Computing power status reporting methods, devices, equipment, media and program products
By adopting a computing power status reporting mode that combines periodic reporting, event triggering, and on-demand querying on the network side, the problem of balancing the real-time nature and overhead of computing power status information in the 6G converged computing network is solved, and dynamic matching and efficient computing power status awareness are achieved.
Patent Information
- Application Number
- CN202611432464.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-15
- Publication Date
- 2026-11-17
AI Technical Summary
In 6G converged computing and communication networks, the real-time nature of terminal computing power status information and air interface signaling overhead cannot be balanced. Existing technologies lack dedicated reporting mechanisms, fixed frequencies lead to an imbalance between real-time performance and overhead, and there is a lack of on-demand query mechanisms on the network side.
Three computing power status reporting modes are adopted: periodic reporting, event-triggered reporting, and on-demand querying on the network side. By adaptively adjusting the reporting frequency of computing power status information, the frequency is dynamically matched with task requirements and network status.
It achieves real-time perception of terminal computing power status information while taking into account computing power overhead, and can automatically reduce air interface signaling overhead under high load conditions.
Smart Images

Figure CN123123952A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, medium and program product for reporting computing power status. Background Technology
[0002] With the fifth generation (5 th Generation 5G mobile communication technology and sixth generation (6G) th With the continuous development of Generation 6G mobile communication technology, terminals are evolving from traditional communication nodes to distributed computing nodes.
[0003] Currently, in scenarios where communication and computing power converge, terminals not only need to complete traditional communication protocol stack processing, but also need to undertake computationally intensive tasks such as artificial intelligence (AI) inference, immersive rendering, and real-time perception. In order to make optimal end-to-end network collaborative scheduling decisions, network devices need to perceive the computing power status of terminals in real time or near real time, including the types of currently available computing resources, computing resource utilization, types of computing tasks being executed, and execution progress.
[0004] Therefore, how to ensure that network devices obtain sufficient and timely computing power status information while avoiding excessive air interface signaling overhead caused by terminals frequently reporting computing power status information has become a technical problem that 6G converged computing and communication networks urgently need to solve. Summary of the Invention
[0005] This application provides a method, apparatus, device, medium, and program product for reporting computing power status, which is used for terminals to report computing power status information while taking into account real-time performance and computing power overhead.
[0006] In a first aspect, embodiments of this application provide a computing power status reporting method applied to a terminal. The method includes: receiving computing power status reporting configuration parameters from a network device; wherein the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; periodically sending first computing power status information of the terminal to the network device based on the periodic reporting parameters and the adaptive adjustment parameters; in the event of detecting a preset event, sending second computing power status information of the terminal to the network device based on the event-triggered reporting parameters; and in the event of receiving a query request from the network device, sending third computing power status information of the terminal to the network device; wherein the first computing power status information, the second computing power status information, and the third computing power status information are used by the network device for scheduling decisions.
[0007] The technical solution provided in this application offers at least the following advantages: When receiving resource reporting configuration information from network devices, it can send three types of computing power status information to the network devices based on three reporting modes: periodic reporting, event-triggered reporting, and on-demand query reporting from the network side. This achieves dynamic matching between reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Therefore, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0008] One possible implementation, which involves periodically sending the terminal's first computing power status information to the network device based on periodic reporting parameters and adaptive adjustment parameters, includes: determining a first reporting frequency based on the periodic reporting parameters; adjusting the first reporting frequency based on the adaptive adjustment parameters and the task type of the terminal's current task to obtain a second reporting frequency; and periodically sending the first computing power status information to the network device based on the second reporting frequency.
[0009] Another possible implementation, the above computing power status reporting method further includes: receiving a load adjustment instruction from a network device, the load adjustment instruction being used to instruct the terminal to adjust the computing power status information reporting frequency; adjusting a second reporting frequency based on the load adjustment instruction to obtain a third reporting frequency; and sending the terminal's computing power status information to the network device based on the third reporting frequency.
[0010] Another possible implementation, in the case of detecting a preset event, sending the terminal's second computing power status information to the network device based on the event triggering reporting parameters, includes: when a preset event is detected and the time interval between the detection of the preset event and the terminal's previous sending of computing power status information to the network device for the preset event is greater than or equal to a first time interval, sending the terminal's second computing power status information to the network device based on the event triggering reporting parameters.
[0011] Another possible implementation is that the above periodic reporting parameters include at least one of the following: whether periodic reporting is enabled, periodic reporting interval, granularity of periodic reporting content, and maximum number of bytes per periodic report; and / or, the event-triggered reporting parameters include at least one of the following: whether event triggering is enabled, length of event configuration list, and list of event triggering conditions; and / or, the adaptive adjustment parameters include at least one of the following: adaptive mode, minimum reporting interval, maximum reporting interval, and default task priority.
[0012] Another possible implementation is that the above query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; the third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0013] Secondly, embodiments of this application provide a computing power status reporting method applied to a network device. The method includes: sending computing power status reporting configuration parameters to a terminal; wherein the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; receiving first computing power status information from the terminal, the first computing power status information being sent by the terminal based on the periodic reporting parameters and the adaptive adjustment parameters; receiving second computing power status information from the terminal, the second computing power status information being sent by the terminal based on the event-triggered reporting parameters when a preset event is detected; receiving third computing power status information from the terminal when a query request is sent to the terminal; and making scheduling decisions based on the first computing power status information, the second computing power status information, and the third computing power status information.
[0014] The technical solution provided in this application offers at least the following advantages: When sending computing power status reporting configuration parameters and query requests to the terminal, it can receive three types of computing power status information from the terminal based on three computing power status reporting modes. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Therefore, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0015] One possible implementation of the above-mentioned computing power status reporting method further includes: obtaining current network load information, terminal subscription information, and terminal service requirement information; and generating computing power status reporting configuration parameters based on the current network load information, terminal subscription information, and terminal service requirement information.
[0016] Another possible implementation is that the above query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; the third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0017] Thirdly, embodiments of this application provide a computing power status reporting device applied to a terminal, comprising: a receiving module and a sending module; the receiving module is configured to receive computing power status reporting configuration parameters from a network device; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; the sending module is configured to periodically send the terminal's first computing power status information to the network device based on the periodic reporting parameters and the adaptive adjustment parameters; the sending module is further configured to send the terminal's second computing power status information to the network device based on the event-triggered reporting parameters when a preset event is detected; the sending module is further configured to send the terminal's third computing power status information to the network device when the receiving module receives a query request from the network device; wherein, the first computing power status information, the second computing power status information, and the third computing power status information are used by the network device for scheduling decisions.
[0018] One possible implementation is that the aforementioned sending module is specifically used to: determine a first reporting frequency based on periodic reporting parameters; adjust the first reporting frequency based on adaptive adjustment parameters and the task type of the terminal's current task to obtain a second reporting frequency; and periodically send first computing power status information to the network device based on the second reporting frequency.
[0019] In another possible implementation, the sending module is further configured to: after the receiving module receives a load adjustment instruction from the network device, which instructs the terminal to adjust the computing power status information reporting frequency, adjust the second reporting frequency based on the load adjustment instruction to obtain a third reporting frequency; and send the terminal's computing power status information to the network device based on the third reporting frequency.
[0020] Another possible implementation is that the aforementioned sending module is specifically used to: when a preset event is detected and the time interval between the terminal's previous sending of computing power status information to the network device for the preset event is greater than or equal to a first time interval, send the terminal's second computing power status information to the network device based on the event triggering reporting parameters.
[0021] Another possible implementation is that the above periodic reporting parameters include at least one of the following: whether periodic reporting is enabled, periodic reporting interval, granularity of periodic reporting content, and maximum number of bytes per periodic report; and / or, the event-triggered reporting parameters include at least one of the following: whether event triggering is enabled, length of event configuration list, and list of event triggering conditions; and / or, the adaptive adjustment parameters include at least one of the following: adaptive mode, minimum reporting interval, maximum reporting interval, and default task priority.
[0022] Another possible implementation is that the above query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; the third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0023] Fourthly, embodiments of this application provide a computing power status reporting device applied to a network device, comprising: a sending module, a receiving module, and a processing module; the sending module is used to send computing power status reporting configuration parameters to a terminal; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; the receiving module is used to receive first computing power status information from the terminal, the first computing power status information being sent by the terminal based on the periodic reporting parameters and the adaptive adjustment parameters; the receiving module is also used to receive second computing power status information from the terminal, the second computing power status information being sent by the terminal based on the event-triggered reporting parameters when a preset event is detected; the receiving module is also used to receive third computing power status information from the terminal when the sending module sends a query request to the terminal; the processing module is used to make scheduling decisions based on the first computing power status information, the second computing power status information, and the third computing power status information.
[0024] In one possible implementation, the aforementioned processing module is further configured to obtain current network load information, terminal subscription information, and terminal service requirement information before the sending module sends the computing power status reporting configuration parameters to the terminal; and generate computing power status reporting configuration parameters based on the current network load information, terminal subscription information, and terminal service requirement information.
[0025] Another possible implementation is that the above query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; the third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0026] Fifthly, this application provides 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 method of the first aspect or the method of the second aspect.
[0027] In a sixth aspect, this application provides a readable storage medium on which a program or instructions are stored, which, when executed by a computer, implement the method of the first aspect or the method of the second aspect.
[0028] In a seventh aspect, this application provides a computer program product that is stored in a storage medium and, when executed by a computer, implements the method of the first aspect or the method of the second aspect.
[0029] Eighthly, embodiments of this application provide a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method described in the first aspect, or to implement the method described in the second aspect above.
[0030] The beneficial effects of aspects five through eight above are described in the corresponding descriptions of aspect one or aspect two, and will not be repeated here. Attached Figure Description
[0031] Figure 1 A schematic diagram of the network architecture for an application of a computing power status reporting method provided in this application embodiment;
[0032] Figure 2 A flowchart illustrating a computing power status reporting method provided in an embodiment of this application;
[0033] Figure 3 A flowchart illustrating another computing power status reporting method provided in this application embodiment;
[0034] Figure 4 A flowchart illustrating another computing power status reporting method provided in this application embodiment;
[0035] Figure 5 A flowchart illustrating another computing power status reporting method provided in this application embodiment;
[0036] Figure 6 A flowchart illustrating a computing power status reporting method provided in an embodiment of this application;
[0037] Figure 7 A flowchart illustrating another computing power status reporting method provided in this application embodiment;
[0038] Figure 8 A flowchart illustrating a computing power status reporting method provided in an embodiment of this application;
[0039] Figure 9 A flowchart illustrating a computing power status reporting method provided in an embodiment of this application;
[0040] Figure 10 A flowchart illustrating the network device reporting configuration distribution in a computing power status reporting method provided in this application embodiment;
[0041] Figure 11 This is a schematic diagram of the structure of a computing power status reporting system provided in an embodiment of this application;
[0042] Figure 12 A flowchart illustrating the periodic reporting of a computing power status reporting method provided in this application embodiment;
[0043] Figure 13 A flowchart illustrating the terminal event-triggered reporting method in a computing power status reporting method provided in this application embodiment;
[0044] Figure 14 A flowchart illustrating on-demand querying of network devices in a computing power status reporting method provided in this application embodiment;
[0045] Figure 15 A schematic diagram of the structure of a computing power status reporting device provided in this application embodiment;
[0046] Figure 16 A schematic diagram of the structure of a computing power status reporting device provided in this application embodiment;
[0047] Figure 17 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0048] 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.
[0049] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged 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," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.
[0050] The terms "at least one," "at least one of," etc., used in the specification and claims of this application refer to any one, any two, or a combination of two or more of the included items. For example, at least one of a, b, and c can mean: "a," "b," "c," "a and b," "a and c," "b and c," and "a, b, and c," where a, b, and c can be single or multiple. Similarly, "at least two" refers to two or more items, and its meaning is similar to that of "at least one."
[0051] In the description of this application, unless otherwise stated, "a plurality of" means two or more.
[0052] The computing power status reporting method, apparatus, device, medium, and program product provided in this application embodiment can be applied to scenarios such as integrated communication, sensing, and computing, autonomous driving, and wireless communication.
[0053] With the continuous development of 5G and 6G mobile communication technologies, User Equipment (UE) is evolving from a traditional "data pipeline terminal" to a "distributed computing node." Under the vision of converged computing and communication, the terminal side not only needs to handle traditional communication protocol stack processing but also needs to undertake computationally intensive tasks such as AI inference, immersive rendering, and real-time perception. To make optimal end-to-end network collaborative scheduling decisions, the network side needs to perceive the computing power status of the terminal side in real-time or near real-time, including the types and utilization rates of currently available computing resources, the types of computing tasks being executed, and their progress. These computing resource types include, but are not limited to: Graphics Processing Units (GPUs), Neural-network Processing Units (NPUs), and Field-Programmable Gate Arrays (FPGAs).
[0054] However, reporting the computing power status on the edge faces completely different challenges than reporting the communication status: the communication status—Reference Signal Received Power (RSRP), Reference Signal Receiving Quality (RSRQ), Radio Resource Control (RRC) status, or registration status—changes at a low frequency, ranging from seconds to minutes; while computing status (such as AI inference progress, GPU utilization, and remaining task time) changes at a high frequency, ranging from milliseconds; and the edge may run multiple computing tasks simultaneously, each with its own independent status change trajectory. Therefore, designing a mechanism that ensures the network side receives sufficiently timely computing status information while avoiding excessive air interface signaling overhead due to frequent reporting has become a critical issue that 6G converged computing and communication networks urgently need to address.
[0055] Currently, there are three relevant technical solutions. Solution 1 is a 5G RRC measurement reporting mechanism. In this solution, the UE needs to report channel quality measurement results (RSRP / RSRQ / signal-to-noise ratio) to the base station (gNB) for mobility decisions and resource scheduling. This mechanism supports two reporting methods: one is periodic reporting, where the UE reports measurement results periodically according to the measurement period configured by the network side (e.g., 120ms, 240ms, 480ms, etc.); the other is event-triggered reporting, where the UE triggers reporting when specific event conditions are met (e.g., the neighboring cell's RSRP is one offset higher than the serving cell's). Solution 2 involves the 5G operating system (OS) querying the modem communication status through the Radio Interface Layer (RIL) in a polling manner (e.g., the pollStateInternal() function). This includes: getOperator(): querying the current public land mobile network registration status, querying the packet-switched domain registration status (getDataRegistrationState), and querying the circuit-switched domain registration status (getVoiceRegistrationState). Meanwhile, the modem can also proactively notify the OS when its state changes via the RIL_UNSOL_* event. Option three involves the UE reporting its capabilities (such as supported frequency bands, slicing, security algorithms, etc.) via a registration request during the 5G registration process. This is a one-time, static capability declaration.
[0056] However, the above-mentioned related technical solutions have the following drawbacks:
[0057] (1) Lack of a dedicated reporting mechanism for computing power status
[0058] The computing power status reporting mechanisms of the above three schemes are all designed for communication status and do not support the reporting of end-side computing power status: RRC measurement reporting only supports channel quality-related measurements (RSRP / RSRQ / signal-to-noise ratio) and cannot carry computing status information such as GPU / NPU utilization, AI inference progress, and remaining task time; RIL polling is only used by the OS to query the modem communication status and does not support the network side to query the end-side computing status; UE capability reporting is a one-time static announcement and cannot reflect the dynamic changes in computing status.
[0059] (2) Fixed frequency leads to an imbalance between real-time performance and overhead.
[0060] Existing mechanisms (such as RRC periodic measurement and RIL periodic polling) all use a fixed-frequency periodic reporting method, which cannot be dynamically adjusted according to the characteristics of the computing task and network load. For high real-time tasks (such as vehicle-to-everything (V2X) inference, which requires millisecond-level state awareness), the fixed period may be too low, and the network side cannot know the state changes in time. For low real-time tasks (such as background AI analysis, which involves state changes from seconds to minutes), the fixed period may be too high, causing unnecessary air interface signaling overhead. There is no mechanism that can adaptively adjust the reporting frequency according to the communication task type of Ultra-Reliable Low-Latency Communications (URLLC), Enhanced Mobile Broadband (eMBB), and Massive Machine-Type Communications (mMTC).
[0061] (3) Lack of a flexible query mechanism driven by the network side
[0062] In the existing mechanism, the network side cannot query the computing status of the end-user side "on demand": the network side can only passively wait for the UE to report at fixed intervals, or report when event conditions are met; when the network side needs to urgently obtain the status information of a certain computing task (such as a sudden end-to-end network collaborative decision), there is no dedicated query signaling available. The query response cannot be targeted at specific computing tasks or specific computing resources (such as querying only the GPU status, not the NPU status).
[0063] To address the aforementioned technical problems, this application provides a method, apparatus, device, medium, and program product for reporting computing power status. Upon receiving resource reporting configuration information from a network device, it can send three types of computing power status information to the network device based on three reporting modes: periodic reporting, event-triggered reporting, and network-side on-demand query reporting. This achieves dynamic matching between reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Therefore, the terminal's reporting of computing power status information can balance real-time performance and computing power consumption.
[0064] The following description, in conjunction with the accompanying drawings, details the computing power status reporting method, apparatus, device, medium, and program products provided in the embodiments of this application.
[0065] Figure 1 The network architecture of a computing power status reporting method provided in an embodiment of this application is illustrated. For example... Figure 1 As shown, the network architecture includes a network device 101 and a terminal 102. The network device 101 and the terminal 102 are interconnected.
[0066] In some embodiments of this application, the network device 101 may be a server or a computer. The server may be a single server or a server cluster composed of multiple servers. It should be noted that the embodiments of this application do not limit the specific device form of the computing power status reporting device 101. Figure 1 The computing power status reporting device 101 is used as an example of a single server.
[0067] In some embodiments, the terminal may be a mobile phone, tablet computer, laptop computer, handheld computer, personal computer (PC), ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., and the embodiments of this application do not specifically limit it. Figure 1 The example shown is a mobile phone, with terminal 102 as the case study.
[0068] In some embodiments of this application, network device 101 sends computing power status reporting configuration parameters to terminal 102; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; terminal 102 receives the computing power status reporting configuration parameters from network device 101 and periodically sends the terminal's first computing power status information to network device 101 based on the periodic reporting parameters and adaptive adjustment parameters; when a preset event is detected, terminal 102 sends terminal 102's second computing power status information to network device 101 based on the event-triggered reporting parameters; network device 101 receives the first and second computing power status information from terminal 102 and sends a query request to terminal 102; when the query request is received from network device 101, terminal 102 sends terminal 102's third computing power status information to network device 101; network device 101 receives the third computing power status information from terminal 102 and makes scheduling decisions based on the first, second, and third computing power status information.
[0069] It should be noted that the network architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As network architectures evolve, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.
[0070] See Figure 2 This is a flowchart illustrating a computing power status reporting method provided in an embodiment of this application. Figure 2 As shown, the computing power status reporting method provided in this application embodiment is applied to a terminal and can be implemented by the computing power status reporting device in the third aspect described above, specifically including the following steps 201 to 204.
[0071] Step 201: The computing power status reporting device receives the computing power status reporting configuration parameters from the network device.
[0072] The computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters.
[0073] In some embodiments of this application, computing power status includes, but is not limited to, computing status information such as GPU and NPU utilization, communication task progress, remaining communication time, and Quality of Service (QoS) priority.
[0074] In some embodiments of this application, the periodic reporting parameters include at least one of the following: whether periodic reporting is enabled, periodic reporting interval, granularity of periodic reporting content, and maximum number of bytes for periodic reporting; and / or, the event-triggered reporting parameters include at least one of the following: whether event triggering is enabled, length of event configuration list, and list of event triggering conditions; and / or, the adaptive adjustment parameters include at least one of the following: adaptive mode, minimum reporting interval, maximum reporting interval, and default task priority.
[0075] In some embodiments of this application, the periodic reporting parameter is used to trigger the reporting of computing power status at a configured period.
[0076] In some embodiments of this application, the periodic reporting interval can be the default of the computing power status reporting device, or it can be arbitrarily set by the user according to actual usage needs.
[0077] For example, taking the above-mentioned periodic reporting interval as arbitrarily set by the user according to actual usage needs, the periodic reporting interval can be 120ms, 240ms, or 480ms, etc.
[0078] In some embodiments of this application, the granularity of the periodic reporting content is determined by which computing power information (such as only the overall GPU utilization rate, or refined to the load, temperature, and memory fragmentation rate of each core) is included in each periodic report, in order to balance the reporting accuracy and signaling overhead.
[0079] In some embodiments of this application, the maximum number of bytes reported periodically is the maximum amount of data allowed to be occupied in a single periodic report (RRC signaling payload limit), which is used to prevent reported data from crowding out effective service data bandwidth.
[0080] In some embodiments of this application, the event triggering reporting parameters are used to calculate whether the communication task state change meets the preset event conditions.
[0081] In some embodiments of this application, the length of the event configuration list is the maximum number of entries for event triggering conditions (such as "NPU utilization is less than 20%) that can be monitored simultaneously within the terminal, in order to limit the resource consumption of concurrent detection by the terminal.
[0082] In some embodiments of this application, the event triggering condition list is a specific threshold rule (such as "remaining time of inference task < 5ms" or "sudden increase in video memory usage of 10%), which is used to trigger the terminal to actively send an event-type report when the condition is met.
[0083] In some embodiments of this application, adaptive adjustment parameters are used to automatically adjust the reporting frequency of the computing power status reporting device according to the environment of the communication task type (e.g., URLLC, eMBB, or mMTC).
[0084] In some embodiments of this application, adaptive modes include off, by task type, by network load, combined adaptive, etc.
[0085] In some embodiments of this application, the minimum reporting interval is the shortest time interval allowed between two reports (periodic or event-based).
[0086] For example, in the environment of URLLC communication task type, the minimum reporting interval is 2ms; in the environment of eMBB communication task type, the minimum reporting interval is 100ms; and in the environment of mMTC communication task type, the minimum reporting interval is 5000ms.
[0087] In some embodiments of this application, the maximum reporting interval is the longest interval between two reports (periodic or event-based).
[0088] For example, in the environment of URLLC communication task type, the maximum reporting interval is 10ms; in the environment of eMBB communication task type, the maximum reporting interval is 500ms; and in the environment of mMTC communication task type, the maximum reporting interval is 60000ms.
[0089] In some embodiments of this application, the default priority of a task is the baseline scheduling level of the communication task (such as high / medium / low), which is used by the network side to determine which communication task's computing power request should be processed first when there is a resource conflict.
[0090] Thus, a dedicated reporting mechanism for computing power status is established by periodically reporting parameters, event-triggered reporting parameters, and adaptively adjusting parameters. By collecting and configuring multi-source computing power status configuration parameters, a solid foundation is provided for terminals to report computing power status to network devices.
[0091] Step 202: The computing power status reporting device periodically sends the terminal's first computing power status information to the network device based on the periodic reporting parameters and adaptive adjustment parameters.
[0092] In some embodiments of this application, the first computing power status information is periodically reported computing power status information. Periodic reporting is a "safety net" mechanism to ensure that network devices can periodically obtain the computing power status of terminals, which is applicable to all communication task types. The first computing power status information includes system-level summary (total computing power utilization, total number of tasks), task-level information (progress of high-priority tasks), and resource-level information (GPU / NPU / FPGA load rate), etc.
[0093] In some embodiments of this application, combined with Figure 2 ,like Figure 3 As shown, step 202 above can be implemented through steps 202a to 202c.
[0094] Step 202a: The computing power status reporting device determines the first reporting frequency based on the periodic reporting parameters.
[0095] In some embodiments of this application, the first reporting frequency is the basic reporting frequency of the terminal periodically reporting the computing power status, and the first reporting frequency is usually 120ms, 240ms or 480ms, etc.
[0096] Step 202b: The computing power status reporting device adjusts the first reporting frequency based on the adaptive adjustment parameters and the task type of the terminal's current task to obtain the second reporting frequency.
[0097] In some embodiments of this application, the task type of the terminal currently executing the task is URLLC, eMBB, and mMTC. URLLC pursues extremely low reporting intervals and extremely high reliability (such as autonomous driving and remote surgery); eMBB pursues extremely high data rates and large traffic (such as 8K video and high-speed download); mMTC pursues ultra-large number of connections (millions per square kilometer) and extremely low power consumption (such as tens of millions of sensors and smart meters).
[0098] In some embodiments of this application, the second reporting frequency is the frequency of the first reporting frequency increased or decreased according to URLLC, eMBB and mMTC; if a high real-time URLLC task is performed, the second reporting frequency approaches the minimum reporting interval (100ms) to ensure real-time performance; if a low real-time mMTC task is performed, the second reporting frequency approaches the maximum reporting interval (500ms) to save signaling overhead.
[0099] Step 202c: The computing power status reporting device periodically sends the first computing power status information to the network device based on the second reporting frequency.
[0100] In some embodiments of this application, the first computing power status information can be reported via extended RRC measurement reports, uplink (UL) information transmission, or network attached storage (NAS) messages.
[0101] In this way, the computing power status reporting device determines the reporting frequency based on the periodic reporting parameters, and then adjusts the reporting frequency according to the adaptive adjustment parameters and the task type of the terminal's current task. The computing power status information is periodically sent to the network device at the adjusted reporting frequency. Based on the type of computing task, the same set of terminals can adaptively adjust the reporting strategy in different scenarios, which improves the dynamic adaptability of the terminal's reporting of computing power status information and the system's flexibility.
[0102] Step 203: When a preset event is detected, the computing power status reporting device sends the terminal's second computing power status information to the network device based on the event triggering reporting parameters.
[0103] In some embodiments of this application, the second computing power status information is the computing power status information reported by event triggering. The event triggering reporting is an "on-demand" mechanism, which is only reported when the conditions for calculating the task status trigger event type are met. The second computing power status information includes event type, triggered task identifier (Identity, ID), and event triggering reason, etc.
[0104] In some embodiments of this application, combined with Figure 2 ,like Figure 4 As shown, step 203 above can be implemented through step 203a as follows.
[0105] Step 203a: When a preset event is detected and the time interval between the terminal's previous sending of computing power status information to the network device in response to the preset event is greater than or equal to the first time interval, the computing power status reporting device sends the terminal's second computing power status information to the network device based on the event triggering reporting parameters.
[0106] In some embodiments of this application, preset events include progress estimation threshold, changes in computing load, latency prediction deviation, and task completion.
[0107] In some embodiments of this application, the first time interval is a minimum time interval threshold for determining whether an event triggers the reporting of status information.
[0108] In some embodiments of this application, the first time interval may be the default setting of the computing power status reporting device, or it may be arbitrarily set by the user according to actual usage needs.
[0109] For example, the first time interval can be arbitrarily set by the user according to actual usage needs. This first time interval can be 1 second, 2 seconds, or 3 seconds, etc.
[0110] For example, if the computing load of a communication task changes and the GPU utilization rate jumps across a preset percentage threshold from 50% to 70%, reaching the preset event trigger threshold of 20%, and the time interval between the last event trigger and the reporting of the event is 5 seconds, which is greater than the first time interval of 3 seconds, the parameter requirements for event trigger reporting are met. The terminal will report the computing power status information, including the event type, the triggering task identifier (Identity, ID), the original GPU prediction value, the new GPU prediction value, and the reason for the event triggering, to the network device.
[0111] In some embodiments of this application, when no preset event is detected, or when a preset event is detected but the time interval is less than the first time interval, the computing power status reporting device does not send the terminal's second computing power status information to the network device.
[0112] Thus, when a preset event is detected and the time interval between the terminal's previous transmission of computing power status information to the network device for the preset event is greater than or equal to the first time interval, the terminal's second computing power status information is sent to the network device based on the event-triggered reporting parameters by setting a time interval threshold, thereby reducing the communication signaling overhead of the terminal in reporting computing power status information.
[0113] Step 204: Upon receiving a query request from a network device, the computing power status reporting device sends the terminal's third computing power status information to the network device.
[0114] Among them, the first computing power status information, the second computing power status information, and the third computing power status information are used by network devices to make scheduling decisions.
[0115] In some embodiments of this application, the third computing power status information is specific computing power status information obtained by the network device according to the query request. The third computing power status information includes task ID, GPU, NPU, and remaining time of communication task, etc.
[0116] In some embodiments of this application, scheduling decision-making is a process by which the network dynamically decides, among multiple computing units, "when, what frequency band, and how much power a certain computing unit transmits data" based on the current channel quality, service requirements, resource load, and QoS level.
[0117] In some embodiments of this application, the query request includes at least one of the following: query identifier, query type, target task identifier, target computing power resource type identifier, query timeout time, and maximum response byte count; third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0118] In some embodiments of this application, the query identifier is a unique serial number for this query task, used to match the response returned by the UE, and to prevent the request and response from being out of order.
[0119] In some embodiments of this application, the query types include system-level summary, specific task, specific resource, etc.
[0120] In some embodiments of this application, the target task identifier specifies which specific computing task to query, which is used for accurate positioning and avoids the UE reporting redundant information about irrelevant tasks.
[0121] In some embodiments of this application, the target computing resource type is identified as the type of hardware to be queried (e.g., querying only the NPU, not the GPU). This is used to reduce UE data collection overhead, retrieving only the resource data that the network is currently interested in.
[0122] In some embodiments of this application, the query timeout is the longest time limit for which the network device is willing to wait for the terminal to report the status information of the third computing power.
[0123] In some embodiments of this application, the query timeout can be the default value of the computing power status reporting device, or it can be arbitrarily set by the user according to actual usage needs.
[0124] For example, taking the query timeout time mentioned above as an example where the user arbitrarily sets it according to actual usage needs, the query timeout time can be 2 seconds, 3 seconds, or 4 seconds, etc.
[0125] In some embodiments of this application, the maximum response byte count is the maximum length of the data packet that the terminal is allowed to reply with.
[0126] In some embodiments of this application, the response status is the terminal's processing result code for this query (such as success, resource busy, task not found).
[0127] In some embodiments of this application, the response data size is the number of bytes of the actual payload in the response. This is used by network devices to verify the integrity of computing power status information.
[0128] In some embodiments of this application, the response content is the actual computing power status information carried.
[0129] In this way, by defining query requests and query responses and establishing rule-based communication signaling, intelligent computing power scheduling is achieved, enabling terminals to report computing power status information in real time based on query requests, thereby improving the efficiency of terminals reporting computing power status information.
[0130] The computing power status reporting method provided in this application can send three types of computing power status information to the network device based on three modes: periodic reporting, event-triggered reporting, and network-side on-demand query reporting, when receiving resource reporting configuration information sent by the network device. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Thus, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0131] In some embodiments of this application, combined with Figure 3 ,like Figure 5 As shown, after step 203c above, the computing power status reporting method provided in this application embodiment may further include steps 203d to 203f.
[0132] Step 203d: The computing power status reporting device receives a load adjustment instruction from the network device. The load adjustment instruction is used to instruct the terminal to adjust the frequency of computing power status information reporting.
[0133] In some embodiments of this application, the load adjustment instruction is a downlink control signaling sent from the network device to the terminal. When the network device issues the instruction, it must explicitly specify the adjustment reporting frequency and determine the adjusted reporting frequency.
[0134] Step 203e: The computing power status reporting device adjusts the second reporting frequency based on the load adjustment indication to obtain the third reporting frequency.
[0135] In some embodiments of this application, the third reporting frequency is the frequency adjusted by increasing or decreasing the second reporting frequency.
[0136] Step 203f: The computing power status reporting device sends the terminal's computing power status information to the network device based on the third reporting frequency.
[0137] In some embodiments of this application, the computing power status information can be first computing power status information, second computing power status information, or third computing power status information.
[0138] In this way, by adjusting the reporting frequency through the load adjustment instructions of network devices, the terminal reports its computing power status information based on the adjusted reporting frequency. Through manual intervention to adjust scheduling decisions, the signaling efficiency of the terminal reporting computing power status information is improved.
[0139] See Figure 6 This is a flowchart illustrating a computing power status reporting method provided in an embodiment of this application. Figure 6 As shown, the computing power status reporting method provided in this application embodiment is applied to network devices and can be implemented by the computing power status reporting device in the fourth aspect above, specifically including the following steps 301 to 305.
[0140] Step 301: The computing power status reporting device sends the computing power status reporting configuration parameters to the terminal.
[0141] The computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters.
[0142] Step 302: The computing power status reporting device receives the first computing power status information from the terminal. The first computing power status information is sent by the terminal based on the periodic reporting parameters and adaptive adjustment parameters.
[0143] Step 303: The computing power status reporting device receives the second computing power status information from the terminal. The second computing power status information is sent by the terminal based on the event-triggered reporting parameters when a preset event is detected.
[0144] Step 304: When a query request is sent to the terminal, the computing power status reporting device receives third computing power status information from the terminal.
[0145] In some embodiments of this application, the query request includes at least one of the following: query identifier, query type, target task identifier, target computing power resource type identifier, query timeout time, and maximum response byte count; third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0146] In this way, by defining query requests and responses and establishing rule-based communication signaling, intelligent computing power scheduling is achieved. This enables network devices to obtain computing power status information reported by terminals in real time based on query requests, thereby improving the efficiency of network device scheduling decisions.
[0147] Step 305: The computing power status reporting device makes scheduling decisions based on the first computing power status information, the second computing power status information, and the third computing power status information.
[0148] The computing power status reporting method provided in this application can receive three types of computing power status information sent by the terminal based on three computing power status reporting modes when sending computing power status reporting configuration parameters and query requests to the terminal. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Thus, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0149] In some embodiments of this application, combined with Figure 6 ,like Figure 7 As shown, prior to step 301 above, the computing power status reporting method provided in this application embodiment may further include the following steps 306 and 307.
[0150] Step 306: The computing power status reporting device obtains the current network load information, the terminal's subscription information, and the terminal's service requirement information.
[0151] In some embodiments of this application, the current network load information includes information such as the current air interface resource utilization rate, uplink interference level, and available bandwidth of the network device.
[0152] In some embodiments of this application, the terminal's subscription information includes information such as allowed QoS levels, computing power service permissions, and maximum signaling overhead.
[0153] In some embodiments of this application, the terminal's service requirement information includes information such as the terminal's current AI inference task, the application being run, and latency and computing power.
[0154] Step 307: The computing power status reporting device generates computing power status reporting configuration parameters based on the current network load information, the terminal's subscription information, and the terminal's service requirements information.
[0155] In some embodiments of this application, the theoretical lower and upper limits of the reporting parameters are determined based on the current network load information, the terminal's subscription information, and the terminal's service requirement information (e.g., if the subscription allows 10ms and the service only requires 20ms, then 20ms is taken as the minimum value). The theoretical lower and upper limit values are then filled into the standardized configuration parameter model to obtain the computing power status reporting configuration parameters.
[0156] In this way, after obtaining the current network load information, the terminal's subscription information, and the terminal's service requirements information, the computing power status reporting configuration parameters are generated using this information. This prevents subsequent network air interface congestion from the root cause, eliminating the need to wait for signaling to accumulate before relying on load adjustment instructions for processing. As a result, the computing power status information reported by the terminal meets the real-time requirements.
[0157] For further descriptions of the computing power status reporting method provided in some embodiments of this application, please refer to the relevant descriptions in the above-described terminal-side method embodiments. To avoid repetition, they will not be repeated here.
[0158] See Figure 8 This is a flowchart illustrating a computing power status reporting method provided in an embodiment of this application. Figure 8 As shown, the computing power status reporting method provided in this application embodiment involves interaction between the terminal and the network device, specifically including the following steps 401 to 406.
[0159] Step 401: The network device sends the computing power status reporting configuration parameters to the terminal.
[0160] The computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters.
[0161] Step 402: The terminal receives the computing power status reporting configuration parameters from the network device, and periodically sends the terminal's first computing power status information to the network device based on the periodic reporting parameters and adaptive adjustment parameters.
[0162] Step 403: When a preset event is detected, the terminal sends its second computing power status information to the network device based on the event trigger reporting parameters.
[0163] Step 404: The network device receives the first computing power status information and the second computing power status information from the terminal, and sends a query request to the terminal.
[0164] Step 405: Upon receiving a query request from a network device, the terminal sends its third computing power status information to the network device.
[0165] Step 406: The network device receives the third computing power status information from the terminal and makes scheduling decisions based on the first computing power status information, the second computing power status information and the third computing power status information.
[0166] In some embodiments of this application, combined with Figure 8 ,like Figure 9 As shown, before step 401, the computing power status reporting method in this application embodiment may further include the following steps 407 and 408.
[0167] Step 407: The network device obtains the current network load information, the terminal's subscription information, and the terminal's service requirement information.
[0168] Step 408: The network device generates computing power status reporting configuration parameters based on the current network load information, the terminal's subscription information, and the terminal's service requirements information.
[0169] The computing power status reporting method provided in this application allows the terminal to send three types of computing power status information to the network device based on three modes: periodic reporting, event-triggered reporting, and network-side on-demand query reporting, when receiving resource reporting configuration information from the network device. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, the terminal can both perceive its computing power status in real time and automatically reduce computing power overhead under high load. Therefore, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0170] For further descriptions of the computing power status reporting methods provided in some embodiments of this application, please refer to the relevant descriptions in the above-described terminal-side and network device-side method embodiments. To avoid repetition, they will not be repeated here.
[0171] It should be noted that for specific details regarding some of the steps described above, please refer to the descriptions in the above embodiments, which will not be repeated here.
[0172] The following describes the computing power status reporting method of this application through specific embodiments.
[0173] I. Data structure definition for reported configuration:
[0174] This application provides a flowchart of the network device reporting configuration distribution in a computing power status reporting method, as shown in the embodiments. Figure 10As shown, the network side sends configuration parameters for computing power status reporting to the end side via extended RRC or NAS signaling. The configuration includes three parts: periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters.
[0175] Report configuration data structure:
[0176] The data structures of the ComputeReportConfig configuration for computing resource reporting include, but are not limited to:
[0177] 1) General configuration section: Configuration ID, Configuration validity period
[0178] 2) Configure periodic reporting: Whether to enable periodic reporting, the periodic reporting interval, the granularity of periodic reporting content (system-level summary, task-level, full report, etc.), and the maximum number of bytes per periodic report (for controlling overhead), etc.
[0179] 3) Configure event triggering and reporting settings: whether to enable event triggering, event configuration list length, event triggering condition list, etc.
[0180] 4) Configure three adaptive adjustment settings: adaptive mode (off, by task type, by network load, combined adaptive, etc.), minimum reporting interval (to prevent signaling storms), maximum reporting interval (to guarantee updates), and default task priority (when no task priority is configured).
[0181] The EventTriggerConfig data structure for event triggering configuration includes, but is not limited to:
[0182] Event type, target resource type, trigger threshold, hysteresis value (to prevent frequent triggering), minimum reporting interval for this type of event, and granularity of the reported content when triggered, etc.
[0183] II. Adaptive Frequency Adjustment Strategy
[0184] Adaptive frequency adjustment is based on three dimensions of information: task type (Level 1), network load indication (Level 2), and boundary constraints (Level 3). Level 1 determines the base frequency based on the type of the currently executing computation task: URLLC tasks use the highest frequency (min_report_interval) (the maximum reporting interval mentioned above), eMBB tasks use the default frequency, and mMTC tasks use the lowest frequency (max_report_interval) (the minimum reporting interval mentioned above). Level 2 fine-tunes the frequency based on load adjustment instructions from the network side: reducing the reporting frequency to alleviate signaling burden during high network load and increasing the reporting frequency to obtain more accurate status information during low network load. Level 3 constrains the adjusted frequency within the range of [min_report_interval, max_report_interval] to prevent extreme cases.
[0185] III. Selection Logic for Three Reporting Methods
[0186] (1) The network side needs to continuously monitor the computing power overview of the terminal side. It is recommended to report periodically, because the bottom-line mechanism ensures that the network side always has the latest data. The typical reporting content is a system-level summary (total computing power, total number of tasks).
[0187] (2) When the status of the calculation task changes significantly, it is recommended to report the event-triggered mode because it is highly timely and only triggered when there are valuable events. The typical reporting content is task-level details (specific event task ID, progress).
[0188] (3) When the network side needs to urgently obtain specific information, it is recommended to report the network side to query on demand. The reason is that it is precise and can be obtained on demand. Typical reporting content is targeted information (specific task or specific resource).
[0189] (4) When the calculation task is completed, the result needs to be sent. It is recommended to report the event-triggered report mode. The reason is that it immediately notifies the network side to prepare to receive. Typical reporting content includes task ID, result data size, and transmission channel requirements.
[0190] (5) The network load is extremely high and it is necessary to reduce the uplink burden. It is recommended to retain only event-triggered reporting in the reporting mode. The reason is that periodic reporting is suspended and only triggered when a critical event occurs. The typical reporting content is only critical event notification.
[0191] (6) When establishing a computing task for the first time, it is recommended to use periodic reporting and event triggering as the reporting mode. The reason is that the baseline is established in the early stage, and key changes are captured at the same time. Typical reporting content includes full task information and event notifications.
[0192] Thus, this application includes the following key points:
[0193] Key Point 1: The "Three-Mode Adaptive" Reporting Architecture for Edge Computing Power Status
[0194] The edge computing power status reporting is designed as an organic combination of three modes: "network configuration-driven periodic reporting, edge-driven event-triggered reporting, and network-side on-demand querying." These different modes complement and coordinate with each other in different scenarios. Periodic reporting provides a guaranteed update, event-triggered reporting provides immediate response, and on-demand querying provides accurate retrieval. All three are coordinated and scheduled through a unified computing power status manager.
[0195] Key Point 2: Adaptive Frequency Adjustment Based on Computational Task Characteristics
[0196] In related technologies, status reporting uses a fixed period, and an adaptive frequency adjustment mechanism based on the computing task type (URLLC / eMBB / mMTC) is introduced (but not limited to these three task types):
[0197] URLLC-type tasks (vehicle-to-everything inference, real-time control): employ high frequency (approaching minimum).
[0198] eMBB-type tasks (cloud rendering, video processing): using medium frequency (default configuration).
[0199] mMTC tasks (IoT sensing, backend computing): employ low frequency (approaching the maximum frequency).
[0200] Key Point 3: Coordinated Adjustment of Reporting Frequency Based on Network Load
[0201] The terminal dynamically adjusts its reporting frequency based on load adjustment instructions issued by the network side. When the network load is too high, it reduces the periodic reporting frequency and event trigger threshold for non-urgent tasks to proactively reduce the load on the air interface; when the network load recovers, it restores the default configuration.
[0202] Key Point 4: Minimum Interval Constraints and Hysteresis Design for Event-Triggered Reporting
[0203] To prevent signaling storms caused by event-triggered reporting, two types of protection mechanisms are introduced:
[0204] Minimum reporting interval constraint: The same type of event is only allowed to be reported once within a specified time interval.
[0205] Hysteresis: Hysteresis is introduced into the event triggering condition to prevent frequent triggering near the critical value.
[0206] Key Point 5: Precise Granularity Control of On-Demand Queries on the Network Side
[0207] The network side can select different granularities of information by query type: system-level summary (supports quick overview), specific task details (supports targeted judgment), and specific resource status (supports resource-level scheduling). This avoids a "one-size-fits-all" full query approach.
[0208] It should be noted that the description of this embodiment can be found in the detailed description of the above embodiments, and will not be repeated here.
[0209] It should be noted that the above-described method embodiments, or the various possible implementations of the method embodiments, can be executed individually, or, provided there is no conflict, they can be combined with each other. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any restrictions on this.
[0210] Figure 11 This is a schematic diagram of a computing power status reporting system provided in an embodiment of this application. Figure 11As shown, the computing power status reporting system 800 may include: a 6G converged computing terminal UE module 801 and a 6G converged computing network side (base station and core network side) module 802.
[0211] Among them, the 6G converged computing terminal UE module 801 is used to uniformly manage the scheduling of three types of reporting mechanisms through the computing power status manager:
[0212] (1) Periodic reporting scheduler is responsible for triggering reporting according to the configured period (mode 1: periodic reporting).
[0213] This application provides a flowchart of a terminal periodic reporting method for computing power status reporting, as shown in the embodiments of this application. Figure 12 As shown, periodic reporting is a "safety net" mechanism to ensure that the network side can periodically obtain the computing power status of the terminal side, applicable to all computing tasks. Periodic reporting is triggered by a timer in the terminal-side computing power status manager. The reported content includes system-level summary (total computing power utilization, total number of tasks), task-level information (progress of the Top N high-priority tasks), and resource-level information (GPU / NPU / FPGA load rate), etc. Before reporting, the computing power status manager will adaptively adjust the reporting interval according to the type of task currently being executed: if high real-time URLLC tasks are being executed, the interval will be moved closer to min_report_interval to ensure real-time performance; if all tasks are low real-time mMTC tasks, the interval will be moved closer to max_report_interval to save signaling overhead. If the network side issues a load adjustment instruction, the terminal will adjust accordingly. The reporting message can be carried through extended RRC MeasurementReport, UL Information Transfer, or NAS messages.
[0214] (2) The event trigger detector is responsible for monitoring whether the changes in the state of the computing task meet the preset event conditions (mode 2: event trigger reporting).
[0215] This application provides a flowchart of a terminal event-triggered reporting method for computing power status reporting, as shown in the embodiments of this application. Figure 13 As shown, event-triggered reporting follows an "on-demand" mechanism, reporting only when a "significant" change occurs in the computation task's state, balancing real-time performance with overhead. Event triggering scenarios are as follows:
[0216] Event type progress crosses threshold, task_progress crosses preset percentage threshold (to prevent ping-pong delay), for example, inference progress from 49% to 50% (threshold 50%), the reported content includes task ID, current progress, remaining time, QoS priority;
[0217] Event type: Change in computing load, GPU / NPU utilization crosses a preset percentage threshold (latency prevention ping-pong), for example, GPU load changes from 50% to 70% (threshold 20% change), the reported content includes resource type, current load rate, top 3 tasks occupying resources;
[0218] Event type delay prediction deviation, estimated_remain_time deviates from the previous prediction by more than the threshold. For example, the original prediction was 50ms, but the actual time is 80ms (deviation > 50%). The reporting content includes task ID, original prediction value, new prediction value and reason for deviation.
[0219] Event type: task completed, task_progress reaches 100% or task terminates early. Example: AI inference completed. Reported content includes task ID, total execution time and result data size.
[0220] (3) The query response processor is responsible for responding to on-demand query requests initiated by the network side.
[0221] This application provides a flowchart of a network device on-demand query in a computing power status reporting method, as shown in the embodiments of this application. Figure 14 As shown, network-side on-demand querying is a "precision-guided" mechanism that allows the network side to directly initiate queries to the endpoint when needed to obtain specific computing power status information. The parameters of the ComputeStatusQueryRequest include, but are not limited to: query ID, query type (system-level summary, specific task, specific resource, etc.), target task ID, target resource type ID, query timeout, and maximum response bytes (control overhead). The parameters of the ComputeStatusQueryResponse include, but are not limited to: query ID, response status (success, timeout, queried object does not exist, etc.), response data size, and response content.
[0222] The aforementioned reporting content assembler collects the necessary information from the heterogeneous computing power resource pool, assembles it into a reporting message, and sends it to the network side. This applies to the relevant schemes used in steps 201 and 204 above.
[0223] The 6G converged computing network side (base station and core network side) module 802 is used to send and receive reported configurations and initiate on-demand queries through the computing power status management unit of the computing management module (this module can be in the base station or the core network). This applies to the relevant schemes in steps 301 and 305 above.
[0224] It should be noted that for a detailed explanation of the steps performed by each module and their beneficial effects, please refer to the description in the above embodiments, which will not be repeated here.
[0225] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0226] This application embodiment can divide the computing power status reporting device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.
[0227] In some embodiments of this application, a computing power status reporting device is also provided. This computing power status reporting device may include one or more functional modules for implementing the computing power status reporting method of the above method embodiments.
[0228] For example, Figure 15 This is a schematic diagram of a computing power status reporting device provided in an embodiment of this application. Figure 15 As shown, the computing power status reporting device 900 is applied to a terminal and includes: a receiving module 901 and a sending module 902;
[0229] The aforementioned receiving module 901 is used to receive computing power status reporting configuration parameters from network devices; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters;
[0230] The aforementioned sending module 902 is used to periodically send the terminal's first computing power status information to the network device based on periodic reporting parameters and adaptive adjustment parameters;
[0231] The aforementioned sending module 902 is also used to send the terminal's second computing power status information to the network device based on the event triggering reporting parameters when a preset event is detected;
[0232] The aforementioned sending module 902 is further configured to send the terminal's third computing power status information to the network device when the receiving module receives a query request from the network device; wherein the first computing power status information, the second computing power status information, and the third computing power status information are used by the network device to make scheduling decisions.
[0233] The computing power status reporting device 900 provided in this application can send three types of computing power status information to the network device based on three modes: periodic reporting, event-triggered reporting, and network-side on-demand query reporting, when receiving resource reporting configuration information sent by the network device. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the computing power status of the terminal in real time and automatically reduce computing power overhead under high load. Thus, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0234] In some embodiments of this application, the sending module 902 is specifically used for: determining a first reporting frequency based on periodic reporting parameters; adjusting the first reporting frequency based on adaptive adjustment parameters and the task type of the task currently being executed by the terminal to obtain a second reporting frequency; and periodically sending first computing power status information to the network device based on the second reporting frequency.
[0235] In some embodiments of this application, the sending module 902 is further configured to: after the receiving module 901 receives a load adjustment instruction from the network device, the load adjustment instruction being used to instruct the terminal to adjust the computing power status information reporting frequency, adjust the second reporting frequency based on the load adjustment instruction to obtain a third reporting frequency; and send the terminal's computing power status information to the network device based on the third reporting frequency.
[0236] In some embodiments of this application, the sending module 902 is specifically used to: when a preset event is detected and the time interval between the terminal's previous sending of computing power status information to the network device for the preset event is greater than or equal to a first time interval, send the terminal's second computing power status information to the network device based on the event triggering reporting parameters.
[0237] In some embodiments of this application, the above-mentioned periodic reporting parameters include at least one of the following: whether periodic reporting is enabled, periodic reporting interval, granularity of periodic reporting content, and maximum number of bytes in periodic reporting; and / or, the event-triggered reporting parameters include at least one of the following: whether event triggering is enabled, length of event configuration list, and list of event triggering conditions; and / or, the adaptive adjustment parameters include at least one of the following: adaptive mode, minimum reporting interval, maximum reporting interval, and default task priority.
[0238] In some embodiments of this application, the query request includes at least one of the following: query identifier, query type, target task identifier, target computing power resource type identifier, query timeout time, and maximum response byte count; third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0239] For example, Figure 16 This is a schematic diagram of a computing power status reporting device provided in an embodiment of this application. Figure 16 As shown, the computing power status reporting device 500 is applied to a terminal and includes: a sending module 501, a receiving module 502, and a processing module 503;
[0240] The sending module 501 is used to send computing power status reporting configuration parameters to the terminal; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters;
[0241] The receiving module 502 is used to receive the first computing power status information from the terminal. The first computing power status information is sent by the terminal based on the periodic reporting parameters and the adaptive adjustment parameters.
[0242] The receiving module 502 is also used to receive second computing power status information from the terminal. The second computing power status information is sent by the terminal based on event-triggered reporting parameters when a preset event is detected.
[0243] The receiving module 502 is also used to receive third computing power status information from the terminal when the sending module 501 sends a query request to the terminal;
[0244] The processing module 503 is used to make scheduling decisions based on the first computing power status information, the second computing power status information and the third computing power status information.
[0245] The computing power status reporting device 500 provided in this application can receive three types of computing power status information sent by the terminal based on three computing power status reporting modes when sending computing power status reporting configuration parameters and query requests to the terminal. This achieves dynamic matching between the reporting frequency and task requirements and network status. By adaptively adjusting the reporting frequency of computing power status information, it can both perceive the terminal's computing power status in real time and automatically reduce computing power overhead under high load. Thus, the terminal's reporting of computing power status information can balance real-time performance and computing power overhead.
[0246] In some embodiments of this application, the processing module 503 is further configured to obtain current network load information, terminal subscription information and terminal service requirement information before the sending module 501 sends the computing power status reporting configuration parameters to the terminal; and generate computing power status reporting configuration parameters based on the current network load information, terminal subscription information and terminal service requirement information.
[0247] Another possible implementation is that the above query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; the third computing power status information is carried in the query response, and the query response includes at least one of the following: query identifier, response status, response data size, and response content.
[0248] It should be noted that the computing power status reporting device can implement all the processes implemented in the above method embodiments and achieve the same beneficial effects. To avoid repetition, it will not be described again here.
[0249] In the case where the functions of the integrated modules described above are implemented in hardware, this application provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 17 As shown, the electronic device 90 includes: a processor 92, a communication interface 93, and a bus 94. Optionally, the electronic device 90 may also include a memory 91.
[0250] Processor 92 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.
[0251] Communication interface 93 is used to connect with other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.
[0252] The memory 91 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.
[0253] As one possible implementation, the memory 91 can exist independently of the processor 92. The memory 91 can be connected to the processor 92 via a bus 94 and is used to store instructions or program code. When the processor 92 calls and executes the instructions or program code stored in the memory 91, it can implement the computing power status reporting method provided in this application embodiment.
[0254] In another possible implementation, memory 91 can also be integrated with processor 92.
[0255] Bus 94 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 94 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 17 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0256] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.
[0257] 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 power status reporting method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.
[0258] 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.
[0259] This application also provides a readable storage medium storing a program or instructions. When executed by a computer, the program or instructions implement the computing power status reporting method provided in the above embodiments. It is understood that all or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The readable storage medium can be any of the foregoing embodiments or memory. The readable storage medium can also be an external storage device of the service invocation device, such as a pluggable hard drive, SmartMedia Card (SMC), Secure Digital (SD) card, flash card, etc., equipped on the service invocation device. Further, the readable storage medium can include both internal storage units of the service invocation device and external storage devices. The readable storage medium is used to store the computer program and other programs and data required by the service invocation device. The readable storage medium can also be used to temporarily store data that has been output or will be output.
[0260] This application also provides a computer program product, which is stored in a storage medium and, when executed by a computer, implements the computing power status reporting method provided in the above embodiments.
[0261] 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.
[0262] Through 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 software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0263] 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 forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.
Claims
1. A method for reporting computing power status, characterized in that, Applied to terminals, including: Receive computing power status reporting configuration parameters from network devices; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; Based on the periodic reporting parameters and the adaptive adjustment parameters, the terminal's first computing power status information is periodically sent to the network device; Upon detecting a preset event, the terminal's second computing power status information is sent to the network device based on the event trigger reporting parameters. Upon receiving a query request from the network device, the terminal's third computing power status information is sent to the network device; The first computing power status information, the second computing power status information, and the third computing power status information are used by the network device to make scheduling decisions.
2. The computing power status reporting method according to claim 1, characterized in that, The step of periodically sending the terminal's first computing power status information to the network device based on the periodic reporting parameters and the adaptive adjustment parameters includes: Based on the periodic reporting parameters, a first reporting frequency is determined; Based on the adaptive adjustment parameters and the task type of the task currently being executed by the terminal, the first reporting frequency is adjusted to obtain the second reporting frequency; Based on the second reporting frequency, the first computing power status information is periodically sent to the network device.
3. The computing power status reporting method according to claim 2, characterized in that, The computing power status reporting method also includes: Receive a load adjustment instruction from the network device, the load adjustment instruction being used to instruct the terminal to adjust the frequency of reporting computing power status information; Based on the load adjustment indication, the second reporting frequency is adjusted to obtain the third reporting frequency; Based on the third reporting frequency, the computing power status information of the terminal is sent to the network device.
4. The computing power status reporting method according to claim 1, characterized in that, The step of sending the terminal's second computing power status information to the network device based on the event trigger reporting parameters when a preset event is detected includes: If a preset event is detected and the time interval between the terminal's previous transmission of computing power status information to the network device in response to the preset event is greater than or equal to a first time interval, the terminal's second computing power status information is transmitted to the network device based on the event trigger reporting parameters.
5. The computing power status reporting method according to any one of claims 1 to 4, characterized in that, The periodic reporting parameters include at least one of the following: whether periodic reporting is enabled, periodic reporting interval, granularity of periodic reporting content, and maximum number of bytes per periodic report. And / or, the event triggering reporting parameters include at least one of the following: whether event triggering is enabled, the length of the event configuration list, and the event triggering condition list; And / or, the adaptive adjustment parameters include at least one of the following: adaptive mode, minimum reporting interval, maximum reporting interval, and task default priority.
6. The computing power status reporting method according to any one of claims 1 to 4, characterized in that, The query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; The third computing power status information is carried in the query response, which includes at least one of the following: query identifier, response status, response data size, and response content.
7. A method for reporting computing power status, characterized in that, Applied to network devices, including: Send computing power status reporting configuration parameters to the terminal; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; Receive first computing power status information from the terminal, wherein the first computing power status information is sent by the terminal based on the periodic reporting parameters and the adaptive adjustment parameters; The terminal receives second computing power status information, which is sent by the terminal based on the event-triggered reporting parameters when a preset event is detected. When a query request is sent to the terminal, third computing power status information is received from the terminal; Scheduling decisions are made based on the first computing power status information, the second computing power status information, and the third computing power status information.
8. The computing power status reporting method according to claim 7, characterized in that, Before sending the computing power status reporting configuration parameters to the terminal, the computing power status reporting method further includes: Obtain current network load information, the terminal's subscription information, and the terminal's service requirements information; Based on the current network load information, the terminal's subscription information, and the terminal's service requirements information, the computing power status reporting configuration parameters are generated.
9. The computing power status reporting method according to claim 7 or 8, characterized in that, The query request includes at least one of the following: query identifier, query type, target task identifier, target computing resource type identifier, query timeout, and maximum response bytes; The third computing power status information is carried in the query response, which includes at least one of the following: query identifier, response status, response data size, and response content.
10. A computing power status reporting device, characterized in that, Applied to terminals, including: a receiving module and a transmitting module; The receiving module is used to receive computing power status reporting configuration parameters from network devices; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; The sending module is used to periodically send the terminal's first computing power status information to the network device based on the periodic reporting parameters and the adaptive adjustment parameters. The sending module is also used to send the terminal's second computing power status information to the network device based on the event trigger reporting parameters when a preset event is detected. The sending module is further configured to send the terminal's third computing power status information to the network device when the receiving module receives a query request from the network device; The first computing power status information, the second computing power status information, and the third computing power status information are used by the network device to make scheduling decisions.
11. A computing power status reporting device, characterized in that, Applied to network devices, including: transmitting module, receiving module and processing module; The sending module is used to send computing power status reporting configuration parameters to the terminal; wherein, the computing power status reporting configuration parameters include periodic reporting parameters, event-triggered reporting parameters, and adaptive adjustment parameters; The receiving module is used to receive first computing power status information from the terminal, wherein the first computing power status information is sent by the terminal based on the periodic reporting parameters and the adaptive adjustment parameters; The receiving module is further configured to receive second computing power status information from the terminal, wherein the second computing power status information is sent by the terminal based on the event-triggered reporting parameters when a preset event is detected; The receiving module is further configured to receive third computing power status information from the terminal when the sending module sends a query request to the terminal; The processing module is used to make scheduling decisions based on the first computing power status information, the second computing power status information, and the third computing power status information.
12. An electronic device, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the computing power status reporting method as described in any one of claims 1 to 6, or to implement the computing power status reporting method as described in any one of claims 7 to 9.
13. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a computer, implement the computing power status reporting method as described in any one of claims 1 to 6, or implement the computing power status reporting method as described in any one of claims 7 to 9.
14. A computer program product, characterized in that, The computer program product is stored in a storage medium, and when the computer program product is executed by a computer, it implements the computing power status reporting method as described in any one of claims 1 to 6, or implements the computing power status reporting method as described in any one of claims 7 to 9.
