is configured to perform data processing. The transceiver module 1601 is configured to implement corresponding communication functions.

[0428] Optionally, the transceiver module 1601 can include a sending module and a receiving module. The sending module is configured to perform the sending operations in the above method embodiments. The receiving module is configured to perform the receiving operations in the above method embodiments.

[0429] It should be noted that the communication apparatus 1600 can include the sending module, but not the receiving module. Alternatively, the communication apparatus 1600 can include the receiving module, but not the sending module. Specifically, whether the sending module and the receiving module are included in the communication apparatus 1600 can depend on whether the sending action and the receiving action are included in the above scheme executed by the communication apparatus 1600.

[0430] Optionally, the communication apparatus 1600 can further include a storage module, which can be configured to store instructions and / or data. The processing module 1602 can read the instructions and / or data in the storage module, so that the communication apparatus 1600 implements the foregoing method embodiments.

[0431] The communication apparatus 1600 can be configured to perform the actions performed by the first access network device side in the embodiments shown in FIG. 9 to FIG. 15. The processing module 1602 is configured to perform processing-related operations of the first access network device side in the embodiment shown in FIG. 9. The transceiver module 1601 is configured to perform receiving- or sending-related operations of the first access network device side in the embodiment shown in FIG. 9.

[0432] The communication apparatus 1600 can be configured to perform the actions performed by the second access network device side in the embodiments shown in FIG. 9 to FIG. 15. The processing module 1602 is configured to perform processing-related operations of the second access network device side in the embodiments shown in FIG. 9 to FIG. 15. The transceiver module 1601 is configured to perform receiving- or sending-related operations of the second access network device side in the embodiments shown in FIG. 9 to FIG. 15.

[0433] In an example, the communication apparatus 1600 is applied to a first access network device, and the communication apparatus 1600 comprises:

[0434] The transceiver 1601 is configured to receive a first request from a second access network device, the first request indicating that a terminal device requests to switch from the second access network device to the first access network device, and the first request comprising context information of a computing task requested to be executed by the terminal device.

[0435] The transceiver 1601 is further configured to, in response to the first request, send a second request to a first computing node, the second request being used to indicate that an access network device related to the computing task is changed to the first access network device.

[0436] In a possible implementation, the second request comprises any one or more of the following information:

[0437] The context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device, the routing information indicating a transmission path carrying the computing task.

[0438] In a possible implementation,

[0439] The processing module 1602 is configured to, in response to the first request, determine the first computing node.

[0440] The transceiver 1601 is further configured to send the second request to the first computing node.

[0441] The transceiver 1601 is further configured to receive a second response from the first computing node, the second response indicating that the first computing node allows the computing task requested to be executed by the terminal device to be executed.

[0442] The processing module 1602 is further configured to, in response to the second response, establish a transmission path related to the terminal device with the first computing node.

[0443] In a possible implementation,

[0444] The transceiver 1601 is further configured to send second indication information to the second access network device, the second indication information being used to indicate that a second computing node migrates the computing task to the first computing node, the second computing node being an original computing node for executing the computing task.

[0445] The transceiver 1601 is further configured to receive data related to the computing task from the second access network device.

[0446] In a possible implementation manner,

[0447] The transceiver 1601 is further configured to send a third request to the core network device, where the third request carries context information of the computing task requested by the terminal device to perform, so that the core network device sends the second request to the first computing node, and the second computing node is an original computing node for performing the computing task.

[0448] The transceiver 1601 is further configured to receive a third response from the core network device, where the third response carries second indication information, and the second indication information is used to indicate that the second computing node migrates the computing task to the first computing node, and the second computing node is an original computing node for performing the computing task.

[0449] The processing module 1602 is further configured to, in response to the third response, establish a transmission path related to the terminal device with the first computing node.

[0450] In a possible implementation manner, the second indication information includes:

[0451] routing information, context information of the computing task, and / or related information of the first computing node, where the routing information indicates a transmission path carrying the computing task.

[0452] In a possible implementation manner, the context information of the computing task includes any one or more of the following information:

[0453] identification information of the computing task,

[0454] identification information of an application related to the computing task,

[0455] the routing information, the routing information indicating a transmission path carrying the computing task,

[0456] identification information of a computing node supporting execution of the computing task,

[0457] or first indication information indicating that the computing task needs to be switched to the first access network device for continuous execution.

[0458] In a possible implementation manner, the context information of the computing task further includes any one or more of the following information:

[0459] identification information of a neural network model related to the computing task,

[0460] type information of the neural network model,

[0461] parameter information of the neural network model,

[0462] key-value storage (KV Cache) information of the neural network model,

[0463] information of one or more intermediate layers included in the neural network model,

[0464] initial input feature information of the neural network model,

[0465] or output feature information of the one or more intermediate layers of the neural network model.

[0466] It should be understood that the specific processes in which the respective modules perform the above-mentioned corresponding processes have been described in detail in the above-mentioned method embodiments, and for the sake of brevity, will not be described here.

[0467] The processing module 1602 in the above embodiment can be implemented by at least one processor or processor-related circuit. The transceiver module 1601 can be implemented by a transceiver or transceiver-related circuit. The transceiver module 1601 can also be referred to as a communication module or a communication interface. The storage module can be implemented by at least one memory.

[0468] The present application also provides another communication device, and FIG. 17 is another structural schematic diagram of a communication device according to an embodiment of the present application. Referring to FIG. 17, the communication device 1700 includes a processor 1701.

[0469] Optionally, the communication device 1700 further includes a memory 1702.

[0470] Optionally, the communication device 1700 further includes a transceiver 1703.

[0471] In a possible implementation, the processor 1701, the memory 1702, and the transceiver 1703 are connected through a bus respectively, and the memory 1702 stores computer instructions.

[0472] In a possible implementation, when the communication device 1700 includes an access network device, or the CU or DU included in the access network device, or a component (for example, a chip), a module or a unit in the access network device, the communication device 1700 can be used to execute the steps performed by the first access network device and / or the second access network device in the above-mentioned method embodiments, and the related description can be referred to the above-mentioned method embodiments.

[0473] Optionally, the processing module 1602 in the above-mentioned embodiment shown in FIG. 16 can be the processor 1701, and the transceiver module 1601 in the above-mentioned embodiment shown in FIG. 16 can be the transceiver 1703. Alternatively, the processing module 1602 in the above-mentioned embodiment shown in FIG. 16 can be the processor 1701, and the transceiver module 1601 in the above-mentioned embodiment shown in FIG. 16 can be the transceiver 1703.

[0474] The embodiment of the present application further provides a communication device. FIG. 18 is another structural schematic diagram of the communication device according to the embodiment of the present application. Referring to FIG. 18, the communication device 1800 can be the terminal device in the method embodiment, or can be a component (for example, a chip), a module or a unit of the terminal device in the method embodiment. The communication device 1800 can be used to execute the operation performed by the terminal device in the method embodiment.

[0475] The processor is mainly used for processing data or signals, controlling the communication device, executing a corresponding software program, processing data of the software program and the like.

[0476] It should be noted that the signal processing algorithm of the processor has weak capability and cannot perform complex signal processing algorithm.

[0477] The memory is mainly used for storing software programs and data. The radio frequency circuit is mainly used for conversion between a baseband signal and a radio frequency signal and processing of the radio frequency signal.

[0478] The antenna is mainly used for receiving and transmitting radio frequency signals in the form of electromagnetic waves.

[0479] Optionally, the communication device 1800 further includes an input and output device, for example, a touch screen, a display screen, a keyboard and the like, which are mainly used for receiving data input by a user and outputting data to the user.

[0480] When data needs to be transmitted, the processor performs baseband processing on the data to be transmitted, and outputs a baseband signal to the radio frequency circuit. The radio frequency circuit performs radio frequency processing on the baseband signal, and transmits a radio frequency signal in the form of electromagnetic waves through the antenna. When data is transmitted to the communication device, the radio frequency circuit receives the radio frequency signal through the antenna, converts the radio frequency signal into a baseband signal, and outputs the baseband signal to the processor. The processor converts the baseband signal into data and processes the data.

[0481] For the convenience of description, only one memory and one processor are shown in FIG. 18. In the actual product of the communication device, one or more processors and one or more memories can exist. The memory can also be referred to as a storage medium or a storage device and the like. The memory can be arranged independently of the processor, or can be integrated with the processor. The embodiment of the present application does not limit this.

[0482] In the embodiments of the present application, the antenna and the radio frequency circuit with transceiving function can be regarded as a transceiving unit of the communication device, and the processor with processing function can be regarded as a processing unit of the communication device. As shown in FIG. 18, the communication device 1800 includes a transceiving unit 1810 and a processing unit 1820. The transceiving unit can also be referred to as a transceiver, a transceiving machine, a transceiving device, etc. The processing unit can also be referred to as a processor, a processing board, a processing module, a processing device, etc.

[0483] Optionally, the device for realizing the receiving function in the transceiving unit 1810 can be regarded as a receiving unit, and the device for realizing the sending function in the transceiving unit 1810 can be regarded as a sending unit, that is, the transceiving unit 1810 includes the receiving unit and the sending unit. The transceiving unit can also be referred to as a transceiver, a transceiving machine, or a transceiving circuit, etc. The receiving unit can also be referred to as a receiver, a receiver, or a receiving circuit, etc. The sending unit can also be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

[0484] It should be understood that the transceiving unit 1810 is configured to perform the sending operation and the receiving operation of the first access network device and / or the second access network device in the method embodiments, and the processing unit is configured to perform other operations of the first access network device and / or the second access network device in the method embodiments, in addition to the transceiving operation.

[0485] When the communication device is a chip, the chip includes the transceiving unit and the processing unit. The transceiving unit can be an input / output circuit or a communication interface, and the processing unit is a processor or a microprocessor integrated on the chip or an integrated circuit or a logic circuit. In the method embodiments, the sending operation corresponds to the output of the input / output circuit, and the receiving operation corresponds to the input of the input / output circuit.

[0486] Please refer to FIG. 19, which is a structural schematic diagram of a computing node provided by an embodiment of the present application. As shown in FIG. 19, the computing node provided by an embodiment of the present application is applied to a first computing node, and the computing node includes:

[0487] The transceiving module 1901 is configured to receive a second request, and the second request is used to indicate that a terminal device requesting to perform the computing task is switched from a second access network device to a first access network device.

[0488] The processing module 1902 is configured to perform the computing task in response to the second request.

[0489] In a possible implementation, the second request includes any one or more of the following information:

[0490] Context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device, and the routing information indicates a transmission path carrying the computing task.

[0491] In a possible implementation manner,

[0492] The transceiver 1901 is further configured to, in response to the second request, send a second response to the first access network device, where the second response indicates that the first computing node allows the execution of the computing task requested by the terminal device.

[0493] The processing module 1902 is further configured to establish a transmission path related to the terminal device with the first computing node.

[0494] In a possible implementation manner,

[0495] The transceiver 1901 is further configured to receive data related to the computing task from the first access network device.

[0496] Alternatively, the transceiver 1901 is further configured to receive the data related to the computing task from a second computing node, where the second computing node is an original computing node for executing the computing task.

[0497] In a possible implementation manner,

[0498] The transceiver 1901 is further configured to receive the second request from the first access network device.

[0499] Alternatively, the transceiver 1901 is further configured to receive the second request from a core network device, where the core network device sends the second request to the first computing node according to a third request sent by the first access network device, and the third request carries context information of the computing task requested by the terminal device to be executed.

[0500] In a possible implementation manner, the context information of the computing task includes any one or more of the following information:

[0501] identification information of the computing task,

[0502] identification information of an application, where the application is related to the computing task,

[0503] the routing information, where the routing information indicates a transmission path carrying the computing task,

[0504] identification information of a computing node supporting the execution of the computing task,

[0505] or first indication information, where the first indication information indicates that the computing task needs to be switched to the first access network device for continuous execution.

[0506] In a possible implementation, the context information of the computing task further includes any one or more of the following:

[0507] identification information of a neural network model related to the computing task,

[0508] type information of the neural network model,

[0509] parameter information of the neural network model,

[0510] key-value storage (KV Cache) information of the neural network model,

[0511] information of one or more intermediate layers included in the neural network model,

[0512] initial input feature information of the neural network model,

[0513] or output feature information of the one or more intermediate layers of the neural network model.

[0514] The transceiver module 1901 and the processing module 1902 described above can be implemented by software or by hardware. For example, the implementation of the processing module 1902 is described below. Similarly, the implementation of the transceiver module 1901 can refer to the implementation of the processing module 1902.

[0515] As an example of the software functional unit, the processing module 1902 can include code running on a computing instance. The computing instance can include at least one of a physical host (compute node), a virtual machine, and a container. Further, the computing instance can be one or more. For example, the processing module 1902 can include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code can be distributed in the same region, or can be distributed in different regions. Further, the multiple hosts / virtual machines / containers used to run the code can be distributed in the same availability zone (AZ), or can be distributed in different AZs, and each AZ includes one data center or multiple data centers with similar geographical locations. Generally, one region can include multiple AZs.

[0516] Likewise, the multiple hosts / virtual machines / containers used to run the code can be distributed in the same virtual private cloud (VPC) or in multiple VPCs. Among them, usually one VPC is set in one region, and a communication gateway needs to be set in each VPC for cross-region communication between two VPCs in the same region and between VPCs in different regions, and the interconnection between VPCs is realized through the communication gateway.

[0517] As an example of a hardware functional unit, the processing module 1902 can include at least one computing node, such as a server, etc. Alternatively, the processing module 1902 can also be a device implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), etc. Among them, the above-mentioned PLD can be implemented by a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0518] The multiple computing nodes included in the processing module 1902 can be distributed in the same region or in different regions. The multiple computing nodes included in the processing module 1902 can be distributed in the same AZ or in different AZs. Likewise, the multiple computing nodes included in the processing module 1902 can be distributed in the same VPC or in multiple VPCs. Among them, the multiple computing nodes can be any combination of servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0519] It should be noted that the information interaction and implementation process between the modules / units of the above device, since it is based on the same concept as the method embodiments of the present application, the technical effects brought about by it are the same as those of the method embodiments of the present application, and the specific content can be referred to the description of the method embodiments described above. Here, it will not be repeated.

[0520] The embodiment of the present application further provides a computing node 2000. Please refer to FIG. 20, which is a structural schematic diagram of the computing node 2000 provided by the embodiment of the present application. As shown in FIG. 20, the computing node 2000 comprises a bus 2002, a processor 2004, a memory 2006 and a communication interface 2008. The processor 2004, the memory 2006 and the communication interface 2008 communicate through the bus 2002. The computing node 2000 can be a server or a terminal device. It should be understood that the number of processors and memories in the computing node 2000 is not limited by the present application. The computing node 2000 can be used to realize the functions of the first computing node and / or the second computing node in the foregoing embodiments.

[0521] The bus 2002 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one line is shown in FIG. 20, but it does not mean that there is only one bus or only one type of bus. The bus 2002 can comprise a channel for transmitting information between various components (for example, the memory 2006, the processor 2004, the communication interface 2008) of the computing node 2000.

[0522] The processor 2004 can comprise any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP) processor, etc.

[0523] The memory 2006 can comprise a volatile memory (for example, a random access memory (RAM)). The processor 2004 can further comprise a non-volatile memory (for example, a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid state drive (SSD)). The memory 2006 is an optional component, that is, the computing node 2000 can not comprise the memory 2006.

[0524] The executable program code stored in the memory 2006 is executed by the processor 2004 to realize the functions of the foregoing functional modules respectively, so as to realize the computing migration method described above. That is, the instructions for executing the computing migration method are stored in the memory 2006.

[0525] The communication interface 2008 uses a transceiving module such as but not limited to a network interface card and a transceiver to realize the communication between the computing node 2000 and other devices or communication networks.

[0526] The first computing node in the embodiments of the present application can be implemented by a computing node cluster, and the second computing node in the embodiments of the present application can also be implemented by a computing node cluster. Therefore, the embodiments of the present application also provide a computing node cluster. The computing node cluster includes at least one computing node. The computing node can be a server, for example, a central server, an edge server, or a local server in a local data center. In some embodiments, the computing node can also be a terminal device such as a desktop computer, a notebook computer, or a smart phone.

[0527] Please refer to FIG. 21, which is a structural schematic diagram of a computing node cluster provided by the embodiments of the present application. As shown in FIG. 21, the computing node cluster includes at least one computing node 2000.

[0528] Optionally, the same instructions for executing the computing migration method can be stored in the memory 2006 of one or more computing nodes 2000 in the computing node cluster.

[0529] In some possible implementations, the memory 2006 of one or more computing nodes 2000 in the computing node cluster can also respectively store partial instructions for executing the computing migration method. In other words, the combination of one or more computing nodes 2000 can collectively execute the instructions for executing the computing migration method.

[0530] It should be noted that the memories 2006 in different computing nodes 2000 in the computing node cluster can store different instructions respectively for executing partial functions of the data processing apparatus. That is, the instructions stored in the memories 2006 in different computing nodes 2000 can realize the functions of one or more of the foregoing functional modules.

[0531] In some possible implementations, one or more of the computing nodes in the cluster of computing nodes can be connected through a network. The network can be a wide area network or a local area network, among others. FIG. 22 illustrates one possible implementation. FIG. 22 is a diagram of the structure of another cluster of computing nodes according to an embodiment of the present application. As shown in FIG. 22, in the cluster of computing nodes 2200, two computing nodes 2000A and 2000B are connected through a network. Specifically, the computing nodes are connected to the network through the communication interfaces in the computing nodes. In this type of possible implementation, the memory 2006 in the computing node 2000A stores instructions for performing the functions of the above-described functional modules. Meanwhile, the memory 2006 in the computing node 2000B stores instructions for performing the functions of the above-described functional modules.

[0532] It should be understood that the functions of the computing node 2000A shown in FIG. 22 can also be performed by multiple computing nodes 2000. Similarly, the functions of the computing node 2000B can also be performed by multiple computing nodes 2000.

[0533] Referring to FIG. 23, FIG. 23 is a diagram of the structure of a computer-readable storage medium according to an embodiment of the present application. The present application also provides a computer-readable storage medium. In some embodiments, the workflow performed by the above-described first computing node and / or second computing node can be implemented as computer program instructions encoded in a machine-readable format on a computer-readable storage medium or on other non-transitory media or articles of manufacture.

[0534] FIG. 23 schematically illustrates a conceptual partial view of an example computer-readable storage medium arranged in accordance with at least some embodiments presented herein, the example computer-readable storage medium including a computer program for executing a computer process on a computing node.

[0535] In one embodiment, the computer-readable storage medium 2300 is provided using a signal-bearing medium 2301. The signal-bearing medium 2301 can include one or more program instructions 2302, which when executed by one or more processors can provide the functionality or some portion of the functionality described above with respect to the above-described first computing node and / or second computing node.

[0536] In some examples, the signal-bearing medium 2301 can include a computer- readable medium 2303, such as but not limited to, a hard disk drive, a compact disc (CD), a digital video disc (DVD), a digital tape, memory, ROM, or RAM, among others.

[0537] In some embodiments, the signal bearing medium 2301 can comprise computer- readable storage media 2304, which comprises, for example, a storage device such as a hard disk drive, a floppy disk drive, a CD-ROM drive, a DVD-ROM drive, a Blu-ray drive, a magnetic tape drive, a flash memory, or a solid-state memory. In some embodiments, the signal bearing medium 2301 can comprise communication media 2305, which comprises, for example, a wireless link, optical link, or a wired link that is implemented with one or more communication protocols. Thus, for example, the signal bearing medium 2301 can be implemented with a wireless form of communication media 2305, such as is the case with wireless communication media 2305 that is compliant with IEEE 802.X standards or other transmission protocols.

[0538] The one or more program instructions 2302 can be, for example, computer-executable instructions or logic-implementing instructions. In some examples, a computing node of the computing nodes can be configured to provide various operations, functions, or actions in response to the program instructions 2302 being conveyed to the computing node by one or more of the computer-readable media 2303, the computer-recordable media 2304, and / or the communication media 2305.

[0539] The present application also provides a communication system, comprising a first access network device, a second access network device, a first computing node and / or a second computing node.

[0540] Embodiments of the present application also provide a computer program product comprising computer instructions, which, when executed on a computer, cause the computer to perform the method of the embodiments shown in FIG. 9 and FIG. 15.

[0541] Embodiments of the present application also provide a computer-readable storage medium comprising computer instructions, which, when executed on a computer, cause the computer to perform the method of the embodiments shown in FIG. 9 and FIG. 15.

[0542] Embodiments of the present application also provide a chip device comprising a processor, configured to invoke a computer program or computer instructions stored in a memory to cause the processor to perform the method of the embodiments shown in FIG. 9 and FIG. 15.

[0543] Optionally, the processor is coupled to the memory through an interface.

[0544] Optionally, the chip device further comprises the memory, and the memory stores the computer program or computer instructions.

[0545] The processor mentioned in any of the above can be a general central processing unit, a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the method of the embodiments shown in FIG. 9 and FIG. 15. The memory mentioned in any of the above can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), and the like.

[0546] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0547] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0548] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0549] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the essential part of the technical solutions of the present application or the whole or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0550] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of computing migration, characterized by, The method is applied to a first access network device, and the method comprises: receiving a first request from a second access network device, the first request indicating that a terminal device requests to switch from the second access network device to the first access network device, the first request comprising context information of a computing task requested to be executed by the terminal device; in response to the first request, sending a second request to a first computing node, the second request being used to indicate that the terminal device requesting to execute the computing task switches to the first access network device.

2. The method of claim 1, wherein, The second request comprises any one or more of the following information: context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device, the routing information indicating a transmission path carrying the computing task.

3. The method according to claim 1 or 2, characterized in that, In response to the first request, sending the second request to the first computing node comprises: in response to the first request, determining the first computing node; sending the second request to the first computing node; receiving a second response from the first computing node, the second response indicating that the first computing node allows the computing task requested to be executed by the terminal device to be executed; in response to the second response, establishing a transmission path related to the terminal device with the first computing node.

4. The method of claim 3, wherein, The method further comprises: sending second indication information to the second access network device, the second indication information being used to indicate that a second computing node migrates the computing task to the first computing node, the second computing node being an original computing node for executing the computing task; receiving data related to the computing task from the second access network device.

5. The method according to claim 1 or 2, characterized in that, In response to the first request, sending the second request to the first computing node comprises: sending a third request to a core network device, the third request carrying context information of the computing task requested to be executed by the terminal device, so that the core network device sends the second request to the first computing node, the second computing node being an original computing node for executing the computing task; receiving a third response from the core network device, the third response carrying second indication information, the second indication information being used to indicate that a second computing node migrates the computing task to the first computing node, the second computing node being an original computing node for executing the computing task; in response to the third response, establishing a transmission path related to the terminal device with the first computing node.

6. The method according to claim 4 or 5, characterized in that, The second indication information comprises: routing information, context information of the computing task, and / or related information of the first computing node, the routing information indicating a transmission path carrying the computing task.

7. The method according to any one of claims 1 to 6, characterized in that, The context information of the computing task comprises any one or more of the following information: identification information of the computing task, identification information of an application, the application being an application related to the computing task, the routing information, the routing information indicating a transmission path carrying the computing task, identification information of a computing node supporting execution of the computing task, Or, the first indication information indicates that the computing task needs to be switched to the first access network device to continue execution.

8. The method of claim 7, wherein, The context information of the computing task further includes any one or more of the following: identification information of a neural network model related to the computing task, type information of the neural network model, parameter information of the neural network model, key-value storage (KV Cache) information of the neural network model, information of one or more intermediate layers included in the neural network model, initial input feature information of the neural network model, or output feature information of the one or more intermediate layers of the neural network model.

9. A method of computing migration, characterized by, The method is applied to a first computing node, and the method includes: receiving a second request for indicating that a terminal device requesting execution of a computing task is switched from a second access network device to a first access network device; in response to the second request, executing the computing task.

10. The method of claim 9, wherein, The second request includes any one or more of the following information: context information of the computing task, address information of the first access network device, routing information, or address information of the terminal device, the routing information indicating a transmission path carrying the computing task.

11. The method according to claim 9 or 10, characterized in that, The method further includes: in response to the second request, sending a second response to the first access network device, the second response indicating that the first computing node allows execution of the computing task requested by the terminal device; establishing a transmission path related to the terminal device with the first computing node.

12. The method according to any one of claims 9-11, characterized in that, The method further includes: receiving data related to the computing task from the first access network device; or receiving data related to the computing task from a second computing node, the second computing node being an original computing node for executing the computing task.

13. The method according to any one of claims 9-12, characterized in that, Receiving the second request includes: receiving the second request from the first access network device; or receiving the second request from a core network device, wherein the core network device sends the second request to the first computing node according to a third request sent by the first access network device, and the third request carries context information of the computing task requested by the terminal device.

14. The method according to any one of claims 10-13, characterized in that, The context information of the computing task includes any one or more of the following information: identification information of the computing task, identification information of an application related to the computing task, routing information indicating a transmission path carrying the computing task, identification information of a computing node supporting execution of the computing task, or first indication information indicating that the computing task needs to be switched to the first access network device to continue execution.

15. The method of claim 14, wherein, The context information of the computing task further includes any one or more of the following: identification information of a neural network model related to the computing task, type information of the neural network model, parameter information of the neural network model, key-value storage (KV Cache) information of the neural network model, information of one or more intermediate layers included in the neural network model, initial input feature information of the neural network model, or output feature information of the one or more intermediate layers of the neural network model. initial input feature information of the neural network model, or output feature information of the one or more intermediate layers of the neural network model.

16. A communications device, characterized by The communication device comprises a module for performing the method of any of claims 1 to 8.

17. A communications device, characterized by The communication device comprises a processor for executing a computer program or computer instructions in a memory to perform the method of any of claims 1 to 8.

18. A computing node, characterized in that, The computing node comprises a module for performing the method of any of claims 9 to 15.

19. A computing node, characterized in that, The computing node comprises a processor for executing a computer program or computer instructions in a memory to perform the method of any of claims 9 to 15.

20. A computer program product, characterised in that, The computer program product, when running on a computer, causes the computer to perform the method of any of claims 1 to 8, or 9 to 15.

21. A computer-readable storage medium, characterized in that, A computer program product having stored thereon a computer program which, when executed by a device, causes the device to perform the method of any of claims 1 to 8, or 9 to 15.

Images