Computing node switching method, apparatus, and device

Abstract

The present application relates to the technical field of communications, and discloses a computing node switching method, an apparatus, and a device. The computing node switching method in embodiments of the present application comprises: a first node receives first information, wherein the first information comprises at least one of measurement information and a computing service request; and on the basis of the first information, the first node determines whether to switch a computing node used for executing a computing task.

Description

Computing node switching methods, devices and equipment

[0001] Cross-reference to related applications

[0002] This application is based on and claims priority to Chinese Patent Application No. 202410909466.7, filed on July 8, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of communication technology, and more specifically, to a method, apparatus, and device for switching computing nodes. Background Technology

[0004] With the development of artificial intelligence technology, an increasing number of application scenarios require the simultaneous application of computing and communication. Examples include assisted autonomous driving, autonomous collaboration between devices in medical assistance applications, cross-device / network computing offloading, the creation and prediction of digital twins, and the implementation of collaborative robots. Mobile network communication systems supporting computing services include computing nodes. Terminals can act as sending nodes, receiving nodes, or computing nodes for computing services, while network-side devices serve the terminals and participate in the computing services. In related technologies, after a computing node is determined, it executes the computing task until the task is completed. However, during the execution of the computing task, the computing node may fail to complete the task effectively (e.g., the quality of the computing service provided by the node may no longer meet the requirements), resulting in poor computing service performance. Summary of the Invention

[0005] This application provides a computing node switching method, apparatus, and device that can solve the problem of poor computing service performance.

[0006] Firstly, a method for switching computing nodes is provided, the method comprising:

[0007] The first node receives first information, which includes at least one of measurement information and a computing service request.

[0008] The first node determines whether to switch the computing node used to perform the computing task based on the first information.

[0009] Secondly, a method for switching computing nodes is provided, the method comprising:

[0010] The terminal sends the first measurement information or computing service request to the first node;

[0011] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0012] Thirdly, a method for switching computing nodes is provided, the method comprising:

[0013] The first access network device sends the second measurement information to the first node. The first access network device is the service access network device of the terminal associated with the computing task.

[0014] The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0015] Fourthly, a method for switching computing nodes is provided, which includes:

[0016] The source computing node sends the third measurement information to the first node;

[0017] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0018] Fifthly, a method for switching computing nodes is provided, which includes:

[0019] The candidate target node receives a first switching request sent by the first node. The first switching request is used to request a switch of computing nodes and includes the identifier of the computing task.

[0020] Sixthly, a method for switching computing nodes is provided, the method comprising:

[0021] The second access network device receives a second handover request sent by the first node. The second handover request is used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

[0022] In a seventh aspect, a computing node switching device is provided, comprising:

[0023] A receiving module is configured to receive first information, the first information including at least one of measurement information and a computing service request;

[0024] The processing module is used to determine whether to switch the computing node used to perform the computing task based on the first information.

[0025] Eighthly, a computing node switching device is provided, comprising:

[0026] The sending module is used to send the first measurement information or calculation service request to the first node;

[0027] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0028] Ninthly, a computing node switching device is provided, comprising:

[0029] The sending module is used to send the second measurement information to the first node;

[0030] The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0031] In a tenth aspect, a computing node switching device is provided, comprising:

[0032] The sending module is used to send third measurement information to the first node;

[0033] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0034] Eleventhly, a computing node switching device is provided, comprising:

[0035] The receiving module is used to receive a first switching request sent by the first node. The first switching request is used to request a switch of computing nodes, and the first switching request includes the identifier of the computing task.

[0036] In a twelfth aspect, a computing node switching device is provided, comprising:

[0037] The receiving module is used to receive a second handover request sent by the first node, the second handover request being used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

[0038] In a thirteenth aspect, a computing node switching apparatus is provided, the apparatus being configured to perform the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect, or implement the steps of the method described in the third aspect, or implement the steps of the method described in the fourth aspect, or implement the steps of the method described in the fifth aspect, or implement the steps of the method described in the sixth aspect.

[0039] In a fourteenth aspect, an electronic device is provided, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the methods described in the first, second, third, fourth, fifth, or sixth aspects.

[0040] In the fifteenth aspect, a first node is provided, including a processor and a communication interface, wherein...

[0041] A communication interface for receiving first information, the first information including at least one of measurement information and a computing service request;

[0042] The processor is configured to determine, based on the first information, whether to switch the computing node used to perform the computing task.

[0043] In a sixteenth aspect, a terminal is provided, including a processor and a communication interface, wherein...

[0044] The communication interface is used to send the first measurement information or calculation service request to the first node;

[0045] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0046] In a seventeenth aspect, a first access network device is provided, including a processor and a communication interface, wherein...

[0047] A communication interface is used to send the second measurement information to the first node;

[0048] The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0049] In the eighteenth aspect, a source computing node is provided, including a processor and a communication interface, wherein...

[0050] The communication interface is used to send third measurement information to the first node;

[0051] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0052] In the nineteenth aspect, a candidate target node is provided, including a processor and a communication interface, wherein...

[0053] A communication interface is used to receive a first switching request sent by a first node. The first switching request is used to request a switch of computing nodes, and the first switching request includes an identifier of the computing task.

[0054] In a twentieth aspect, a second access network device is provided, including a processor and a communication interface, wherein...

[0055] A communication interface is used to receive a second handover request sent by the first node, the second handover request being used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

[0056] In a twenty-first aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or the steps of the method described in the second aspect, or the steps of the method described in the third aspect, or the steps of the method described in the fourth aspect, or the steps of the method described in the fifth aspect, or the steps of the method described in the sixth aspect.

[0057] In a twenty-second aspect, a wireless communication system is provided, comprising: a first node, a terminal, a first access network device, a source computing node, and a candidate target node; or, the wireless communication system comprises: a first node, a terminal, a first access network device, a source computing node, a candidate target node, and a second access network device; or the wireless communication system comprises: a first node, a terminal, and a second access network device; wherein the first node is available to implement the steps of the method described in the first aspect, the terminal is available to implement the steps of the method described in the second aspect, the first access network device is available to implement the steps of the method described in the third aspect, the source computing node is available to implement the steps of the method described in the fourth aspect, the candidate target node is available to implement the steps of the method described in the fifth aspect, and the second access network device is available to implement the steps of the method described in the sixth aspect.

[0058] In a twentieth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run programs or instructions to implement the method as described in the first aspect, or the method as described in the second aspect, or the method as described in the third aspect, or the method as described in the fourth aspect, or the method as described in the fifth aspect, or the method as described in the sixth aspect.

[0059] In a twentieth aspect, a computer program / program product is provided, the computer program / program product being stored in a storage medium, the computer program / program product being executed by at least one processor to implement the method as described in the first aspect, or the method as described in the second aspect, or the method as described in the third aspect, or the method as described in the fourth aspect, or the method as described in the fifth aspect, or the method as described in the sixth aspect.

[0060] In this embodiment, a first node receives first information, which includes at least one of measurement information and a computing service request. Based on the first information, the first node determines whether to switch the computing node used to perform the computing task. Thus, by determining whether to switch the computing node using the first information, the decision to switch the computing node can be made while the computing node is performing the computing task, taking into account either measurement information or a computing service request, thereby improving the computing service performance. Attached Figure Description

[0061] Figure 1 is a block diagram of a wireless communication system applicable to an embodiment of this application;

[0062] Figure 2 is a switching flowchart in a related technology;

[0063] Figure 3 is a flowchart of one of the computing node switching methods provided in an embodiment of this application;

[0064] Figure 4 is a second flowchart of a computing node switching method provided in an embodiment of this application;

[0065] Figure 5 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0066] Figure 6 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0067] Figure 7 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0068] Figure 8 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0069] Figure 9 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0070] Figure 10 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0071] Figure 11 is a flowchart of a computing node switching method provided in an embodiment of this application;

[0072] Figure 12 is a schematic diagram of one of the structures of a computing node switching device provided in an embodiment of this application;

[0073] Figure 13 is a second structural schematic diagram of a computing node switching device provided in an embodiment of this application;

[0074] Figure 14 is a third structural schematic diagram of a computing node switching device provided in an embodiment of this application;

[0075] Figure 15 is a fourth structural schematic diagram of a computing node switching device provided in an embodiment of this application;

[0076] Figure 16 is a fifth structural schematic diagram of a computing node switching device provided in an embodiment of this application;

[0077] Figure 17 is a sixth schematic diagram of a computing node switching device provided in an embodiment of this application;

[0078] Figure 18 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0079] Figure 19 is a schematic diagram of the structure of a terminal provided in an embodiment of this application;

[0080] Figure 20 is one of the structural schematic diagrams of a network-side device provided in an embodiment of this application;

[0081] Figure 21 is a second schematic diagram of the structure of a network-side device provided in an embodiment of this application. Detailed Implementation

[0082] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0083] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, the scope of protection for "A or B" covers at least three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. In addition, the terms "A and / or B," "at least one of A and B," and "at least one of A or B" also cover at least the above three scenarios. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0084] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as the sender explicitly informing the receiver of specific information, the required operation, or the requested result in the instruction sent. An indirect instruction can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the required operation or requested result based on the judgment result.

[0085] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used with the systems and radio technologies mentioned above, as well as with other systems and radio technologies. The following description describes New Radio (NR) systems for illustrative purposes, and the term NR is used in most of the following description; however, these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) radio systems. th Generation 6G communication system.

[0086] Figure 1 shows a block diagram of a wireless communication system applicable to an embodiment of this application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can also be referred to as User Equipment (UE), and can be a mobile phone, tablet computer, laptop computer, notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR), virtual reality (VR) device, robot, wearable device, flight vehicle, vehicle user equipment (VUE), shipboard equipment, pedestrian user equipment (PUE), smart home (home devices with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game console, personal computer (PC), ATM, or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among these, in-vehicle devices can also be referred to as in-vehicle terminals, in-vehicle controllers, in-vehicle modules, in-vehicle components, in-vehicle chips, or in-vehicle units, etc. It should be noted that the specific type of terminal 11 is not limited in this application embodiment. Network-side equipment 12 may include access network equipment or core network equipment, wherein access network equipment may also be referred to as Radio Access Network (RAN) equipment, radio access network function, or radio access network unit. Access network equipment may include base stations, Wireless Local Area Network (WLAN) access points (APs), or Wireless Fidelity (WiFi) nodes, etc.Among them, base stations can be referred to as Node B (NB), Evolved Node B (eNB), Next Generation Node B (gNB), New Radio Node B (NR Node B), Access Point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), Radio Base Station, Radio Transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B (HNB), Home Evolved Node B, Transmit / Receive Point (TRP), Non-Terrestrial Network (NTN) equipment (such as satellite or high altitude platform stations). The term "base station" can be any suitable term in the field, such as "station" or any other appropriate term in the relevant field, as long as the same technical effect is achieved. The term "base station" is not limited to specific technical terms. It should be noted that the embodiments of this application only use the base station in the NR system as an example for introduction, and do not limit the specific type of base station.

[0087] Core network equipment, also known as core network nodes, core network functions, or core network elements, includes, but is not limited to, at least one of the following: Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (or L-NEF), and Binding Support. Functions include BSF, Application Function (AF), Location Management Function (LMF), Gateway Mobile Location Centre (GMLC), Network Data Analytics Function (NWDAF), and Non-Terrestrial Network (NTN) equipment (such as satellite or high altitude platform station).It should be noted that the embodiments of this application only use the core network equipment in the NR system as an example for introduction, and do not limit the specific type of core network equipment. If the name of the core network equipment mentioned in the embodiments of this application changes in subsequent protocol versions (e.g., 6G), it is also within the scope of protection of this application.

[0088] Optionally, the core network equipment can be implemented by one or more functional modules in a single device, or by multiple devices working together; this application does not specifically limit this. It is understood that the aforementioned functional modules can be network elements in hardware devices, software functional modules running on dedicated hardware, or virtualized functional modules instantiated on a platform (e.g., a cloud platform).

[0089] For ease of understanding, the following explains some aspects of the embodiments of this application:

[0090] 1. Switch

[0091] Handover is triggered by the movement of User Equipment (UE, i.e., terminal) while in a connected state. The basic goal of handover is:

[0092] It instructs the UE to communicate with a cell that has better channel quality than the current serving cell; it provides the UE with continuous, uninterrupted communication service, effectively preventing dropped calls due to deteriorating cell signal quality. The handover process in 5G includes the following steps:

[0093] (1) Triggering Measurement: After the UE completes access or successfully hands over, the gNodeB will send measurement control information to the UE via RRC Connection Reconfiguration. Furthermore, if the measurement configuration information is updated, the gNodeB will also send updated measurement control information via Radio Resource Control (RRC) Connection Reconfiguration messages. The most important information in the measurement control message includes the measurement object, Measurement Report (MR) configuration, and measurement events.

[0094] The measurement events defined by 3GPP for NR (with the addition of A6 compared to LTE) are shown in Table 1.

[0095] Table 1

[0096] The specific criteria for determining 5G measurement events are shown in Table 2.

[0097] Table 2

[0098] in:

[0099] Ms represents the measurement results of the serving cell;

[0100] Mn represents the measurement results of the neighboring cell;

[0101] TimeToTrig represents the duration for which the event entry condition is continuously met, i.e., time delay;

[0102] Off indicates the bias of the measurement result, with a step size of 0.5 dB;

[0103] Hys indicates the amplitude hysteresis of the measurement result, with a step size of 0.5 dB;

[0104] Ofs represents the frequency offset of the serving cell;

[0105] Ofn represents the frequency offset of the neighboring cell;

[0106] Ocs stands for Cell Individual Offset (CIO).

[0107] Ocn represents the cell-specific bias CIO of neighboring cells within the system;

[0108] Thresh corresponds to the threshold value configured for the event.

[0109] The measured event values ​​are shown in Table 3.

[0110] Table 3

[0111] The event value range in 5G NR differs from that in LTE. Range corresponds to the value reported in the measurement report, while Value corresponds to its actual value. Taking Reference Signal Received Power (RSRP) as an example, the Value range in LTE is -140 to -44 dBm, while in NR it is -156 to -31 dBm. NR allows for higher and lower receive levels. The actual value in LTE is -140 (MR report value), while in NR it is -156 (MR report value). Assuming the RSRP reported in MR is 50, the actual value is 50 - 140 = -90 dBm. However, the actual RSRP value in NR is 50 - 156 = -106 dBm.

[0112] (2) Perform measurement: According to the relevant configuration of measurement control, the UE monitors the radio channel, but when the measurement reporting conditions (A1-A6, B1 and B2) are met, it reports to the gNB through an event. The number of measurement reports / events can be triggered by RSRP, Reference Signal Received Quality (RSRQ) or Signal-to-noise and interference ratio (SINR).

[0113] (3) Target decision: The gNB selects the handover cell based on measurement as the basic resource and follows the first-report-first-process method, and selects the corresponding handover strategy (such as handover and redirection).

[0114] (4) Handover execution: The original base station applies for and allocates resources to the target base station, and then the source gNodeB makes a handover execution decision and sends the handover command to the UE. The UE performs the handover and data forwarding.

[0115] The handover process defined in 3GPP TS 38300 is shown in Figure 2.

[0116] 2. Artificial Intelligence (AI) and Communication Application Scenarios

[0117] AI and communications are among the 6G application scenarios outlined in ITU-R. Typical use cases include IMT-2030 (6G) assisted autonomous driving, autonomous collaboration between devices in medical assistance applications, cross-device / network computing offloading, creation and prediction of digital twins, and IMT-2030 (6G) assisted collaborative robots.

[0118] These application scenarios will require support for high mobile network capacity and high user experience data rates, as well as low latency and high reliability. Beyond communications, this application scenario is expected to include a suite of new functionalities integrating artificial intelligence and computing into 6G systems, including data acquisition, preparation, and processing from various sources; distributed AI model training; model sharing and distributed inference across mobile communication systems; and computing resource orchestration.

[0119] In related technologies, cloud services mainly define the static, long-term (e.g., monthly) service quality between cloud service providers and users through service level agreements (SLAs). Users are mainly enterprise users, with fewer direct consumer-facing users, especially mobile terminal consumers.

[0120] 3. Event configuration for traditional switching

[0121] The event configuration for traditional switching is shown in Table 4.

[0122] Table 4

[0123] Taking event A3 as an example, the meanings of the parameters in the entry and exit conditions are as follows:

[0124] Mn: Neighboring cell measurement results, without considering any offset.

[0125] Ofn: Specific offset of the neighboring measurement object

[0126] Ocn: Neighboring cell-level specific offset

[0127] Mp: Measurement results for Special Cells (SpCells) (such as the primary serving cell), without considering any offset.

[0128] Ofp: SpCell measures a specific offset of an object.

[0129] Ocp: SpCell cell-level specific offset

[0130] Hys: The hysteresis parameter of the event

[0131] Off: The offset parameter for the event.

[0132] To avoid ping-pong handover, the base station CondTriggerConfig configures a timeToTrigger parameter for each event. When the Layer 3 (L3) filtered signal quality of one or more candidate cells meets the event entry conditions within the timeToTrigger time, the UE will use the cell that meets the conditions as the trigger cell and select an execution condition reconfiguration in the trigger cell.

[0133] The following description, in conjunction with the accompanying drawings, details the computing node switching method, apparatus, and related equipment provided in this application through some embodiments and application scenarios.

[0134] Referring to Figure 3, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0135] Step 101: The first node receives first information, which includes at least one of measurement information and a computing service request;

[0136] Step 102: The first node determines whether to switch the computing node used to perform the computing task based on the first information.

[0137] The first node can be a network-side device. For example, the first node can be a newly added network function node, an enhanced AMF, an enhanced SMF, or an enhanced base station, etc. It is understood that "enhancement" here can refer to adding or enhancing computing-related functions to the entity of the relevant 5G network-side device. It should be noted that this embodiment does not limit the name of the first node. For example, the first node can be called a computing management node, computing management function, computing control function, computing management and control function, computing service control function, or computing service management function, etc. The first node can be a radio access network node or a core network node. The first node is a network node responsible for receiving computing service requests or processing computing service requests, computing resource scheduling, computing information interaction, computing data processing, etc., at least one of these functions. The first node can be an enhanced AMF or an enhanced SMF, or it can be an enhanced other network node or a newly defined network node.

[0138] In the embodiments of this application, a computing task can refer to one or a group of computing services initiated by a computing request node. Specifically, a computing task can be mapped to a Quality of Service (QoS) flow, or a computing task can be mapped to multiple QoS flows, or multiple computing tasks can be mapped to a single QoS flow. Alternatively, a computing task can be mapped to a radio bearer or physical layer resource.

[0139] The first node's determination of whether to switch the computing node used to perform the computing task based on the first information can also be described as follows: the first node determines whether to switch the computing node used to provide computing services based on the first information; or, the first node determines whether to switch the service computing node of the terminal associated with the computing task based on the first information; or, the first node determines whether to switch the service computing node of the terminal associated with the computing service based on the first information. The computing task is associated with the computing service.

[0140] In addition, computing tasks can also be described as AI tasks, and computing services can also be described as AI services.

[0141] It should be noted that the terminal associated with a computing task (or computing service) can refer to the terminal acting as either a sending node or a receiving node for the computing task (or computing service). For example, the terminal sends data for a computing task (or computing service), which is then transmitted over the network and processed, and the application function receives the data; or, the application function sends data for a computing task (or computing service), which is then transmitted over the network and processed, and the terminal receives the data; or, the terminal sends data for a computing task (or computing service), which is then transmitted over the network and processed, and another terminal receives the data; or, the terminal sends data for a computing task (or computing service), which is then transmitted over the network and processed, and the terminal receives the data.

[0142] The measurement information can be included in the measurement report.

[0143] In one implementation, taking the first node as a computing management node as an example, the computing management node sends first handover measurement configuration information to the terminal, and the computing management node receives the first measurement information sent by the terminal; or

[0144] The computing management node sends second handover measurement configuration information to the first access network device, and the computing management node receives the second measurement information sent by the first access network device; or

[0145] The computing management node sends third switching measurement configuration information to the source computing node, and the computing management node receives the third measurement information sent by the source computing node;

[0146] The computing management node determines whether to switch the computing node used to perform the computing task based on at least one of the first measurement information, the second measurement information, and the third measurement information.

[0147] The first measurement information can be carried through the first measurement report, the second measurement information can be carried through the second measurement report, and the third measurement information can be carried through the third measurement report.

[0148] It should be noted that the first node can decide whether to switch the computing node used to perform the computing task based on the measurement information. The first node may decide to switch the computing node used to perform the computing task when the computing node representing the measurement information cannot perform the computing task well; or, the first node may decide to switch the computing node used to perform the computing task when the computing node representing the measurement information cannot meet the performance requirements of the computing task; or, the first node may decide to switch the computing node used to perform the computing task when the computing node representing the measurement information cannot achieve the required quality of service; etc., this embodiment does not limit this.

[0149] Furthermore, the first node can decide whether to switch the computing node used to perform the computing task based on the computing service request. The first node may decide to switch the computing node used to perform the computing task when the source computing node cannot adequately fulfill the computing service request; or, the first node may decide to switch the computing node used to perform the computing task when the source computing node cannot satisfy the computing service request; or, the first node may decide to switch the computing node used to perform the computing task when the source computing node cannot achieve the computing service quality required by the computing service request; and so on. This embodiment does not limit this to any specific instance.

[0150] In related technologies, communication handover is used to ensure the continuity of communication services in order to address terminal mobility. AI and communication are among the application scenarios of 6G, but current mobile network protocols do not yet support the continuity of AI services or the continuity of AI and communication. Therefore, if 6G is to provide AI services or AI and communication services, a solution that ensures the continuity of these services is required.

[0151] In addition, the AI ​​and communication scenarios proposed by the International Telecommunication Union (ITU) may also include non-AI computing, so computing services can include both AI services and non-AI computing.

[0152] Potential computing service flows include on-demand compute offloading from mobile terminals and on-demand on-network compute from mobile terminals. On-demand compute offloading refers to the mobile terminal determining whether to offload computation to the network for processing based on local computing power, computation task requirements, power consumption, etc. If compute offloading is required, the data to be computed is sent to the network, and the computation result is obtained from the network. The performance of on-demand compute offloading is related to the performance of computing nodes, uplink transmission performance, downlink transmission performance, etc. On-demand on-network compute refers to providing computation services on demand for transmitted data during network transmission. Furthermore, based on the sending and receiving nodes that can transmit data, the scenarios can be divided as follows:

[0153] (1) The mobile terminal sends data, which is then transmitted and processed by the network, and the application functions receive the data.

[0154] (2) The application sends data, which is then transmitted over the network and processed by calculation before the mobile terminal receives the data.

[0155] (3) The mobile terminal sends data, which is then transmitted and processed by the network, and then another mobile terminal receives the data.

[0156] The performance of scenario (1) is related to the performance of computing nodes and uplink transmission performance; the performance of scenario (2) is related to the performance of computing nodes and downlink transmission performance; the performance of scenario (3) is related to the performance of computing nodes, the uplink transmission performance of the sending data terminal, and the downlink transmission performance of the receiving data terminal.

[0157] In computing and communication services, a switchover of the base station and / or computing node involved in computing and communication occurs due to one of the following reasons:

[0158] Changes in UE mobility or channel quality increase communication latency in computation and communication service processes, thus requiring the reduction of computation latency and leading to the switching of computation nodes.

[0159] Changes in computing service performance experienced by computing request / receive nodes (such as UEs), and the inability of computing service quality provided by computing service nodes to meet demand, leading to the switching of computing nodes participating in computing and communication service processes;

[0160] Load balancing, compute node failure, compute node power consumption, and other reasons related to compute management nodes or compute nodes can lead to the switching of compute nodes involved in the compute and communication service process.

[0161] Wherein, UE is the sending node and / or receiving node, or computing node, of the computing service. The base station is the serving base station of the UE.

[0162] In related technologies, there is currently no mature method for the switching process of computing nodes involved in computing and communication service processes.

[0163] In this embodiment, the computing service request can be used to determine the computing node or base station handover caused by changes in computing service performance experienced by the computing requesting node or computing receiving node (such as UE). The measurement information corresponding to the first handover measurement configuration information and the second handover measurement configuration information can be used to determine the computing node or base station handover caused by changes in terminal mobility or channel quality, and the computing node or base station handover caused by changes in computing service performance experienced by the terminal; the measurement information corresponding to the third handover measurement configuration can be used to determine the computing node or base station handover caused by reasons related to the computing node.

[0164] In this embodiment of the application, in a mobile network communication system that supports computing services, a terminal can act as a sending node, receiving node, or computing node for computing services. The first information is used to determine whether to switch the computing node used to perform computing tasks. Thus, when the first terminal acts as a sending node or receiving node for computing services, it can support switching the computing node connected to the first terminal through the service access network device; or, when the second terminal acts as a computing node, it can support switching the second terminal connected to the first terminal through the service access network device. This embodiment of the application can solve the terminal switching problem in computing and communication services.

[0165] In this embodiment, a first node receives first information, which includes at least one of measurement information and a computing service request. Based on the first information, the first node determines whether to switch the computing node used to perform the computing task. Thus, by determining whether to switch the computing node using the first information, the decision to switch the computing node can be made while the computing node is performing the computing task, taking into account either measurement information or a computing service request, thereby improving the computing service performance.

[0166] Optionally, before the first node receives the first information, the method further includes at least one of the following:

[0167] The first node sends first handover measurement configuration information to the terminal, wherein the measurement information includes first measurement information corresponding to the first handover measurement configuration information;

[0168] The first node sends second handover measurement configuration information to the first access network device, wherein the first access network device is the service access network device of the terminal associated with the computing task, and the measurement information includes second measurement information corresponding to the second handover measurement configuration information;

[0169] The first node sends third handover measurement configuration information to the source computing node, wherein the measurement information includes third measurement information corresponding to the third handover measurement configuration information.

[0170] The first access network device can be the current serving access network device of the terminal associated with the computing task. The serving access network device of the terminal can be an access network device that serves the terminal, through which the terminal accesses the network.

[0171] The first measurement information may include: the uplink bandwidth or uplink bit rate corresponding to the computing task, the downlink bandwidth or downlink bit rate corresponding to the computing task, the location of the terminal, the AI ​​performance of the AI ​​model used for the computing task, the inference throughput of the AI ​​model used for the computing task, the remaining duration of the computing task, the end-to-end latency of the computing task, the computing failure rate of the computing task, the computing speed of the computing task, or the computing intensity of the computing task, etc. The first measurement information can be obtained by the terminal through measurement and can be carried in a first measurement report. This first measurement report can be an event-triggered report or a periodic report. For example, the first measurement report can be triggered by an event-triggered report.

[0172] The second measurement information may include: the uplink bandwidth or uplink bit rate corresponding to the computing task, the downlink bandwidth or downlink bit rate corresponding to the computing task, or the uplink channel quality of the terminal, etc. The second measurement information can be obtained by the first access network device and can be carried in a second measurement report. This second measurement report can be an event-triggered report or a periodic report. For example, the second measurement report can be triggered by an event.

[0173] The third measurement information may include: the computation failure rate of the computing task, the computation power consumption of the computing task, the computation energy efficiency of the computing task, the computation resource utilization rate of the computing task, the computation speed of the computing task, the computation efficiency of the computing task, the computation intensity of the computing task, the computation latency of the computing task, and the load balancing or failure of computing nodes. The third measurement information can be obtained by the source computing node and can be carried in a third measurement report, which can be an event-triggered report or a periodic report. For example, the third measurement report can be triggered by an event.

[0174] In this embodiment, the decision to switch the computing node used to perform the computing task is made by using at least one of the first measurement information corresponding to the first switching measurement configuration information, the second measurement information corresponding to the second switching measurement configuration information, and the third measurement information corresponding to the third switching measurement configuration information. This allows for the determination of whether the computing node can perform the computing task well during the execution of the computing task, and the decision to switch the computing node, thereby improving the computing service effect.

[0175] Optionally, the first switching measurement configuration information includes at least one of the following: a first measurement object; a first measurement report configuration; a first measurement report triggering event; and related parameters of the first measurement report triggering event.

[0176] or

[0177] The second switching measurement configuration information includes at least one of the following: a second measurement object; a second measurement report configuration; a second measurement report trigger event; and related parameters of the second measurement report trigger event.

[0178] or

[0179] The third switching measurement configuration information includes at least one of the following: third measurement object; third measurement report configuration; third measurement report trigger event; and related parameters of the third measurement report trigger event.

[0180] The first measurement object can be represented by a computation task identifier, which identifies one or more computation tasks transmitted through the source cell that the UE needs to measure. If the computation task corresponds to a QoS flow, the first measurement object can also be represented by a QoS flow identifier (ID). If the computation task corresponds to a radio bearer, the first measurement object can also be represented by a radio bearer ID.

[0181] The second measurement object is represented by a computation task identifier, which identifies one or more computation tasks transmitted in one or more cells of the base station that need to be measured by the first access network device (e.g., a base station). If the computation task corresponds to a QoS flow, the second measurement object can also be represented by a QoS flow ID. If the computation task corresponds to a radio bearer, the second measurement object can also be represented by a radio bearer ID.

[0182] The third measurement object can be represented by a computation task identifier, which identifies one or more computation tasks processed by the computation node that need to be measured. If the computation task corresponds to a QoS flow, the third measurement object can also be represented by a QoS flow ID. If the computation task corresponds to a radio bearer, the third measurement object can also be represented by a radio bearer ID.

[0183] The first measurement report configuration includes the report type (or the reporting principle), which can be periodic reporting or event-triggered reporting; or the measurement report format, such as the maximum number of computational tasks to be reported.

[0184] The second measurement report configuration includes the report type, which can be periodic reporting or event-triggered reporting; or the measurement report format, such as the maximum number of computational tasks to be reported.

[0185] The third measurement report configuration includes the report type, which can be periodic reporting or event-triggered reporting; or the measurement report format, such as the maximum number of computational tasks to be reported.

[0186] In one implementation, the measurement report configuration (e.g., a first measurement report configuration, a second measurement report configuration, or a third measurement report configuration) may include a report type (reportType), which includes event-triggered reporting or periodic reporting. When the report type includes event-triggered reporting, the measurement configuration information is switched to include a measurement report triggering event (e.g., a first measurement report triggering event, a second measurement report triggering event, or a third measurement report triggering event).

[0187] In this embodiment, through the aforementioned first handover measurement configuration information, the terminal can perform measurements based on the first handover measurement configuration information to obtain first measurement information, which can be used to determine whether to switch the computing node used to perform the computing task; or, through the aforementioned second handover measurement configuration information, the first access network device can perform measurements based on the second handover measurement configuration information to obtain second measurement information, which can be used to determine whether to switch the computing node used to perform the computing task; or, through the aforementioned third handover measurement configuration information, the source computing node can perform measurements based on the third handover measurement configuration information to obtain third measurement information, which can be used to determine whether to switch the computing node used to perform the computing task.

[0188] Optionally, the first measurement report triggering event includes at least one of the following:

[0189] The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions.

[0190] The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions.

[0191] The location of the terminal meets preset conditions;

[0192] The AI ​​performance of the artificial intelligence (AI) model used for the computational task meets the preset conditions;

[0193] The inference throughput of the AI ​​model used for the computational task meets the preset conditions;

[0194] The remaining time of the computing task meets the preset conditions;

[0195] The end-to-end computation latency of the computation task meets the preset conditions;

[0196] The failure rate of the computation task meets the preset conditions;

[0197] The computation speed of the computation task meets the preset conditions;

[0198] The computational intensity of the computational task meets the preset conditions.

[0199] For example, the uplink bandwidth or uplink bit rate corresponding to the computing task meeting the preset conditions may mean that the uplink bandwidth of the computing task (such as the first computing task) of the source cell measured by the terminal is always at or below a preset threshold within a preset time period, or the number of times it is below the preset threshold within the preset time period reaches a preset number.

[0200] For example, the downlink bandwidth or downlink bit rate corresponding to the computing task meeting the preset conditions may mean that the downlink bandwidth of the computing task (such as the first computing task) of the source cell measured by the terminal is always at or below a preset threshold within a preset time period, or the number of times it is below the preset threshold within the preset time period reaches a preset number.

[0201] For example, the location of the terminal meeting the preset conditions may mean that the location of the terminal consistently exceeds the preset area within a preset time period.

[0202] Specifically, the AI ​​performance of the AI ​​model used for the computational task meets preset conditions. AI performance can include AI model training performance and / or AI model inference performance. For AI models, performance parameters typically differ across different scenarios. For example, image recognition and object detection performance corresponds to top-1 accuracy or average accuracy, semantic segmentation performance corresponds to mean intersection over union (MIOU), and speech recognition corresponds to word error rate (WER).

[0203] Specifically, the inference throughput of the AI ​​model used for the computational task meets preset conditions. The inference throughput of visual AI models is typically the number of images inferred per second (images / s), while the inference throughput of natural language AI models is the number of sentences inferred per second (sentences / s) or tokens / s. Here, "tokens" refers to words, punctuation marks, or other text units in the input text processed by the model.

[0204] It should be noted that the AI ​​performance meeting the preset conditions and the AI ​​model inference throughput meeting the preset conditions are mainly related to the quality of computing services and do not involve communication performance.

[0205] In addition, for the three scenarios of terminal data transmission mentioned above (computation offloading, terminal data transmission during on-network computing, and terminal / AF data reception), optionally, the first measurement report triggering event may also include the remaining duration of the computing task (or computing service) meeting a preset condition. For example, the remaining duration of the computing task (or computing service) is not less than a preset threshold.

[0206] Additionally, for the terminal performing the calculation unloading, the first measurement report triggering event may optionally include at least one of the following:

[0207] (1) The end-to-end computation latency of the computation task meets preset conditions. End-to-end computation latency is used to define the total latency of the computation task during computation and transmission. Specifically, for example, the end-to-end computation latency for AI model inference includes at least one of the following:

[0208] Total end-to-end inference latency: Specifically, it refers to the total end-to-end latency of multiple consecutive inference operations. The calculation method is as follows: the time before the requesting node sends the first byte of the first computation task (or computation job) is denoted as T. IS The last byte received by the receiving node from all computing tasks (or computing jobs) is denoted as T. IE Then the computational latency budget for AI model inference is T. IE -T IS .

[0209] End-to-end inference latency: Specifically, it refers to the difference between the time it takes to send a sample and the time it takes to receive a result. That is, the time elapsed before the requesting node sends the first byte of a computation task (or job) is denoted as t. TIS The last byte received by the receiving node for the computation task (or job) is denoted as t. TIE Then the computational latency budget for AI model inference is t. TIE -t TIS .

[0210] (2) The computation failure rate (or expressed as failure rate) of the computation task meets the preset conditions. The failure rate refers to the number of failed requests for the computation task per unit time (e.g., per second, per 5 minutes, per day, etc.) divided by the total number of valid requests per unit time. Failed requests include incomplete computation requests and computation requests that have exceeded the latency threshold. Valid requests refer to computation requests accepted and processed by the first node.

[0211] (3) The computation speed of the computation task meets the preset conditions. Computation speed refers to the computation time complexity divided by the aforementioned end-to-end computation latency. For example, computation time complexity is usually represented by operations, and computation usually corresponds to a data type. 2 TFLOPs (floating-point operations) means that the time complexity of the computation is 2 × 10^12 floating-point operations. If the end-to-end computation latency is 20ms, then the computation speed is 2 × 10^12 / (20 × 10^(-3)) = 100 TFLOPS, where TFLOPS refers to the number of floating-point operations per second.

[0212] (4) The computational intensity of the computational task meets the preset conditions. Computational intensity refers to the computational speed divided by the bandwidth. The computational speed is as described above, and the bandwidth is the minimum value of the uplink bandwidth and the downlink bandwidth.

[0213] Additionally, when the computing management node is a radio access network node, especially in the case of an integrated 6G node B, the first measurement report triggering event may optionally include at least one of the conventional handover measurement events and parameters.

[0214] In this embodiment, the terminal can send the first measurement information obtained by measurement to the first node when the first measurement report triggering event is triggered, so that the first node can determine whether to switch the computing node used to perform the computing task based on the first measurement information. It can determine whether the computing node can complete the computing task well during the computing task execution process and decide whether to switch the computing node, thereby improving the computing service effect.

[0215] Optionally, the second measurement report triggering event includes at least one of the following:

[0216] The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions.

[0217] The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions.

[0218] The uplink channel quality of the terminal meets the preset conditions.

[0219] For example, the uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions, which may mean that the uplink bandwidth of the source cell's computing task (such as the first computing task) measured by the first access network device (such as a base station) is always at or below a preset threshold within a preset time period, or the number of times it is below the preset threshold within a preset time period reaches a preset number.

[0220] For example, the downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions, which may mean that the downlink bandwidth of the source cell's computing task (such as the first computing task) measured by the first access network device (such as a base station) is always at or below a preset threshold within a preset time period, or the number of times it is below the preset threshold within a preset time period reaches a preset number.

[0221] It should be noted that the uplink bandwidth (or uplink bit rate) or downlink bandwidth (or downlink bit rate) can be measured by either the base station or the terminal.

[0222] For example, the uplink channel quality (e.g., RSRP, SINR, etc.) of the terminal meeting the preset conditions can mean that the uplink channel quality of the terminal corresponding to the computation task (e.g., the first computation task) of the source cell, as measured by the first access network device (e.g., a base station), remains at or below a preset threshold within a preset time period, or the number of times it falls below the preset threshold within the preset time period reaches a preset number. There can be one or more terminals corresponding to the computation task, and these terminals may differ for different computation tasks.

[0223] In this embodiment, the first access network device can send the second measurement information obtained by measurement to the first node under the triggering of the second measurement report triggering event, so that the first node can determine whether to switch the computing node used to perform the computing task based on the second measurement information. It can determine whether the computing node can complete the computing task well during the computing task execution process and decide whether to switch the computing node, thereby improving the computing service effect.

[0224] Optionally, the third measurement report triggering event includes at least one of the following:

[0225] The failure rate of the computation task meets the preset conditions;

[0226] The computational power consumption of the computational task meets the preset conditions;

[0227] The computational energy efficiency of the computational task meets the preset conditions;

[0228] The computing resource utilization rate of the computing task meets the preset conditions;

[0229] The computation speed of the computation task meets the preset conditions;

[0230] The computational efficiency of the computational task meets the preset conditions;

[0231] The computational intensity of the computational task meets the preset conditions;

[0232] The computation time delay of the computation task meets the preset conditions;

[0233] The computing nodes of the computing task meet the load balancing preset conditions;

[0234] The computing node of the computing task failed.

[0235] Specifically, the failure rate (or simply failure rate) of a computation task must meet preset conditions. The failure rate is defined as the number of failed requests for the computation task per unit time (e.g., per second, per 5 minutes, per day, etc.) divided by the total number of valid requests per unit time. Failed requests include incomplete computation requests and computation requests that have exceeded the latency threshold. Valid requests refer to computation requests accepted and processed by the first node.

[0236] It should be noted that the failure rate for computing nodes is primarily calculated from a computational processing perspective. For UEs with computation offloading, the failure rate measured can be calculated from both computational and transmission end-to-end perspectives.

[0237] For the computational task to meet preset conditions, one method for calculating the computational power consumption is to calculate the power consumption of the computation node within the aforementioned computational delay time. Another method is: let P1 be the power consumption of the computation node per unit time when it is powered on, and P2 be the power consumption per unit time when processing the computational task. Then the computational power consumption is P2-P1 or the ratio of P2 to P1.

[0238] Among them, for computing tasks, the computing energy efficiency meets the preset conditions. One way to calculate computing energy efficiency is to divide the computing speed by the computing power consumption per unit time, or to divide the AI ​​model inference throughput by the computing power consumption to obtain the computing energy efficiency.

[0239] For example, the computing resource utilization rate of a computing task meets the preset conditions, which may mean that the utilization rate of the central processing unit (CPU) / graphics processing unit (GPU), or the memory utilization rate or storage utilization rate is always at or above a preset threshold within a preset time period, or the number of times the preset threshold is exceeded within a preset time period reaches a preset number.

[0240] The computation speed of a computation task must meet preset conditions. Computation speed refers to the number of operations or computations a computation node can process per unit of time. Optionally, computation speed includes theoretical computation speed and / or actual computation speed. Theoretical computation speed is calculated as processor clock frequency × number of operations performed per processor clock cycle × total number of system cores. Actual computation speed is the actual computation speed measured through the computation task. For example, meeting preset conditions for computation speed could mean that the actual computation speed is consistently at or below a preset threshold within a preset time period, or that the number of times the speed falls below the preset threshold within the preset time period reaches a preset number.

[0241] For example, the computational efficiency of a computational task meeting a preset condition can mean that the computational efficiency is consistently at or below a preset threshold within a preset time period, or that the number of times the efficiency is below the preset threshold within the preset time period reaches a preset number. The computational efficiency includes at least one of the following definitions:

[0242] Computational efficiency is defined as the ratio of actual computational speed under ideal conditions to theoretical computational speed.

[0243] Another form of computational efficiency is the ratio of the computational speed measured based on the computational task to the theoretical computational speed.

[0244] Another form of computational efficiency is the ratio of the computational speed measured based on the computational task to the maximum computational speed measured under ideal conditions.

[0245] Among these, the computational intensity of the computational task must meet the preset conditions. Computational intensity refers to the computational speed divided by the bandwidth. As mentioned earlier, the computational speed is defined in two ways: one is memory bandwidth, and the other is the minimum value among memory bandwidth, uplink bandwidth, and downlink bandwidth.

[0246] Among these, the computation latency of a computation task must meet preset conditions. Computation latency is used to define the total latency of the computation task during computation. For example, meeting the preset conditions for computation latency could mean that the computation latency of the computation task (such as the first computation task) measured by the computing node is consistently at or above a preset threshold within a preset time period, or that the latency exceeds the preset threshold a preset number of times within the preset time period. Specifically, for example, computation latency for AI model inference includes at least one of the following situations:

[0247] Total inference latency: Specifically, it refers to the total latency of multiple consecutive inference operations. The calculation method is as follows: the time before inference for the first computational task (or computational job) is denoted as T. ITS The time when all computational tasks (or computational jobs) finish inference is denoted as T. ITE Then the computational latency budget for AI model inference is T. ITE -T ITS .

[0248] Inference latency: Specifically, it refers to the difference between the start time and the end time of inference for a given sample. That is, the time taken before inference for a computational task (or job) is denoted as t. INS The time when the inference of this computational task ends is denoted as t. INE Then the computational latency budget for AI model inference is t. INE -t INS .

[0249] Specifically, for the computing nodes of the aforementioned computing task to meet the preset load balancing conditions, load balancing is used to distribute the load in the computing cluster to optimize resource utilization, maximize throughput, minimize response time, or avoid overload. Specifically, if a computing node has a high load, the computing task is switched to another computing node; or, if a computing node is processing only the aforementioned computing task, the computing task is switched to another computing node to shut down the node. For example, if the load of a computing node remains at or above a preset threshold within a preset time period, or exceeds the preset threshold a preset number of times within the preset time period, then the first node determines whether to switch the computing node based on whether the computing node meets the preset load balancing conditions; or if the load of a computing node remains at or below the preset threshold within a preset time period, or falls below the preset threshold a preset number of times within the preset time period, then the first node determines whether to switch the computing node based on whether the computing node meets the preset load balancing conditions.

[0250] Optionally, when the computing node is a terminal, the third measurement report triggering event may also include uplink bandwidth meeting preset conditions, or downlink bandwidth meeting preset conditions.

[0251] In this embodiment, the source computing node can send the third measurement information obtained from the measurement to the first node when the third measurement report triggering event is triggered. This allows the first node to determine whether to switch the computing node used to perform the computing task based on the third measurement information. The first node can determine whether the computing node can complete the computing task well during the execution of the computing task and decide whether to switch the computing node, thereby improving the computing service effect.

[0252] Optionally, the method further includes:

[0253] If the first node determines to switch the computing node, the first node sends a first switch request to the candidate target node. The first switch request is used to request to switch the computing node and includes the identifier of the computing task.

[0254] Among them, the candidate target node is a candidate computing node, or it can be described as a candidate target computing node.

[0255] In this embodiment, the first node sends a first switching request to the candidate target node. The candidate target node can determine whether to agree to be the computing node to be switched based on the first switching request, thereby realizing the switching of computing nodes and improving the computing service effect.

[0256] Optionally, the first handover request further includes at least one of the following:

[0257] The identifier of the source computing node; the maximum failure rate of the computing task; the maximum computing power consumption of the computing task; the minimum computing energy efficiency of the computing task; the minimum computing speed of the computing task; the minimum computing intensity of the computing task; the minimum computing efficiency of the computing task; and the maximum computing latency of the computing task.

[0258] The identifier of the source computing node may include Internet Protocol (IP) address, port number, etc.

[0259] The maximum failure rate of a computing task represents the maximum failure rate that a computing node can provide when processing that computing task.

[0260] The maximum computing power consumption of a computing task represents the maximum computing power consumption provided by the computing node in processing the computing task.

[0261] The minimum computational energy efficiency of a computation task represents the minimum computational energy efficiency provided by a computing node in processing that computation task.

[0262] The minimum computing speed of a computing task represents the minimum computing speed that a computing node can provide to process that computing task.

[0263] The minimum computational intensity of a computation task represents the minimum computational intensity that a computing node provides to process that computational task.

[0264] The minimum computational efficiency of a computation task represents the minimum computational efficiency provided by a computing node in processing that computation task.

[0265] The maximum computational latency of a computation task represents the maximum computational latency provided by the computing node in processing the computation task.

[0266] In this embodiment, through the information carried in the first switching request, the candidate target node can determine whether it can meet the needs of the computing task based on the first switching request, and thus decide whether to agree to be the computing node to be switched, which can better realize the switching of computing nodes and improve the computing service effect.

[0267] Optionally, after the first node sends a first handover request to the candidate target node, the method further includes:

[0268] The first node receives a first switching response sent by the candidate target node, the first switching response being used to indicate whether to agree to switch computing nodes.

[0269] The candidate target node may send the first switching response only when it agrees to switch the computing node, or it may send the first switching response regardless of whether the candidate target node agrees to switch the computing node.

[0270] In one embodiment, the first switching response may further include receiving updated status information of candidate computing nodes after the computing task, such as computing load, computing speed, computing intensity, available memory, available storage, computing power consumption, or computing energy efficiency.

[0271] In one implementation, when the first switching response indicates agreement to switch computing nodes, the first switching response may further include the computing performance that the candidate target node can support, including at least one of the following:

[0272] Supported maximum failure rate indication, such as 0.1% etc.;

[0273] Supported maximum computing power consumption indication, such as 10 watts, etc.;

[0274] Supported minimum computational energy efficiency indicator;

[0275] Minimum supported computing speed indication;

[0276] Supported minimum computational intensity indicator;

[0277] Supported minimum computational efficiency indicator.

[0278] In this embodiment, the first node receives a first switching response sent by the candidate target node. The first node can know whether the candidate target node agrees to switch computing nodes through the first switching response, which facilitates the first node to make the next decision.

[0279] Optionally, the method further includes:

[0280] When the first node determines to switch the computing node, the first node sends a first switch command to the source computing node. The first switch command carries a first switch indication, which is used to indicate the switching of the computing node.

[0281] The first switching command may further include information about the target computing node, such as its identifier. The target computing node may be the new computing node after the switching.

[0282] In this embodiment, the first node sends a first switching command to the source computing node, and the source computing node can switch computing nodes according to the first switching command, which can improve the computing service effect.

[0283] Optionally, the method further includes:

[0284] When the first node determines to switch the computing node, the first node sends a second switching command to the first access network device. The second switching command carries a second switching indication, which is used to indicate the switching of the computing node. The first access network device is the service access network device of the terminal associated with the computing task.

[0285] In this embodiment, the first node sends a second handover command to the first access network device, and the first access network device can assist in the handover of computing nodes according to the second handover command, which can improve the computing service effect.

[0286] Optionally, the second switching command further includes at least one of the following:

[0287] The information includes the identifier of the computing task, the identifier of the target computing node, and the switching time indication information.

[0288] Upon receiving the second handover command, the first access network device can forward computing task data to the target computing node based on the target computing node's identifier. For example, after receiving the second handover command, the first access network device sends the computing task data to the target computing node but not to the source computing node; or, if the second handover command includes handover time indication information T, then after receiving the second handover command, the first access network device sends the computing task data to both the target computing node and the source computing node within time T. After time T, it sends the computing task data to the target computing node but not to the source computing node.

[0289] Optionally, the method further includes:

[0290] The first node determines whether to switch the service access network device of the terminal associated with the computing task based on the first information;

[0291] When the first node determines to switch the serving access network device, the first node sends a second switching request to the second access network device, the second switching request being used to request the second access network device to be switched to the serving access network device.

[0292] It should be noted that the first node can decide whether to switch the service access network device of the terminal associated with the computing task based on the measurement information. The first node may decide to switch the service access network device when the measurement information indicates that the current service access network device cannot transmit the computing task data well; or, the first node may decide to switch the service access network device when the measurement information indicates that the data transmitted through the current service access network device cannot meet the performance requirements of the computing task; or, the first node may decide to switch the service access network device when the measurement information indicates that the data transmitted through the current service access network device cannot achieve the required computing service quality; etc., this embodiment does not limit this.

[0293] Furthermore, the first node can decide whether to switch the service access network device of the terminal associated with the computing task based on the computing service request. The first node may determine to switch the service access network device when the current service access network device cannot transmit the computing task data well; or, the first node may determine to switch the service access network device when transmitting the computing task data through the current service access network device cannot satisfy the computing service request; or, the first node may determine to switch the service access network device when transmitting the computing task data through the current service access network device cannot achieve the computing service quality required by the computing service request; and so on. This embodiment does not limit this to any specific instance.

[0294] In this embodiment of the application, in a mobile network communication system that supports computing services, a terminal can act as a sending node, receiving node, or computing node for computing services. The first information is used to determine whether to switch the service access network device of the terminal. Thus, when the first terminal acts as a sending node, receiving node, or computing node for computing services, it can support switching the access network device used to transmit computing service data. This embodiment of the application can solve the terminal switching problem in computing and communication services.

[0295] In this embodiment, determining whether to switch the terminal's service access network device based on the first information allows for decision-making on whether to switch the terminal's service access network device while the computing node is performing computing tasks, taking into account measurement information or computing service requests, thereby improving the computing service effect.

[0296] Optionally, the second handover request includes at least one of the following:

[0297] The minimum bit rate corresponding to the computing task; the maximum bit rate corresponding to the computing task; the maximum latency corresponding to the computing task.

[0298] The minimum bit rate corresponding to the computation task can represent the bit rate that the network guarantees to provide to the computation task within the average window. The minimum bit rate can be represented by data, a range of values, or a list of values ​​(i.e., multiple values).

[0299] The maximum bit rate corresponding to the computation task can represent the maximum bit rate that the network can provide for that computation task. The maximum bit rate can be represented by data, a range of values, or a list of values ​​(i.e., multiple values).

[0300] The maximum latency corresponding to the computation task can represent the maximum latency (e.g., uplink latency, downlink latency, or round-trip latency) that the network guarantees to provide to the computation task within an average window. The maximum latency can be represented by data, a range of values, or a list of values ​​(i.e., multiple values).

[0301] In this embodiment, the second access network device can determine whether it can meet the needs of the computing task by using the information carried in the second handover request, and thus decide whether to agree to switch to the access network device used to serve the terminal associated with the computing task.

[0302] Optionally, after the first node sends a second handover request to the second access network device, the method further includes:

[0303] The first node receives a second handover response sent by the second access network device, the second handover response being used to indicate whether it agrees to hand over to the serving access network device.

[0304] The second handover response may further include communication capabilities that the second access network device can support, including at least one of the following:

[0305] Supported minimum bit rate indication, such as 3Mbps, etc.;

[0306] The maximum supported bit rate is indicated, such as 10 Mbps;

[0307] Supported maximum latency indication, such as 10ms.

[0308] In this embodiment, the first node receives a second handover response sent by the second access network device. The first node can know whether the second access network device agrees to switch to the access network device used to serve the terminal associated with the computing task through the second handover response, so as to facilitate the first node to make the next decision.

[0309] Optionally, the computing service request includes at least one of the following:

[0310] The following parameters are specified: computing service identifier; resource type of computing resources required by the computing task; minimum number of operations required by the computing task; minimum computing speed required by the computing task; minimum computing intensity required by the computing task; computing latency budget required by the computing task; maximum failure rate of the computing task; average window of the computing task statistics; maximum duration of the computing task; computing power type of the computing task; data type of the computing task; minimum memory required by the computing task; minimum storage space required by the computing task; minimum transmission bandwidth required by the computing task; arrival mode of the computing task; parameters of the arrival mode of the computing task; training accuracy of the AI ​​model corresponding to the computing task; AI performance parameters of the AI ​​model corresponding to the computing task; minimum inference throughput of the AI ​​model corresponding to the computing task; computing power consumption threshold of the computing task; computing energy efficiency threshold of the computing task.

[0311] The computing service identifier is used to identify computing services. Examples include one-dimensional Discrete Fourier Transform (DFT), two-dimensional Fast Fourier Transform (FFT), AI model training, and AI model inference. AI model training or inference can further include image recognition, object detection, semantic segmentation, recommendation, natural language processing, speech recognition, optical character recognition, face recognition, beam management, localization, perception, and Channel State Information (CSI) feedback.

[0312] The resource types of computing resources required for a computing task can also be referred to as computing resource types, or computing and communication resource types. Potential resource types include guaranteed computing rate (GCR or guaranteed operational rate, GOR), guaranteed computing intensity (GCI or guaranteed operational intensity, GOI), non-guaranteed computing rate (Non-GCR or Non-GOR), non-guaranteed computing intensity (Non-GCI or Non-GOI), delay-critical guaranteed computing rate (delay-critical GCR or delay-critical GOR), and delay-critical guaranteed computing intensity (delay-critical GCI or delay-critical GOI).

[0313] The minimum number of operations (or operands) for a computation task is defined as the minimum number of operations or operands required for that computation task.

[0314] The minimum computational speed required for a computational task is defined as the computational time complexity divided by the tolerable upper limit of computation time. For example, computational time complexity is typically represented by operations, and computations often correspond to a data type (see data type description below). 2 TFLOPs (floating-point operations) indicates that the computational time complexity is 2 × 10^12 floating-point operations. If the tolerable upper limit of computation time is 20ms, meaning the computation can be completed in a maximum of 20ms, then the computational speed is 2 × 10^12 / (20 × 10^(-3)) = 100 TFLOPS.

[0315] Optionally, the minimum computing speed includes the theoretical minimum computing speed and / or the actual minimum computing speed. The actual minimum computing speed can be represented by test case indicators and the actual minimum computing speed itself. Alternatively, it can be represented by the theoretical minimum computing speed, test case indicators, and computational efficiency. Or, it can be represented by the actual minimum computing speed, test case indicators, and computational efficiency itself.

[0316] In this context, the minimum computational intensity required for a computational task is defined as computational speed divided by bandwidth. As mentioned earlier, in the process of providing computational services in 6G, bandwidth consists of multiple components, including the transmission bandwidth between the requesting node (e.g., UE) and the computational node, and the memory bandwidth of the computational node. One method of defining computational intensity is segmented, such as computational speed divided by memory bandwidth. Another method is to take the minimum of the aforementioned related bandwidths as the numerator, i.e., computational speed divided by min{memory bandwidth, transmission bandwidth between UE and computational node}. The transmission bandwidth between the UE and computational node can be further divided into the air interface bandwidth between the UE and the access network node, and the wired transmission bandwidth between the access network node and the computational node. Alternatively, the transmission bandwidth between the UE and computational node can be further divided into the bandwidth between the UE and the User Plane Function Node (UPF) node, and the wired transmission bandwidth between the UPF and the computational node.

[0317] The computing delay budget (CDB) required for the computing task can be defined as the upper limit of the latency (i.e., the maximum latency) that the computing task can tolerate when computing between the computing request node (such as the UE) and the computing node.

[0318] The upper limit of tolerable latency during computation refers to the computation latency of the computing node.

[0319] The upper limit of tolerable latency during transmission refers to the sum of the transmission latency from the requesting node to the computing node and the transmission latency from the computing node to the receiving node.

[0320] One definition is the length of the time interval between the first data packet of a single computational task being sent and the last data packet of the received computational task. Another definition is the length of the time interval between the first data packet of a group of computational tasks being sent and the last data packet being received. For example, for image recognition computational tasks, one approach is to treat single image recognition as a single computational task, while another approach is to treat multiple images (e.g., 100 images) as a group of computational tasks.

[0321] Specifically, computational latency budgets for AI model inference include at least one of the following:

[0322] Total end-to-end inference latency: Specifically, it refers to the total end-to-end latency of multiple consecutive inference operations. The calculation method is as follows: the time before the requesting node sends the first byte of the first computation task (or computation job) is denoted as T. IS The last byte received by the receiving node from all computing tasks (or computing jobs) is denoted as T. IE Then the computational latency budget for AI model inference is T. IE -T IS .

[0323] Total inference latency: Specifically, it refers to the total latency of multiple consecutive inference operations. The calculation method is as follows: the time before inference for the first computational task (or computational job) is denoted as T. ITS The time when all computational tasks (or computational jobs) finish inference is denoted as T. ITE Then the computational latency budget for AI model inference is T. ITE -T ITS .

[0324] End-to-end inference latency: Specifically, it refers to the difference between the time it takes to send a sample and the time it takes to receive a result. That is, the time elapsed before the requesting node sends the first byte of a computation task (or job) is denoted as t. TIS The last byte received by the receiving node for the computation task (or job) is denoted as t. TIE Then the computational latency budget for AI model inference is t. TIE -t TIS .

[0325] Inference latency: Specifically, it refers to the difference between the start time and the end time of inference for a given sample. That is, the time taken before inference for a computational task (or job) is denoted as t. INSThe time when the inference of this computational task ends is denoted as t. INE Then the computational latency budget for AI model inference is t. INE -t INS .

[0326] The maximum failure rate for a computation task is calculated by dividing the number of failed requests per unit time (e.g., per second, per 5 minutes, per day) by the total number of valid requests per unit time. Failed requests include incomplete computation requests and computation requests that have exceeded the latency threshold. Valid requests refer to computation requests accepted and processed by the first node.

[0327] Among them, the average window for the statistics of computing tasks can be the average statistical time window of computing speed, computing intensity, or computing delay budget.

[0328] The maximum duration of the computation task can be represented by one or more of the start time, duration, and end time.

[0329] The computing power type of the computing task can be at least one of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Field Programmable Gate Array (FPGA), Data Processing Unit (DPU), Smart Network Interface Card (SmartNIC), Tensor Processing Unit (TPU), and Neural Network Processing Unit (NPU) (e.g., a combination of two or more). Optionally, the computing power type may also include at least one of the following:

[0330] Clock speed;

[0331] Number of cores.

[0332] The data type of the computation task can be at least one of integers (such as int8, int4, etc.) and floating-point numbers (such as half-precision floating-point numbers, single-precision floating-point numbers, double-precision floating-point numbers, half-precision floating-point numbers, etc.).

[0333] The minimum memory required for the computing task can be the lower limit of the memory required for the computing task, and can be represented by at least one of the following: memory size (e.g., 8G), sustainable memory bandwidth, and memory random access rate.

[0334] The minimum storage space required for the computing task can refer to minimum storage, which defines the lower limit of storage required for the computing task. This minimum storage can be represented by at least one of storage size (e.g., 1T) and storage bandwidth (the maximum input / output (IO) flow per unit time).

[0335] The minimum transmission bandwidth required for the computation task can be used to define the tolerable lower limit of uplink bandwidth and / or downlink bandwidth for the computation task. It can be expressed by at least one of the following:

[0336] Uplink or downlink bandwidth indicator, for example, 0 represents uplink bandwidth and 1 represents downlink bandwidth;

[0337] The uplink and downlink bandwidth are the same, for example, 0 means that the uplink bandwidth and downlink bandwidth are different, and 1 means that the uplink bandwidth and downlink bandwidth are the same;

[0338] Minimum uplink bandwidth;

[0339] Minimum downlink bandwidth.

[0340] This includes the computation task arrival mode (or computation job arrival mode) and parameters. A computation task may contain multiple computation jobs, and the potential modes and parameters include at least one of the following:

[0341] Consecutive arrival mode or single arrival mode: The i-th job (where i is a positive integer) arrives immediately after the (i-1)-th job is completed. Job i is not sent if job (i-1) is not completed or the delay budget threshold is not met.

[0342] Fixed-period arrival mode: Jobs arrive at a fixed period T, with n jobs arriving at a time (n is a positive integer);

[0343] Poisson distribution arrival pattern: Operations based on Where k is the number of jobs arriving per unit time (k is a positive integer), and λ (λ is a positive integer) is the average number of jobs arriving per unit time (e.g., per second);

[0344] Peak arrival pattern: In the Poisson distribution arrival pattern, there are j short periods, each period has a sudden surge in a large number of jobs, and the period lasts for a certain duration T. G (e.g., 5s-10s), and maintain a certain concurrency level σ (σ is a positive integer, e.g., σ>2). 5 (Number of jobs / second), jobs arriving within a short period conform to the fixed-period arrival pattern;

[0345] Offline arrival mode: All items arrive at once;

[0346] Mixed arrival mode: Composed of more than one of the above arrival modes.

[0347] The training precision of the AI ​​model corresponding to the computation task can be a training precision, which can be represented by the data type of the AI ​​model parameters output, such as single-precision floating-point numbers or half-precision floating-point numbers.

[0348] The AI ​​performance parameters of the AI ​​model corresponding to the computational task can include AI performance thresholds, which are the tolerable limits for AI model training and / or the lower limit of AI model inference performance. For AI models, performance parameters typically differ across different scenarios. For example, image recognition and object detection performance corresponds to top-1 accuracy and average accuracy, semantic segmentation performance corresponds to Mean Intersection Over Union (MIOU), and speech recognition performance corresponds to Word Error Rate (WER). For large models, the performance lower limit can be represented by the score of large model benchmark tests. Optionally, the AI ​​performance lower limit can also be represented by at least one of the following:

[0349] The test dataset indicator indicates which dataset, when used, yields performance at least equal to the stated AI performance lower limit. Commonly used AI datasets include ImageNet2012, Pascal VOC2012, LibriSpeech ASR Corpus, or Criteo, etc.

[0350] The minimum inference throughput of the AI ​​model corresponding to the computation task can refer to the minimum inference throughput rate of the AI ​​model. For vision-related tasks, this is usually the number of images inferred per second (images / s), while for natural science tasks, it is the number of sentences inferred per second (sentences / s) or tokens / s. Here, "tokens" refers to words, punctuation marks, or other text units in the input text processed by the model.

[0351] Among these, the power consumption threshold for the computing task is, for example, the maximum computing power consumption. One approach is to calculate the power consumption of the computing node within the aforementioned computing latency budget. Another approach is to denote the power consumption of the computing node per unit time when it is powered on as P1, and the power consumption per unit time when processing a computing task as P2. Then, the computing power consumption is P2 - P1 or the ratio of P2 to P1.

[0352] Among these, the energy efficiency threshold for computing tasks includes, for example, minimum energy efficiency. Minimum energy efficiency is the computing speed per unit time divided by the power consumption, or the AI ​​model inference throughput divided by the power consumption.

[0353] In this embodiment, the first node can determine the performance requirements of the computing task through the information carried in the computing service request, and thus decide whether to switch the computing node used to execute the computing task based on the performance requirements of the computing task, thereby improving the computing service effect.

[0354] The following examples further illustrate the compute node switching method:

[0355] In the example below, we will use the first node as the computing management node, the first access network device as the base station (such as the serving base station or the source base station), and the second access network device as the target base station for illustration.

[0356] Example 1:

[0357] This example demonstrates how to switch computing nodes based on measurement information.

[0358] In this example, for a UE, base station, or computing node that is currently providing computing services, the system determines whether to switch the UE's serving computing node based on measurement information (this example is limited to the case where the serving base station remains unchanged), thereby solving the UE handover problem in computing and communication services.

[0359] The aforementioned base stations providing computing services refer to base stations transmitting data for computing services provided by the 6G network. The aforementioned computing nodes providing computing services refer to computing nodes providing computing services. The potential computing service flows for the aforementioned UEs providing computing services include on-demand computing offloading and on-demand on-network computing for the mobile terminal. On-demand computing offloading refers to the mobile terminal determining whether to offload computing to the network for processing based on local computing power, computing task requirements, power consumption, etc. If computing offloading is required, then the data to be computed is sent to the network, and the computation result is obtained. The performance of on-demand computing offloading is related to the performance of the computing node, uplink transmission performance, downlink transmission performance, etc. On-demand on-network computing refers to providing computing services on demand for the transmitted data during network transmission. Furthermore, based on the sending and receiving nodes that can transmit data, the following scenarios can be identified:

[0360] (1) The mobile terminal sends data, which is then transmitted and processed by the network, and the application functions receive the data.

[0361] (2) The application sends data, which is then transmitted over the network and processed by calculation before the mobile terminal receives the data.

[0362] (3) The mobile terminal sends data, which is then transmitted and processed by the network, and then another mobile terminal receives the data.

[0363] The performance of scenario (1) is related to the performance of computing nodes and uplink transmission performance; the performance of scenario (2) is related to the performance of computing nodes and downlink transmission performance; the performance of scenario (3) is related to the performance of computing nodes, the uplink transmission performance of the sending data terminal, and the downlink transmission performance of the receiving data terminal.

[0364] As shown in Figure 4, the computing node switching method includes the following process:

[0365] Step (11): The computing management node sends the first handover measurement configuration information to the UE. After receiving the handover measurement configuration, the UE performs handover measurement and sends the first handover measurement report back to the computing management node.

[0366] or

[0367] The computing management node sends the second handover measurement configuration information to the base station. After receiving the handover measurement configuration, the base station performs handover measurement and sends the second handover measurement report back to the computing management node.

[0368] or

[0369] The computing management node sends the third handover measurement configuration information to the source computing node (which can be a core network node or a radio access network node). After receiving the handover measurement configuration, the base station performs handover measurement and sends the third handover measurement report back to the computing management node.

[0370] The first switching configuration information includes at least one of the following:

[0371] First measurement object; first measurement report configuration; first measurement report triggering event and related parameters.

[0372] The first measurement object is described in the preceding description of the first measurement object, and will not be repeated here.

[0373] The configuration of the first measurement report is described in the preceding description of the configuration of the first measurement report, and will not be repeated here.

[0374] The first measurement report triggering event is described in the preceding description of the first measurement report triggering event, and will not be repeated here.

[0375] The second switching configuration information includes at least one of the following:

[0376] Second measurement object; Second measurement report configuration; Second measurement report triggering events and related parameters.

[0377] The second measurement object is described in the preceding description of the second measurement object, and will not be repeated here.

[0378] The configuration of the second measurement report is described in the preceding section and will not be repeated here.

[0379] The second measurement report triggering event is described in the preceding description of the second measurement report triggering event, and will not be repeated here.

[0380] The third switching configuration information includes at least one of the following:

[0381] Third measurement object; third measurement report configuration; third measurement report triggering events and related parameters.

[0382] The third measurement object is described in the preceding section on the third measurement object, and will not be repeated here.

[0383] The configuration of the third measurement report is described in the preceding section on the configuration of the third measurement report, and will not be repeated here.

[0384] The third measurement report triggering event is described in the aforementioned description of the third measurement report triggering event, and will not be repeated here.

[0385] Step (12): Based on the measurement report from step (11), the computing management node decides whether to switch computing nodes. For example, if it does not affect the quality of the computing service, it does not switch, maintaining the current computing service or ending the current computing service. For example, if the remaining time of the computing service is long and it affects the quality of the computing service, it switches computing nodes. Or, if the source computing node of the current service fails, the computing management node determines to switch.

[0386] Step (13) If a switch is decided, the computing management node sends a first switch request to at least one candidate target computing node based on information such as computing power type and computing node load. The first switch request includes a computing task identifier.

[0387] Optionally, the first handover request may also include at least one of the following:

[0388] Source computing node identifiers, such as Internet Protocol (IP) address, port number, etc.;

[0389] Maximum failure rate represents the maximum failure rate that a compute node can provide when processing this compute task;

[0390] Maximum computing power consumption indicates the maximum computing power consumption provided by the computing node to process the computing task;

[0391] Minimum computing energy efficiency represents the minimum computing energy efficiency provided by a computing node in processing the computing task.

[0392] Minimum computing speed represents the minimum computing speed that a computing node can provide to handle this computing task.

[0393] Minimum computational intensity represents the minimum computational intensity that a computing node provides to handle the computational task.

[0394] Minimum computational efficiency refers to the minimum computational efficiency that a computing node can provide in processing this computational task.

[0395] Maximum computation latency indicates the maximum computation latency provided by the computing node in processing the computation task.

[0396] Step (14): Optionally, at least one of the candidate target computing nodes sends a first switching response to the computing management node. Optionally, the first switching response may also include updated status information of the computing node after accepting the computing task, such as computing load, computing speed, computing intensity, available memory, available storage, computing power consumption, computing energy efficiency, etc.

[0397] Step (15): The computing management node sends a first switching command to the source computing node, the first switching command including a switching indication.

[0398] Optionally, the first switching command may also include target computing node information, such as the target computing node identifier.

[0399] Step (16): The computing management node sends a second handover command to the serving base station. The second handover command includes a computing node handover indication.

[0400] Optionally, the second switching command may also include: a computation task identifier or target computation node information.

[0401] The computation task identifier can be a newly defined computation task identifier, or it can be represented by PDU session ID, QoS flow ID, or Radio Bearer (RB) ID based on the mapping relationship with Protocol Data Unit (PDU) session, QoS flow, or radio bearer.

[0402] Target computing node information, such as target computing node identifier. The serving base station forwards computing task data to the target computing node based on the target computing node information. Optionally, this includes the following two methods:

[0403] After receiving the handover instruction, the serving base station sends the computing task data to the target computing node instead of the source computing node; or

[0404] If the second handover command includes a handover time indication T, then after receiving the handover indication, the serving base station will send the computing task data to both the target computing node and the source computing node within time T. After time T, it will send the computing task data to the target computing node but not to the source computing node.

[0405] Example 2:

[0406] This example demonstrates the handover between computing nodes and base stations based on measurement information.

[0407] Example 1 addresses a UE, base station, or computing node currently providing computing services by determining, based on measurement information, whether to switch the UE's serving computing node (limited to the case where the serving base station remains unchanged). This example, however, determines whether to switch the UE's serving computing node and serving base station, thus resolving the UE handover issue in computing and communication services.

[0408] As shown in Figure 5, the computing node switching method includes the following process:

[0409] Step (21) is the same as step (21) in Example 1, and will not be repeated here.

[0410] Step (22): Based on the measurement report from step (21), the computing management node decides whether to switch computing nodes and base stations. If switching computing nodes and not switching base stations cannot meet the requirements, or if not switching computing nodes and switching base stations also cannot meet the requirements, then it is determined to switch computing nodes and switch base stations.

[0411] Steps (23) to (24):

[0412] Similar to Example 1, the potential difference is that the relevant parameters in the first switching request in step (23) can be a range or multiple values ​​or levels.

[0413] Maximum failure rate, such as [0.1%, 1%], or 0.1%, 1%, 5%, etc.;

[0414] Maximum computing power consumption, for example [1,10], or 1, 5, 10, etc.;

[0415] Minimum calculated energy efficiency, expressed as a range of values ​​or different values;

[0416] Minimum computation speed, expressed as a range of values ​​or different values;

[0417] Minimum computational intensity, expressed as a numerical range or different numerical values;

[0418] Minimum computational efficiency, expressed as a range of values ​​or different values;

[0419] Maximum computation delay, expressed as a range of values ​​or different values.

[0420] A potential difference is that, when agreeing to switch, optionally, the first switching response sent by the candidate computing node in step (24) includes the computing performance that the candidate target computing node can support, specifically including at least one of the following:

[0421] Supported maximum failure rate indication, such as 0.1% etc.;

[0422] Supported maximum computing power consumption indication, such as 10 watts, etc.;

[0423] Supported minimum computational energy efficiency indicator;

[0424] Minimum supported computing speed indication;

[0425] Supported minimum computational intensity indicator;

[0426] Supported minimum computational efficiency indicator.

[0427] Step (25): If a handover decision is made, the computation management node sends a second handover request to at least one candidate target base station. The second handover request includes at least one of the following:

[0428] (1) Minimum bit rate, which represents the bit rate that the network guarantees to provide to the computation task within the average window. It can be represented by data, a range of values, or a list of values ​​(i.e., multiple values).

[0429] (2) Maximum bit rate, which represents the maximum bit rate that the network can provide for this computational task. It can be represented by data, a range of values, or a list of values ​​(i.e., multiple values).

[0430] (3) Maximum latency, which represents the maximum latency (uplink latency, downlink latency, or round-trip latency) that the network guarantees to provide to the computing task within the average window. It can be represented by data, numerical ranges, or numerical lists (i.e., multiple values).

[0431] Step (26): The candidate target base station sends a second handover response to the computing management node to indicate whether it agrees to the handover. If the handover is agreed to, the second handover response may optionally include the communication performance that the target candidate base station can support, specifically including at least one of the following:

[0432] Supported minimum bit rate indication, such as 3Mbps, etc.;

[0433] The maximum supported bit rate is indicated, such as 10 Mbps;

[0434] Supported maximum latency indication, such as 10ms.

[0435] The computing management node determines the target computing node and the target base station based on the information from steps (24) and (26). Optionally, based on the performance indicators of the target computing node and the target base station, it determines whether to switch the computing node first or the base station first.

[0436] Example 3:

[0437] This example demonstrates the handover of a computationally and communications-integrated base station (such as an Integrated Node B base station) based on measurement information.

[0438] Example 1 addresses a UE, base station, or computing node currently providing computing services, determining whether to switch the UE's serving computing node based on measurement information (limited to the case where the serving base station remains unchanged). This example, however, addresses the scenario where the base station (source / target base station) is a computing management node, a communication serving base station, and a computing node, determining whether to switch the UE's communication serving base station and serving computing node, thus resolving the UE handover issue in computing and communication services.

[0439] In this example, the source base station implements the functions of the first node and the computing node.

[0440] As shown in Figure 6, the computing node switching method includes the following process:

[0441] Step (31): The base station (e.g., the source base station) sends the first handover measurement configuration information to the UE. After receiving the first handover measurement configuration, the UE performs handover measurement and sends the first handover measurement report back to the computing management node. The first handover configuration information is the same as in Example 1 and will not be described again.

[0442] Alternatively, the base station receives a computing service request sent by the UE.

[0443] The computing service request is described in the preceding description of the computing service request, and will not be repeated here.

[0444] Step (32): Based on the measurement report from step (31), the base station (such as the source base station) decides whether to switch over.

[0445] Alternatively, the base station may decide whether to switch over based on the received computing service requests.

[0446] Step (33): If a handover decision is made, the base station (source base station) sends a handover request to at least one candidate target base station. The handover request includes at least one of the following: computing task identifier, computing service identifier, resource type, minimum number of operations, minimum computing speed, minimum computing intensity, computing latency budget, maximum failure rate, average window, maximum time, computing power type, data type, minimum memory, minimum storage, minimum transmission bandwidth, computing task arrival mode or computing job arrival mode and parameters, AI model training accuracy, AI performance threshold, minimum AI model inference throughput, computing power consumption threshold, and computing energy efficiency threshold.

[0447] In this context, computing service identifiers and computing tasks can be mapped one-to-one, or multiple computing services can be mapped to a single computing task identifier.

[0448] Optionally, the switching request may also include at least one of the following:

[0449] Calculate the remaining service time;

[0450] Calculate the service end time;

[0451] UE location.

[0452] Step (34): The candidate target base station determines whether to accept the handover based on the handover request information. A handover response is sent, which falls into one of the following two categories:

[0453] If agreed, the candidate target base station sends a response message to the source base station, the response message indicating agreement to perform the handover.

[0454] If the candidate base station disagrees, it may optionally send a rejection message to the source base station, the rejection message indicating a refusal to perform the handover.

[0455] Step (35): The UE, source base station, and target base station perform a handover. Specifically, this is divided into the following two cases:

[0456] (1) A soft handover method is adopted. The target base station configures parameters based on the handover request information to perform computing and communication services. After completing at least one computing job or computing task, the target base station sends a handover success message to the source base station or the first device.

[0457] Furthermore, after receiving the handover success message, the source base station releases the resources (including computing resources, time and frequency resources, antenna port resources, etc.) occupied by the computing and communication.

[0458] (2) Use hard handover method. While performing step (34), the source base station does not need to wait for the handover success message and releases the resources occupied by the computing and communication (including computing resources, time and frequency resources, antenna port resources, etc.).

[0459] In this embodiment, the decision to switch computing nodes is based on measurement information. Optionally, the base station can also be switched, thereby solving the UE handover problem in computing and communication services. This embodiment is applicable to providing computing services to AF (AF-based) as well as to providing computing services (such as CSI feedback) to UE and NF (Network Functions). Furthermore, it better ensures the continuity and performance of computing services.

[0460] Referring to Figure 7, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0461] Step 201: The terminal sends the first measurement information or calculation service request to the first node;

[0462] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0463] Optionally, before the terminal sends the first measurement information to the first node, the method further includes:

[0464] The terminal receives the first handover measurement configuration information sent by the first node.

[0465] It should be noted that this embodiment is an implementation method of the terminal corresponding to the embodiment shown in FIG3. For the specific implementation method, please refer to the relevant description of the embodiment shown in FIG3. To avoid repeated description, this embodiment will not be repeated.

[0466] Referring to Figure 8, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0467] Step 301: The first access network device sends the second measurement information to the first node. The first access network device is the service access network device of the terminal associated with the computing task.

[0468] The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0469] Optionally, before the first access network device sends the second measurement information to the first node, the method further includes:

[0470] The first access network device receives the second handover measurement configuration information sent by the first node.

[0471] Optionally, the method further includes:

[0472] The first access network device receives a second handover command sent by the first node. The second handover command carries a second handover indication, which is used to indicate the switching of computing nodes.

[0473] It should be noted that this embodiment is an implementation of the first access network device corresponding to the embodiment shown in Figure 3. For the specific implementation, please refer to the relevant description of the embodiment shown in Figure 3. To avoid repetition, this embodiment will not be described again.

[0474] Referring to Figure 9, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0475] Step 401: The source computing node sends the third measurement information to the first node;

[0476] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0477] Optionally, before the source computing node sends the third measurement information to the first node, the method further includes:

[0478] The source computing node receives the third switching measurement configuration information sent by the first node.

[0479] Optionally, the method further includes:

[0480] The source computing node receives a first switching command sent by the first node. The first switching command carries a first switching indication, which is used to indicate the switching of computing nodes.

[0481] It should be noted that this embodiment is an implementation of the source computing node corresponding to the embodiment shown in Figure 3. For the specific implementation method, please refer to the relevant description of the embodiment shown in Figure 3. To avoid repetition, this embodiment will not be described again.

[0482] Referring to Figure 10, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0483] Step 501: The candidate target node receives a first switching request sent by the first node. The first switching request is used to request a switching of the computing node and includes the identifier of the computing task.

[0484] Optionally, after the candidate target node receives the first handover request sent by the first node, the method further includes:

[0485] The candidate target node sends a first switching response to the first node, the first switching response being used to indicate whether it agrees to switch computing nodes.

[0486] It should be noted that this embodiment is an implementation of the candidate target node corresponding to the embodiment shown in Figure 3. For the specific implementation, please refer to the relevant description of the embodiment shown in Figure 3. To avoid repetition, this embodiment will not be described again.

[0487] Referring to Figure 11, which is a flowchart of a computing node switching method provided in an embodiment of this application, the computing node switching method includes the following steps:

[0488] Step 601: The second access network device receives a second handover request sent by the first node. The second handover request is used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

[0489] Optionally, after the second access network device receives the second handover request sent by the first node, the method further includes:

[0490] The second access network device sends a second handover response to the first node, the second handover response being used to indicate whether it agrees to hand over to the serving access network device.

[0491] It should be noted that this embodiment is an implementation of the second access network device corresponding to the embodiment shown in Figure 3. For the specific implementation, please refer to the relevant description of the embodiment shown in Figure 3. To avoid repetition, this embodiment will not be described again.

[0492] The compute node switching method provided in this application can be executed by a compute node switching device. This application uses the execution of the compute node switching method by a compute node switching device as an example to illustrate the compute node switching device provided in this application.

[0493] This application provides a compute node switching device. As an example, the compute node switching device can be a communication device or a component within a communication device, such as a chip. The communication device can be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal can be, but is not limited to, the type of terminal 11 listed above, and the network-side device can be, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0494] The computing node switching device includes a receiving module, a transmitting module, and a processing module. These modules can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), Artificial Intelligence (AI) processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving and transmitting modules can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0495] Specifically, referring to Figure 12, when the computing node switching device is the first node or a component within the first node, the computing node switching device 700 includes:

[0496] Receiving module 701 is configured to receive first information, the first information including at least one of measurement information and a calculation service request;

[0497] The processing module 702 is used to determine whether to switch the computing node used to perform the computing task based on the first information.

[0498] Optionally, the device further includes a transmitting module for at least one of the following:

[0499] Send first handover measurement configuration information to the terminal, wherein the measurement information includes first measurement information corresponding to the first handover measurement configuration information;

[0500] Send second handover measurement configuration information to a first access network device, wherein the first access network device is the serving access network device of the terminal associated with the computing task, and the measurement information includes second measurement information corresponding to the second handover measurement configuration information;

[0501] Send third handover measurement configuration information to the source computing node, wherein the measurement information includes third measurement information corresponding to the third handover measurement configuration information.

[0502] Optionally, the first switching measurement configuration information includes at least one of the following: a first measurement object; a first measurement report configuration; a first measurement report triggering event; and related parameters of the first measurement report triggering event.

[0503] or

[0504] The second switching measurement configuration information includes at least one of the following: a second measurement object; a second measurement report configuration; a second measurement report trigger event; and related parameters of the second measurement report trigger event.

[0505] or

[0506] The third switching measurement configuration information includes at least one of the following: third measurement object; third measurement report configuration; third measurement report trigger event; and related parameters of the third measurement report trigger event.

[0507] Optionally, the first measurement report triggering event includes at least one of the following:

[0508] The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions.

[0509] The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions.

[0510] The location of the terminal meets preset conditions;

[0511] The AI ​​performance of the artificial intelligence (AI) model used for the computational task meets the preset conditions;

[0512] The inference throughput of the AI ​​model used for the computational task meets the preset conditions;

[0513] The remaining time of the computing task meets the preset conditions;

[0514] The end-to-end computation latency of the computation task meets the preset conditions;

[0515] The failure rate of the computation task meets the preset conditions;

[0516] The computation speed of the computation task meets the preset conditions;

[0517] The computational intensity of the computational task meets the preset conditions.

[0518] Optionally, the second measurement report triggering event includes at least one of the following:

[0519] The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions.

[0520] The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions.

[0521] The uplink channel quality of the terminal meets the preset conditions.

[0522] Optionally, the third measurement report triggering event includes at least one of the following:

[0523] The failure rate of the computation task meets the preset conditions;

[0524] The computational power consumption of the computational task meets the preset conditions;

[0525] The computational energy efficiency of the computational task meets the preset conditions;

[0526] The computing resource utilization rate of the computing task meets the preset conditions;

[0527] The computation speed of the computation task meets the preset conditions;

[0528] The computational efficiency of the computational task meets the preset conditions;

[0529] The computational intensity of the computational task meets the preset conditions;

[0530] The computation time delay of the computation task meets the preset conditions;

[0531] The computing nodes of the computing task meet the load balancing preset conditions;

[0532] The computing node of the computing task failed.

[0533] Optionally, the sending module is further configured to:

[0534] If the first node determines to switch the computing node, a first switch request is sent to the candidate target node. The first switch request is used to request to switch the computing node and includes the identifier of the computing task.

[0535] Optionally, the first handover request further includes at least one of the following:

[0536] The identifier of the source computing node; the maximum failure rate of the computing task; the maximum computing power consumption of the computing task; the minimum computing energy efficiency of the computing task; the minimum computing speed of the computing task; the minimum computing intensity of the computing task; the minimum computing efficiency of the computing task; and the maximum computing latency of the computing task.

[0537] Optionally, after the first node sends a first handover request to the candidate target node, the receiving module is further configured to:

[0538] Receive a first switching response sent by the candidate target node, the first switching response being used to indicate whether to agree to switch computing nodes.

[0539] Optionally, the sending module is further configured to:

[0540] When the first node determines to switch the computing node, the first node sends a first switch command to the source computing node. The first switch command carries a first switch indication, which is used to indicate the switching of the computing node.

[0541] Optionally, the sending module is further configured to:

[0542] When the first node determines to switch the computing node, the first node sends a second switching command to the first access network device. The second switching command carries a second switching indication, which is used to indicate the switching of the computing node. The first access network device is the service access network device of the terminal associated with the computing task.

[0543] Optionally, the second switching command further includes at least one of the following:

[0544] The information includes the identifier of the computing task, the identifier of the target computing node, and the switching time indication information.

[0545] Optionally, the processing module is further configured to: determine whether to switch the service access network device of the terminal associated with the computing task based on the first information;

[0546] The sending module is further configured to send a second switching request to the second access network device when the first node determines to switch the service access network device, the second switching request being used to request the second access network device to be switched to the service access network device.

[0547] Optionally, the second handover request includes at least one of the following:

[0548] The minimum bit rate corresponding to the computing task; the maximum bit rate corresponding to the computing task; the maximum latency corresponding to the computing task.

[0549] Optionally, after sending the second handover request to the second access network device, the receiving module is further configured to:

[0550] The system receives a second handover response from the second access network device, the second handover response indicating whether it agrees to switch to the serving access network device.

[0551] Optionally, the computing service request includes at least one of the following:

[0552] The following parameters are specified: computing service identifier; resource type of computing resources required by the computing task; minimum number of operations required by the computing task; minimum computing speed required by the computing task; minimum computing intensity required by the computing task; computing latency budget required by the computing task; maximum failure rate of the computing task; average window of the computing task statistics; maximum duration of the computing task; computing power type of the computing task; data type of the computing task; minimum memory required by the computing task; minimum storage space required by the computing task; minimum transmission bandwidth required by the computing task; arrival mode of the computing task; parameters of the arrival mode of the computing task; training accuracy of the AI ​​model corresponding to the computing task; AI performance parameters of the AI ​​model corresponding to the computing task; minimum inference throughput of the AI ​​model corresponding to the computing task; computing power consumption threshold of the computing task; computing energy efficiency threshold of the computing task.

[0553] Referring to Figure 13, when the compute node switching device is a terminal or a component within a terminal, the compute node switching device 800 includes:

[0554] Sending module 801 is used to send first measurement information or a calculation service request to the first node;

[0555] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0556] Optionally, the device further includes:

[0557] The receiving module is used to receive the first handover measurement configuration information sent by the first node.

[0558] Referring to Figure 14, when the computing node switching device is a first access network device or a component of the first access network device, the computing node switching device 900 includes:

[0559] The sending module 901 is used to send second measurement information to the first node, wherein the first access network device is the service access network device of the terminal associated with the computing task;

[0560] The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0561] Optionally, the device further includes:

[0562] The receiving module is used to receive the second handover measurement configuration information sent by the first node.

[0563] Optionally, the receiving module is further configured to:

[0564] The system receives a second switching command sent by the first node. The second switching command carries a second switching indication, which is used to indicate the switching of computing nodes.

[0565] Referring to Figure 15, when the compute node switching device is a source compute node or a component within a source compute node, the compute node switching device 1000 includes:

[0566] The sending module 1001 is used to send third measurement information to the first node;

[0567] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0568] Optionally, the device further includes:

[0569] The receiving module is used to receive the third handover measurement configuration information sent by the first node.

[0570] Optionally, the receiving module is further configured to:

[0571] The system receives a first switching command sent by the first node, the first switching command carrying a first switching indication, the first switching indication being used to indicate the switching of computing nodes.

[0572] Referring to Figure 16, when the compute node switching device is a candidate target node or a component within a candidate target node, the compute node switching device 1100 includes:

[0573] The receiving module 1101 is used to receive a first switching request sent by the first node. The first switching request is used to request a switching of computing nodes and includes the identifier of the computing task.

[0574] Optionally, the device further includes:

[0575] The sending module is used to send a first switching response to the first node, the first switching response being used to indicate whether the switching of computing nodes is agreed upon.

[0576] Referring to Figure 17, when the computing node switching device is a second access network device or a component of the second access network device, the computing node switching device 1200 includes:

[0577] The receiving module 1201 is used to receive a second handover request sent by the first node. The second handover request is used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

[0578] Optionally, the device further includes:

[0579] The sending module is used to send a second handover response to the first node, the second handover response being used to indicate whether to agree to switch to the service access network device.

[0580] The computing node switching device provided in this application embodiment can implement the various processes implemented in the method embodiments of Figures 7 to 11 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0581] As shown in Figure 18, this application embodiment also provides a communication device 1300, including a processor 1301 and a memory 1302. The memory 1302 stores programs or instructions that can run on the processor 1301. For example, when the communication device 1300 is a first node, when the program or instructions are executed by the processor 1301, they implement the various steps of the above-described computing node switching method embodiment applied to the first node, and achieve the same technical effect. When the communication device 1300 is a terminal, when the program or instructions are executed by the processor 1301, they implement the various steps of the above-described computing node switching method embodiment applied to the terminal, and achieve the same technical effect. To avoid repetition, this will not be repeated here. When the communication device 1300 is a first access network device, when the program or instructions are executed by the processor 1301, they implement the various steps of the above-described computing node switching method embodiment applied to the first access network device, and achieve the same technical effect. To avoid repetition, this will not be repeated here. When the communication device 1300 is a source computing node, the program or instructions executed by the processor 1301 implement the various steps of the computing node switching method embodiment applied to the source computing node described above, and achieve the same technical effect. To avoid repetition, they will not be described again here. When the communication device 1300 is a candidate target node, the program or instructions executed by the processor 1301 implement the various steps of the computing node switching method embodiment applied to the candidate target node described above, and achieve the same technical effect. To avoid repetition, they will not be described again here. When the communication device 1300 is a second access network device, the program or instructions executed by the processor 1301 implement the various steps of the computing node switching method embodiment applied to the second access network device described above, and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0582] This application also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the method embodiment shown in FIG7. This terminal embodiment corresponds to the above-described terminal-side method embodiment, and all implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and can achieve the same technical effect. The terminal may be the computing node switching device shown in FIG13. Specifically, FIG19 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of this application.

[0583] The terminal 1400 includes, but is not limited to, at least some of the following components: radio frequency unit 1401, network module 1402, audio output unit 1403, input unit 1404, sensor 1405, display unit 1406, user input unit 1407, interface unit 1408, memory 1409, and processor 1410.

[0584] Those skilled in the art will understand that terminal 1400 may also include a power supply (such as a battery) for powering various components. The power supply may be logically connected to processor 1410 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The terminal structure shown in Figure 19 does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0585] It should be understood that, in this embodiment, the input unit 1404 may include a graphics processor 14041 and a microphone 14042. The graphics processor 14041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 1406 may include a display panel 14061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 1407 includes at least one of a touch panel 14071 and other input devices 14072. The touch panel 14071 is also called a touch screen. The touch panel 14071 may include a touch detection device and a touch controller. Other input devices 14072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0586] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 1401 can transmit it to the processor 1410 for processing; in addition, the radio frequency unit 1401 can send uplink data to the network-side device. Typically, the radio frequency unit 1401 includes, but is not limited to, antennas, amplifiers, transceivers, couplers, low-noise amplifiers, duplexers, etc.

[0587] The memory 1409 can be used to store software programs or instructions, as well as various data. The memory 1409 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 1409 may include volatile memory or non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 1409 in this embodiment includes, but is not limited to, these and any other suitable types of memory.

[0588] Processor 1410 may include one or more processing units; optionally, processor 1410 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 1410.

[0589] In one embodiment, the radio frequency unit 1401 is used for:

[0590] Send the first measurement information or computing service request to the first node;

[0591] The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

[0592] Optionally, the radio frequency unit 1401 is also used for:

[0593] Receive the first handover measurement configuration information sent by the first node.

[0594] In one implementation, when the source computing node is a terminal:

[0595] The radio frequency unit 1401 is used to: send third measurement information to the first node;

[0596] The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

[0597] Optionally, the radio frequency unit 1401 is further configured to: receive third switching measurement configuration information sent by the first node.

[0598] Optionally, the radio frequency unit 1401 is further configured to: receive a first switching command sent by the first node, the first switching command carrying a first switching indication, the first switching indication being used to indicate switching computing nodes.

[0599] In one implementation, when the candidate target node is a terminal:

[0600] The radio frequency unit 1401 is used to: receive a first switching request sent by a first node, the first switching request being used to request a switching of computing nodes, the first switching request including an identifier of a computing task.

[0601] Optionally, the radio frequency unit 1401 is further configured to: send a first handover response to the first node, the first handover response being used to indicate whether to agree to switch computing nodes.

[0602] It is understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant descriptions in Figures 7, 9 or 10 of the method embodiment, and achieve the same or corresponding technical effects. To avoid repetition, it will not be described again here.

[0603] This application also provides a network-side device, including a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps of the method embodiments shown in Figures 3, 7, 9, 10, or 11. This network-side device embodiment corresponds to the above-described network-side device method embodiments. All implementation processes and methods of the above-described method embodiments can be applied to this network-side device embodiment and achieve the same technical effects.

[0604] Specifically, this application embodiment also provides a network-side device, which may be the computing node switching device shown in Figures 12, 14, 15, 16, or 17. As shown in Figure 20, the network-side device 1500 includes: an antenna 1501, a radio frequency (RF) device 1502, a baseband device 1503, a processor 1504, and a memory 1505. The antenna 1501 is connected to the RF device 1502. In the uplink direction, the RF device 1502 receives information through the antenna 1501 and sends the received information to the baseband device 1503 for processing. In the downlink direction, the baseband device 1503 processes the information to be transmitted and sends it to the RF device 1502, which then processes the received information and transmits it through the antenna 1501.

[0605] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 1503, which includes a baseband processor.

[0606] The baseband device 1503 may include at least one baseband board, on which multiple chips are disposed, as shown in FIG20. One of the chips is, for example, a baseband processor, which is connected to the memory 1505 via a bus interface to call the program in the memory 1505 and execute the network device operation shown in the above method embodiment.

[0607] The network-side device may also include a network interface 1506, such as a Common Public Radio Interface (CPRI).

[0608] Specifically, the network-side device 1500 in this application embodiment further includes: instructions or programs stored in memory 1505 and executable on processor 1504. Processor 1504 calls the instructions or programs in memory 1505 to execute the methods executed by the modules shown in FIG12, FIG14, FIG15, FIG16 or FIG17 and achieve the same technical effect. To avoid repetition, they will not be described in detail here.

[0609] Specifically, this application embodiment also provides a network-side device. As shown in FIG21, the network-side device 1600 includes: a processor 1601, a network interface 1602, and a memory 1603. The network-side device may be a computing node switching device as shown in FIG12, FIG14, FIG15, FIG16, or FIG17. The network interface 1602 is, for example, a Common Public Radio Interface (CPRI).

[0610] Specifically, the network-side device 1600 in this application embodiment further includes: instructions or programs stored in memory 1603 and executable on processor 1601. Processor 1601 calls the instructions or programs in memory 1603 to execute the methods executed by the modules shown in FIG12, FIG14, FIG15, FIG16 or FIG17 and achieve the same technical effect. To avoid repetition, they will not be described in detail here.

[0611] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described computing node switching method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0612] The processor mentioned above is the processor in the terminal or network-side device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. In some examples, the readable storage medium may be a non-transient readable storage medium.

[0613] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described computing node switching method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0614] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0615] This application also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described computing node switching method embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0616] This application also provides a wireless communication system, including: a first node, a terminal, a first access network device, a source computing node, and a candidate target node; or, the wireless communication system includes: a first node, a terminal, a first access network device, a source computing node, a candidate target node, and a second access network device; or the wireless communication system includes: a first node, a terminal, and a second access network device. The first node can be used to execute the steps of the computing node switching method applied to the first node as described above; the terminal can be used to execute the steps of the computing node switching method applied to the terminal as described above; the first access network device can be used to execute the steps of the computing node switching method applied to the first access network device as described above; the source computing node can be used to execute the steps of the computing node switching method applied to the source computing node as described above; the candidate target node can be used to execute the steps of the computing node switching method applied to the candidate target node as described above; and the second access network device can be used to execute the steps of the computing node switching method applied to the second access network device as described above.

[0617] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0618] From the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of computer software products plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0619] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A method for switching compute nodes, comprising: The first node receives first information, which includes at least one of measurement information and a computing service request. The first node determines whether to switch the computing node used to perform the computing task based on the first information.

2. The method according to claim 1, wherein, Before the first node receives the first information, the method further includes at least one of the following: The first node sends first handover measurement configuration information to the terminal, wherein the measurement information includes first measurement information corresponding to the first handover measurement configuration information; The first node sends second handover measurement configuration information to the first access network device, wherein the first access network device is the service access network device of the terminal associated with the computing task, and the measurement information includes second measurement information corresponding to the second handover measurement configuration information; The first node sends third handover measurement configuration information to the source computing node, wherein the measurement information includes third measurement information corresponding to the third handover measurement configuration information.

3. The method according to claim 2, wherein, The first switching measurement configuration information includes at least one of the following: a first measurement object; a first measurement report configuration; a first measurement report trigger event; and related parameters of the first measurement report trigger event. or The second switching measurement configuration information includes at least one of the following: a second measurement object; a second measurement report configuration; a second measurement report trigger event; and related parameters of the second measurement report trigger event. or The third switching measurement configuration information includes at least one of the following: third measurement object; third measurement report configuration; third measurement report trigger event; and related parameters of the third measurement report trigger event.

4. The method according to claim 3, wherein, The first measurement report triggering event includes at least one of the following: The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions. The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions. The location of the terminal meets preset conditions; The AI ​​performance of the artificial intelligence (AI) model used for the computational task meets the preset conditions; The inference throughput of the AI ​​model used for the computational task meets the preset conditions; The remaining time of the computing task meets the preset conditions; The end-to-end computation latency of the computation task meets the preset conditions; The failure rate of the computation task meets the preset conditions; The computation speed of the computation task meets the preset conditions; The computational intensity of the computational task meets the preset conditions.

5. The method according to claim 3 or 4, wherein, The second measurement report triggering event includes at least one of the following: The uplink bandwidth or uplink bit rate corresponding to the computing task meets the preset conditions. The downlink bandwidth or downlink bit rate corresponding to the computing task meets the preset conditions. The uplink channel quality of the terminal meets the preset conditions.

6. The method according to any one of claims 3-5, wherein, The third measurement report triggering event includes at least one of the following: The failure rate of the computation task meets the preset conditions; The computational power consumption of the computational task meets the preset conditions; The computational energy efficiency of the computational task meets the preset conditions; The computing resource utilization rate of the computing task meets the preset conditions; The computation speed of the computation task meets the preset conditions; The computational efficiency of the computational task meets the preset conditions; The computational intensity of the computational task meets the preset conditions; The computation time delay of the computation task meets the preset conditions; The computing nodes of the computing task meet the load balancing preset conditions; The computing node of the computing task failed.

7. The method according to any one of claims 1-6, wherein, The method further includes: If the first node determines to switch the computing node, the first node sends a first switch request to the candidate target node. The first switch request is used to request to switch the computing node and includes the identifier of the computing task.

8. The method according to claim 7, wherein, The first switching request also includes at least one of the following: The identifier of the source computing node; the maximum failure rate of the computing task; the maximum computing power consumption of the computing task; the minimum computing energy efficiency of the computing task; the minimum computing speed of the computing task; the minimum computing intensity of the computing task; the minimum computing efficiency of the computing task; and the maximum computing latency of the computing task.

9. The method according to claim 7 or 8, wherein, After the first node sends a first handover request to the candidate target node, the method further includes: The first node receives a first switching response sent by the candidate target node, the first switching response being used to indicate whether to agree to switch computing nodes.

10. The method according to any one of claims 1-9, wherein, The method further includes: When the first node determines to switch the computing node, the first node sends a first switch command to the source computing node. The first switch command carries a first switch indication, which is used to indicate the switching of the computing node.

11. The method according to any one of claims 1-10, wherein, The method further includes: When the first node determines to switch the computing node, the first node sends a second switching command to the first access network device. The second switching command carries a second switching indication, which is used to indicate the switching of the computing node. The first access network device is the service access network device of the terminal associated with the computing task.

12. The method according to claim 11, wherein, The second switching command also includes at least one of the following: The information includes the identifier of the computing task, the identifier of the target computing node, and the switching time indication information.

13. The method according to any one of claims 1-12, wherein, The method further includes: The first node determines whether to switch the service access network device of the terminal associated with the computing task based on the first information; When the first node determines to switch the serving access network device, the first node sends a second switching request to the second access network device, the second switching request being used to request the second access network device to be switched to the serving access network device.

14. The method according to claim 13, wherein, The second handover request includes at least one of the following: The minimum bit rate corresponding to the computing task; the maximum bit rate corresponding to the computing task; the maximum latency corresponding to the computing task.

15. The method according to claim 13 or 14, wherein, After the first node sends a second handover request to the second access network device, the method further includes: The first node receives a second handover response sent by the second access network device, the second handover response being used to indicate whether it agrees to hand over to the serving access network device.

16. The method according to any one of claims 1-15, wherein, The computing service request includes at least one of the following: The following parameters are specified: computing service identifier; resource type of computing resources required by the computing task; minimum number of operations required by the computing task; minimum computing speed required by the computing task; minimum computing intensity required by the computing task; computing latency budget required by the computing task; maximum failure rate of the computing task; average window of the computing task statistics; maximum duration of the computing task; computing power type of the computing task; data type of the computing task; minimum memory required by the computing task; minimum storage space required by the computing task; minimum transmission bandwidth required by the computing task; arrival mode of the computing task; parameters of the arrival mode of the computing task; training accuracy of the AI ​​model corresponding to the computing task; AI performance parameters of the AI ​​model corresponding to the computing task; minimum inference throughput of the AI ​​model corresponding to the computing task; computing power consumption threshold of the computing task; computing energy efficiency threshold of the computing task.

17. A method for switching computing nodes, comprising: The terminal sends the first measurement information or computing service request to the first node; The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

18. The method according to claim 17, wherein, Before the terminal sends the first measurement information to the first node, the method further includes: The terminal receives the first handover measurement configuration information sent by the first node.

19. A method for switching computing nodes, comprising: The first access network device sends the second measurement information to the first node. The first access network device is the service access network device of the terminal associated with the computing task. The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

20. The method according to claim 19, wherein, Before the first access network device sends the second measurement information to the first node, the method further includes: The first access network device receives the second handover measurement configuration information sent by the first node.

21. The method according to claim 19 or 20, wherein, The method further includes: The first access network device receives a second handover command sent by the first node. The second handover command carries a second handover indication, which is used to indicate the switching of computing nodes.

22. A method for switching computing nodes, comprising: The source computing node sends the third measurement information to the first node; The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

23. The method according to claim 22, wherein, Before the source computing node sends the third measurement information to the first node, the method further includes: The source computing node receives the third switching measurement configuration information sent by the first node.

24. The method according to claim 22 or 23, wherein, The method further includes: The source computing node receives a first switching command sent by the first node. The first switching command carries a first switching indication, which is used to indicate the switching of computing nodes.

25. A method for switching computing nodes, comprising: The candidate target node receives a first switching request sent by the first node. The first switching request is used to request a switch of computing nodes and includes the identifier of the computing task.

26. The method according to claim 25, wherein, After the candidate target node receives the first handover request sent by the first node, the method further includes: The candidate target node sends a first switching response to the first node, the first switching response being used to indicate whether it agrees to switch computing nodes.

27. A method for switching computing nodes, comprising: The second access network device receives a second handover request sent by the first node. The second handover request is used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

28. The method according to claim 27, wherein, After the second access network device receives the second handover request sent by the first node, the method further includes: The second access network device sends a second handover response to the first node, the second handover response being used to indicate whether it agrees to hand over to the serving access network device.

29. A computing node switching device, comprising: A receiving module is configured to receive first information, the first information including at least one of measurement information and a computing service request; The processing module is used to determine whether to switch the computing node used to perform the computing task based on the first information.

30. The apparatus according to claim 29, wherein, The device further includes a transmitting module for at least one of the following: Send first handover measurement configuration information to the terminal, wherein the measurement information includes first measurement information corresponding to the first handover measurement configuration information; Send second handover measurement configuration information to a first access network device, wherein the first access network device is the serving access network device of the terminal associated with the computing task, and the measurement information includes second measurement information corresponding to the second handover measurement configuration information; Send third handover measurement configuration information to the source computing node, wherein the measurement information includes third measurement information corresponding to the third handover measurement configuration information.

31. The apparatus according to claim 29, wherein, The device further includes a transmitting module for... If the first node determines to switch the computing node, a first switch request is sent to the candidate target node. The first switch request is used to request to switch the computing node and includes the identifier of the computing task.

32. A computing node switching device, comprising: The sending module is used to send the first measurement information or calculation service request to the first node; The first measurement information is used to determine whether to switch the computing node used to perform the computing task, and the computing service request is used to determine whether to switch the computing node used to perform the computing task.

33. A computing node switching device, comprising: The sending module is used to send the second measurement information to the first node; The second measurement information is used to determine whether to switch the computing node used to perform the computing task.

34. A computing node switching device, comprising: The sending module is used to send third measurement information to the first node; The third measurement information is used to determine whether to switch the computing node used to perform the computing task.

35. A computing node switching device, comprising: The receiving module is used to receive a first switching request sent by the first node. The first switching request is used to request a switch of computing nodes, and the first switching request includes the identifier of the computing task.

36. A computing node switching device, comprising: The receiving module is used to receive a second handover request sent by the first node, the second handover request being used to request the second access network device to be switched to the service access network device of the terminal associated with the computing task.

37. An electronic device comprising a processor and a memory, the memory storing a program or instructions executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the computing node switching method as claimed in any one of claims 1-16, or the steps of the computing node switching method as claimed in any one of claims 17-18, or the steps of the computing node switching method as claimed in any one of claims 19-21, or the steps of the computing node switching method as claimed in any one of claims 22-24, or the steps of the computing node switching method as claimed in any one of claims 25-26, or the steps of the computing node switching method as claimed in any one of claims 27-28.

38. A readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the compute node switching method as claimed in any one of claims 1-16, or the steps of the compute node switching method as claimed in any one of claims 17-18, or the steps of the compute node switching method as claimed in any one of claims 19-21, or the steps of the compute node switching method as claimed in any one of claims 22-24, or the steps of the compute node switching method as claimed in any one of claims 25-26, or the steps of the compute node switching method as claimed in any one of claims 27-28.

39. A chip, the chip comprising a processor and a communication interface, the communication interface being coupled to the processor, the processor being configured to run a program or instructions to implement the steps of the computing node switching method as described in any one of claims 1-16, or the steps of the computing node switching method as described in any one of claims 17-18, or the steps of the computing node switching method as described in any one of claims 19-21, or the steps of the computing node switching method as described in any one of claims 22-24, or the steps of the computing node switching method as described in any one of claims 25-26, or the steps of the computing node switching method as described in any one of claims 27-28.

40. A computer program / program product stored in a storage medium, the computer program / program product being executed by at least one processor to implement the steps of the computing node switching method as claimed in any one of claims 1-16, or the steps of the computing node switching method as claimed in any one of claims 17-18, or the steps of the computing node switching method as claimed in any one of claims 19-21, or the steps of the computing node switching method as claimed in any one of claims 22-24, or the steps of the computing node switching method as claimed in any one of claims 25-26, or the steps of the computing node switching method as claimed in any one of claims 27-28.

Images