Computing resource and task scheduling method and device, equipment and readable storage medium

Abstract

The invention provides a computing resource and task scheduling method, device and equipment and a storable medium, and the method applied to a computing control function comprises the steps: determining at least one candidate computing node for providing a computing service according to the computing power of a registered computing node and a computing task request sent by a computing service consumer, the computing node comprises at least one of a terminal, a base station, an edge computing node and a core network computing node; sending a computing resource state request to the candidate computing node, wherein the computing resource state request is used for requesting computing resource state information; determining at least one target computing node and a corresponding computing task in the candidate computing nodes according to the computing task demand and computing resource state information fed back by the candidate computing nodes; and sending the corresponding calculation task to the target calculation node. The computing control function can comprehensively consider the computing resource states of the computing nodes, unified scheduling of computing resources and tasks is carried out, and global efficient computing power resource utilization is achieved.

Description

Technical Field

[0001] The present invention relates to the field of wireless technology, and in particular to a computing resource and task scheduling method, device, equipment and readable storage medium. Background Art

[0002] There are a large number of computing resource nodes of different sizes throughout the network. Using computing resources can provide users with more efficient services. Computing resource nodes include edge computing nodes, terminals, and base stations. At present, the computing power of terminals and base stations is gradually increasing, and the remaining computing resources are increasing. Using the computing resources of terminals and base stations can optimize artificial intelligence (AI) model processing (AI model processing may include: AI model training, AI model reasoning, and AI optimization, etc.) or deploy third-party business applications, realize deep edge business deployment, and achieve lower business latency. However, there is no way to achieve efficient perception and utilization of computing resources in the existing technology. Summary of the invention

[0003] The purpose of the present invention is to provide a computing resource and task scheduling method, device, equipment and readable storage medium, which are used to solve the problem of low efficiency in the utilization of computing resources in the network in the prior art.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a computing resource and task scheduling method, which is applied to computing control functions, including:

[0005] Determine at least one candidate computing node providing computing services according to the computing capabilities of the registered computing nodes and the computing task requests sent by the computing service consumers, wherein the computing nodes include at least one of a terminal, a base station, an edge computing node, and a core network computing node;

[0006] Sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information;

[0007] Determine at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes;

[0008] Send the corresponding computing task to the target computing node.

[0009] Optionally, the method further comprises:

[0010] When initially establishing a communication connection with the base station, receiving base station computing capability information sent by the base station;

[0011] Alternatively, receiving the base station computing capability information sent by the base station through the NG interface;

[0012] The base station computing capability is registered according to the base station computing capability information.

[0013] Optionally, before receiving the base station computing capability information sent by the base station through the NG interface, the method further includes:

[0014] A base station computing capability request is sent to the base station through the NG interface, where the base station computing capability request is used to request base station computing capability information, where the base station computing capability request includes a base station computing capability subscription request, and the base station capability subscription request is based on periodic triggering or based on preset event triggering.

[0015] Optionally, the method further comprises:

[0016] Receiving terminal computing capability information sent by the terminal through a signaling process of non-access layer NAS signaling;

[0017] The terminal computing capability is registered according to the terminal computing capability information.

[0018] Optionally, the method further comprises:

[0019] Receiving terminal device information sent by the terminal, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity (IMEI);

[0020] According to the terminal device model and / or the International Mobile Equipment Identity code IMEI, the data storage function is queried for a pre-stored mapping relationship table to obtain the corresponding terminal computing capability information, wherein the mapping relationship table is a mapping relationship table between the terminal device model and / or the International Mobile Equipment Identity code IMEI and the terminal computing capability information;

[0021] The terminal computing capability is registered according to the terminal computing capability information.

[0022] Optionally, the computing task request includes one or more of a computing task deployment location range, a computing resource type, and a computing resource size;

[0023] The computing capacity includes computing resource type and computing resource size;

[0024] The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer includes:

[0025] According to the computing capabilities of the computing nodes and the computing task requests sent by the computing service consumers, computing nodes that are within the computing task deployment location range and meet the computing resource type and / or computing resource size are determined as candidate computing nodes.

[0026] Optionally, the computing task request includes a latency requirement, where the latency requirement is a latency requirement for the computing node to transmit the computing task result to the computing service consumer;

[0027] The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer further includes:

[0028] According to the computing capability of the computing node, latency information is acquired through a network data analysis unit or an AI function, where the latency information includes a latency for the computing node to transmit a computing result to the computing service consumer obtained through statistics or prediction by the network data analysis unit or the AI ​​function;

[0029] According to the latency requirement sent by the computing service consumer, at least one computing node whose latency meets the latency requirement is selected as a candidate computing node.

[0030] Optionally, the computing resource status request includes a computing resource status subscription request;

[0031] The sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information, includes:

[0032] A computing resource status subscription request is sent to the candidate computing node, where the computing resource status subscription request is used to subscribe to the computing resource status information of the candidate computing node within a time window T, where the computing resource status information is predicted by the candidate computing node based on historical computing resource information and / or determined based on the prediction results and according to a preset or configured computing resource sharing strategy.

[0033] Optionally, determining at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes includes:

[0034] According to the computing resource status information fed back by the candidate computing nodes, a quality prediction result of the computing task executed by the candidate computing nodes is obtained through a data analysis function and / or an AI function; according to the prediction result and the computing task requirement, at least one target computing node and a corresponding computing task are determined among the candidate computing nodes;

[0035] Alternatively, based on the computing resource status information fed back by the candidate computing node, the mobility prediction result of the candidate computing node is obtained through the computing service consumer; based on the mobility prediction result and the computing task requirements, at least one target computing node and the corresponding computing task are determined among the candidate computing nodes.

[0036] The embodiment of the present invention further provides a computing resource and task scheduling method, which is applied to a base station, and includes:

[0037] In the case of receiving a computing resource status request sent by a computing control function, obtaining computing resource status information according to the computing resource status request; wherein the base station that receives the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered base stations and the computing task request sent by the computing service consumer;

[0038] sending the computing resource status information to the computing control function;

[0039] When receiving a computing task sent by a computing control function, a computing result is generated according to the computing task; wherein the base station that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

[0040] Optionally, the method further comprises:

[0041] When initially establishing a communication connection with the computing control function, sending base station computing capability information to the computing control function;

[0042] Alternatively, the base station computing capability information is sent to the computing control function via an NG interface, where the base station computing capability information is used to register the base station computing capability.

[0043] Optionally, the sending the base station computing capability information to the computing control function through the NG interface includes:

[0044] According to the base station computing capacity request sent by the computing control function, the base station computing capacity information is fed back to the computing control function through the NG interface; wherein the base station computing capacity request is used to request base station computing capacity information, and the base station computing capacity request includes a base station computing capacity subscription request, and the base station capacity subscription request is based on periodic triggering or based on preset event triggering.

[0045] Optionally, the computing resource status request includes a computing resource status subscription request;

[0046] The obtaining computing resource status information according to the computing resource status request includes:

[0047] The computing resource status information corresponding to the candidate computing node within a time window T is calculated based on the prediction results obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, the computing resource status information corresponding to the candidate computing node within a time window T is calculated according to a preset or configured computing resource sharing strategy.

[0048] Optionally, the computing resource status information includes computing resource size;

[0049] The step of calculating the computing resource status information corresponding to the candidate computing node within a time window T based on the prediction result and according to a preset or configured computing resource sharing strategy includes:

[0050] Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0051] The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

[0052] The embodiment of the present invention further provides a computing resource and task scheduling method, which is applied to a terminal and includes:

[0053] In the case of receiving a computing resource status request sent by a computing control function, obtaining computing resource status information according to the computing resource status request; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered terminal and the computing task request sent by the computing service consumer;

[0054] sending the computing resource status information to the computing control function;

[0055] When receiving a computing task sent by a computing control function, a computing result is generated according to the computing task; wherein the terminal receiving the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and computing task requirements.

[0056] Optionally, the method further comprises:

[0057] The terminal computing capability information is sent to the computing control function through a signaling process of a non-access layer NAS signaling, where the terminal computing capability is used to register the terminal computing capability.

[0058] Optionally, the method further comprises:

[0059] Send terminal device information to the computing control function, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity code IMEI, the terminal device model and / or the International Mobile Equipment Identity code IMEI are used to query and obtain the corresponding terminal computing capability information, and the terminal computing capability information is used to register the terminal computing capability.

[0060] Optionally, the computing resource status request includes a computing resource status subscription request;

[0061] The obtaining computing resource status information according to the computing resource status request includes:

[0062] The computing resource status information corresponding to the candidate computing node within a time window T is calculated based on the prediction results obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, the computing resource status information corresponding to the candidate computing node within a time window T is calculated according to a preset or configured computing resource sharing strategy.

[0063] Optionally, the computing resource status information includes computing resource size;

[0064] The step of calculating the computing resource status information corresponding to the candidate computing node within a time window T based on the prediction result and according to a preset or configured computing resource sharing strategy includes:

[0065] Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0066] The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

[0067] The embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to computing control functions, and the device includes:

[0068] A first determination module is used to determine at least one candidate computing node providing computing services according to the computing capabilities of the registered computing nodes and the computing task requests sent by the computing service consumers, wherein the computing nodes include at least one of terminals, base stations, edge computing nodes and core network computing nodes;

[0069] A first sending module, used to send a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information;

[0070] A second determination module is used to determine at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes;

[0071] The second sending module is used to send the corresponding computing task to the target computing node.

[0072] The embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to a base station, and the device includes:

[0073] A first computing module, configured to, upon receiving a computing resource status request sent by a computing control function, obtain computing resource status information according to the computing resource status request; wherein the base station receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capabilities of the registered base stations and the computing task request sent by the computing service consumer;

[0074] A third sending module, used to send the computing resource status information to the computing control function;

[0075] The second computing module is used to generate a computing result according to the computing task when receiving the computing task sent by the computing control function; wherein the base station that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

[0076] The embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to a terminal, and the device includes:

[0077] a third computing module, configured to, upon receiving a computing resource status request sent by a computing control function, obtain computing resource status information according to the computing resource status request; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capability of the registered terminal and the computing task request sent by the computing service consumer;

[0078] A fourth sending module, used to send the computing resource status information to the computing control function;

[0079] The fourth computing module is used to generate a computing result according to the computing task when receiving the computing task sent by the computing control function; wherein, the terminal that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

[0080] An embodiment of the present invention also provides a computing resource and task scheduling device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the computing resource and task scheduling method as described in any one of the above items is implemented.

[0081] An embodiment of the present invention further provides a readable storage medium, comprising: a program is stored on the readable storage medium, and when the program is executed by a processor, the computing resource and task scheduling method as described in any one of the above items is implemented.

[0082] The beneficial effects of the above technical solution of the present invention are as follows:

[0083] In the above scheme, the computing control function performs a rough match of computing tasks according to the computing capabilities of the registered computing nodes, determines at least one candidate computing node, and determines at least one target computing node according to the computing resource status information sent by the candidate computing node, and matches and sends the computing tasks and the target computing nodes. The embodiment of the present invention can dynamically guarantee the computing resources used by the computing nodes for communication processing, and provide a guaranteed relatively stable supply of computing resources to the outside world. Therefore, the computing control function can comprehensively consider the computing resource status and network status of wireless networks such as base stations, terminals, core networks and edge computing, and perform unified scheduling of computing resources and tasks to achieve global and efficient utilization of computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 A flowchart of a computing resource and task scheduling method provided by an embodiment of the present invention;

[0085] Figure 2 A flowchart of a computing resource and task scheduling method provided by another embodiment of the present invention;

[0086] Figure 3 A flowchart of a computing resource and task scheduling method provided by another embodiment of the present invention;

[0087] Figure 4 A schematic diagram of the structure of the wireless access network control function and the core network computing control function based on the service-oriented architecture;

[0088] Figure 5 One of the flow charts of the computing resource and task scheduling method provided by the embodiment of the present invention;

[0089] Figure 6 The second flowchart of the computing resource and task scheduling method provided by the embodiment of the present invention is as follows;

[0090] Figure 7 A schematic diagram of the structure of a computing resource and task scheduling device provided by an embodiment of the present invention;

[0091] Figure 8 A schematic diagram of the structure of a computing resource and task scheduling device provided by another embodiment of the present invention;

[0092] Fig. 9 A schematic diagram of the structure of a computing resource and task scheduling device provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0093] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0094] like Figure 1 As shown, an embodiment of the present invention provides a computing resource and task scheduling method, which is applied to a computing control function (CCF), including:

[0095] Step S101, determining at least one candidate computing node providing computing services according to the computing capabilities of the registered computing nodes and the computing task requests sent by the computing service consumers, wherein the computing nodes include at least one of a terminal, a base station, an edge computing node, and a core network computing node;

[0096] In an embodiment of the present invention, the computing control function CCF includes but is not limited to the core network CCF, wherein the core network CCF can not only directly obtain the computing power of the core network computing nodes, reuse the relevant interfaces of the existing core network network exposure function (NEF) to obtain the computing power of the edge computing nodes, but also obtain the computing power of the terminal and the base station, realize the global perception of various computing nodes, and uniformly schedule the computing resources and tasks of each computing node, so as to realize the efficient utilization of various computing power resources existing in the network.

[0097] This embodiment further illustrates the method for determining candidate computing nodes. At least one candidate computing node providing computing services is determined according to the computing capabilities of registered computing nodes, computing task requests sent by computing service consumers, and other registered computing capabilities, wherein other registered computing capabilities include but are not limited to multi-access edge computing (MEC) capabilities and computing capabilities of other network elements in the core network, etc.; computing service consumers include but are not limited to at least one of third-party business applications, other core network elements, and network management platforms, wherein when the computing service consumer is a third-party business application, the third-party business application sends the computing task request to the computing control function through the Network Exposure Function (NEF); when the computing service consumer is other core network elements or a network management platform, the computing service consumer directly sends the computing task request to the computing control function;

[0098] The computing task request includes one or more of the following: relevant information of the computing task consumer, the type of computing task, the size of computing power resources required for the computing task, the execution time of the computing task, the location where the computing task is expected to be deployed, the computing algorithm model, information related to the input and output data of the computing, and the latency requirement for the computing task result to be transmitted to the computing service consumer; wherein the location where the computing task is expected to be deployed can be: a central cloud, an edge cloud, a base station or a terminal, and the location where the computing task is expected to be deployed can be the range of a geographical area, which is used to provide corresponding computing services for a certain localized area; the information related to the input and output data of the computing includes one or more of the following: the data identifier or type of the computing input data, the data identifier or type of the computing output data, the data size of the computing input data and the output data, the time granularity, etc.; the relevant information of the computing task consumer includes one or more of the following: the computing service consumer type, the identifier of the computing service consumer, the location information of the computing service consumer, etc.; wherein the location information of the computing service consumer can be latitude and longitude information or the public land mobile network (Public Land Mobile Network, PLMN) where the computing service consumer is located, the transmitter address (Transmitter Address, TA) and other information.

[0099] Step S102, sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information;

[0100] In an embodiment of the present invention, the computing resource status request includes a computing resource status subscription request or a computing resource status query request, which is used to request computing resource status information;

[0101] The computing resource status request also includes computing resource status measurement configuration information, where the computing resource status measurement configuration information includes one or more of the following: an identifier of a computing node to be measured, an identifier of a measurement task, a measurement time window T, a measurement object, a measurement method, a measurement start time and an end time, and a measurement reporting condition; wherein the measurement reporting condition is a preset event trigger report and a periodic report, and the measurement object is used to indicate the content of the computing resource status information to be measured and reported, and the measurement object includes one or more of the following: a computing type supported by a computing node, a size of available computing resources for each computing type, a size of storage resources, a time required to execute a computing task of a certain unit size, and (rate transmission size statistics or predictions) and / or (average transmission delay statistics or prediction results) between a computing node and a potential computing service consumer; and the measurement method includes one or more of the following:

[0102] Method 1: measure the average value of each information in the measurement object within the time window T;

[0103] Method 2: Measure the minimum value of each information in the measurement object within the time window T;

[0104] Method 3: The guaranteed value provided by each information in the measurement object within the measurement time window T can be determined.

[0105] The computing resource status information includes one or more of the following: computing type, available computing resource size, storage resource size, and (rate transmission size statistics or predictions) and / or (average transmission delay statistics or predictions) between the computing node and potential computing service consumers.

[0106] It should be noted that in step S102, when the candidate computing node is a terminal and the terminal is in an idle state, it is necessary to first initiate a paging process for the terminal user in the idle state. A computing resource status request can be sent after the paging process, or an indication information for instructing the candidate computing node to report computing resource status information can be added to the paging process message. The paging reason in the paging message indicates that the paging is triggered by a computing task.

[0107] Step S103, determining at least one target computing node and a corresponding computing task from among the candidate computing nodes according to the computing task requirements and the computing resource status information fed back by the candidate computing nodes;

[0108] In an embodiment of the present invention, at least one target computing node is determined from the candidate computing nodes according to the computing task requirements and computing resource status information, and a corresponding computing task is generated according to the computing task requirements; wherein the computing task requirements include one or more of the following: the type of computing task, the size of computing power resources required for the computing task, the execution time of the computing task, the computing algorithm model, the relevant information of the input and output data of the computing task, and the information of the recipient of the computing result;

[0109] Among them, the computing algorithm model can be an algorithm model in a pre-configured algorithm model library or a commonly used machine learning algorithm model, such as: Convolutional Neural Network (CNN) model, Recurrent Neural Networks (RNN) model, transformer model, etc., or it can be the Uniform Resource Locator (URL) address or mirror address information of the computing algorithm model; the relevant information of the computing result recipient at least includes the address information of the computing task result recipient, such as Internet Protocol Address (IP) or URL address, etc., and may also include the category and / or identification information of the recipient.

[0110] Step S104: Send the corresponding computing task to the target computing node.

[0111] In an embodiment of the present invention, a method for sending a computing task to a target computing node is described, wherein the method for sending to an edge computing node and a core network computing node is not limited, and the method for sending a task to a terminal target computing node is that a computing control function sends a computing task to a terminal target computing node by enhancing an existing non-access stratum (NAS) signaling, for example, by sending a computing task by adding a computing task assignment NAS signaling process;

[0112] The method for sending tasks to the target computing node of the base station is that the computing control function sends computing tasks to the target computing node of the base station by enhancing the existing NG interface (the interface between the wireless access network and the 5G core network), for example, sending computing tasks through a newly added computing task assignment signaling process.

[0113] Optionally, the method further comprises:

[0114] Receive the calculation result of the calculation task fed back by the target computing node.

[0115] In summary, in the embodiment of the present invention, the computing control function performs a rough match of computing tasks according to the computing capabilities of the registered computing nodes, determines at least one candidate computing node, and determines at least one target computing node according to the computing resource status information sent by the candidate computing node, and matches and sends the computing task and the target computing node. The embodiment of the present invention can dynamically guarantee the computing resources of the computing nodes for communication processing, and provide a guaranteed and relatively stable supply of computing resources to the outside. Therefore, the computing control function can comprehensively consider the computing resource status and network status of wireless networks such as base stations, terminals, core networks and edge computing, and perform unified scheduling of computing resources and tasks to achieve global and efficient utilization of computing resources.

[0116] In the embodiment of the present invention, before executing step S101, the computing control function first needs to register the computing capabilities of the computing nodes for selecting candidate computing nodes and target computing nodes. The registration method may be one or more of the following:

[0117] Mode 1: The base station and / or the terminal can register the computing capability by interacting with the Access and Mobility Management Function (AMF) entity based on the N2 interface and the N1 interface, or by completing the registration of the computing capability through the computing control function transmitted to the AMF;

[0118] Method 2: Based on the end-to-end service-oriented architecture design, directly interact with CCF to register computing capabilities.

[0119] The computing capability information required for registering computing capability includes: (1) the type and identifier of the computing node; for example, when the computing node type is a base station, the identifier of the computing node can be the ID of the base station, such as a base station identifier (gNB ID) or a global base station identifier (global gNB ID); when the computing node type is a terminal, the identifier of the computing node can be the terminal fifth-generation system temporary mobile user identity code (5G-S-TMSI).

[0120] Optionally, the computing capability information also includes at least one of the following:

[0121] (2) The types of computation supported by the compute node;

[0122] (3) The range of computing power resources or the level of computing power that the computing node can provide. The range of computing power resources can be represented by maximum and minimum value information. The range of computing power resources or the level of computing power can be provided for computing types.

[0123] (4) Calculate the location information of the node, such as the TA information to which the base station belongs or the latitude and longitude information of the base station;

[0124] (5) Calculating node mobility attributes, such as: stationary terminal or mobile terminal; the mobile terminal may also provide information related to the mobile speed, such as the level of mobile speed (fast, medium, and slow);

[0125] (6) Time information during which the computing resources of the computing node can be scheduled, which is used to indicate the time period during which the computing resources of the computing node can be shared, for example, from 12:00 p.m. to 6:00 a.m.

[0126] The following describes a specific method for registering computing capabilities of a base station and / or a terminal computing node to a computing control function:

[0127] Optionally, the method further comprises:

[0128] When initially establishing a communication connection with the base station, receiving base station computing capability information sent by the base station;

[0129] Alternatively, receiving the base station computing capability information sent by the base station through the NG interface;

[0130] The base station computing capability is registered according to the base station computing capability information.

[0131] In an embodiment of the present invention, a method for registering computing capacity of a base station to a computing control function is described, wherein the base station computing capacity information received through the NG interface may be a computing capacity registration actively sent by the base station, or may be base station computing capacity information fed back by a computing capacity query or request sent by the computing control function.

[0132] Optionally, before receiving the base station computing capability information sent by the base station through the NG interface, the method further includes:

[0133] A base station computing capability request is sent to the base station through the NG interface, where the base station computing capability request is used to request base station computing capability information, where the base station computing capability request includes a base station computing capability subscription request, and the base station capability subscription request is based on periodic triggering or based on preset event triggering.

[0134] In an embodiment of the present invention, the base station computing capacity request includes a base station computing capacity query request for requesting the current base station computing capacity information, or a base station computing capacity subscription request for multiple requests for the current base station computing capacity information; wherein, the preset events for triggering the computing control function to send the base station computing capacity subscription request include but are not limited to: the computing power availability of the base station changes, such as from non-shareable computing power to shareable computing power, or the size of the computing power resources that can be shared by the base station is greater than the first threshold, or the type of computing power that the base station can provide changes. It should be noted that the computing control function registers the computing capacity according to the computing capacity information sent by the base station for the first time, and the computing capacity information received thereafter is used to update the computing capacity.

[0135] Optionally, the method further comprises:

[0136] Receiving terminal computing capability information sent by the terminal through a signaling process of non-access layer NAS signaling;

[0137] The terminal computing capability is registered according to the terminal computing capability information.

[0138] In an embodiment of the present invention, the terminal computing capability registration can be achieved by receiving the terminal computing capability information carried in the service request (Service Request) process of NAS signaling, or by introducing new NAS signaling, such as the terminal computing capability (UE ComputingCapability) receiving terminal computing capability information reported by the terminal.

[0139] Optionally, the method further comprises:

[0140] Receiving terminal device information sent by the terminal, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity (IMEI);

[0141] According to the terminal device model and / or the International Mobile Equipment Identity code IMEI, the data storage function is queried for a pre-stored mapping relationship table to obtain the corresponding terminal computing capability information, wherein the mapping relationship table is a mapping relationship table between the terminal device model and / or the International Mobile Equipment Identity code IMEI and the terminal computing capability information;

[0142] The terminal computing capability is registered according to the terminal computing capability information.

[0143] The preset condition of this embodiment is that the computing capability information of the terminal device has been stored in the data storage function. Through the embodiment of the present invention, (2), (3) and (6) in the computing capability information can be obtained; the computing control function obtains the terminal computing capability information by searching the mapping relationship table between the terminal device model or IMEI and the terminal computing capability stored in the data storage function according to the IMEI and / or the model information of the equipment manufacturer and the equipment model sent by the terminal.

[0144] Optionally, the method further comprises:

[0145] After the computing capability is successfully registered, the computing capability information is stored in the data storage function.

[0146] In the embodiment of the present invention, after the computing capacity is successfully registered, the computing capacity information of the computing node is stored in the database of the data storage function, and when the computing control function receives the updated computing capacity information, the corresponding computing capacity information in the data storage function is also updated.

[0147] Optionally, the computing task request includes one or more of a computing task deployment location range, a computing resource type, and a computing resource size;

[0148] The computing capacity includes computing resource type and computing resource size;

[0149] The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer includes:

[0150] According to the computing capabilities of the computing nodes and the computing task requests sent by the computing service consumers, computing nodes that are within the computing task deployment location range and meet the computing resource type and / or computing resource size are determined as candidate computing nodes.

[0151] The embodiment of the present invention provides a method for determining a candidate computing node. When the computing capability of a computing node satisfies a computing task request, the computing node is determined as a candidate computing node.

[0152] Optionally, the computing task request includes a latency requirement, where the latency requirement is a latency requirement for the computing node to transmit the computing task result to the computing service consumer;

[0153] The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer further includes:

[0154] According to the computing capability of the computing node, latency information is acquired through a network data analysis unit or an AI function, where the latency information includes a latency for the computing node to transmit a computing result to the computing service consumer obtained through statistics or prediction by the network data analysis unit or the AI ​​function;

[0155] According to the latency requirement sent by the computing service consumer, at least one computing node whose latency meets the latency requirement is selected as a candidate computing node.

[0156] The embodiment of the present invention provides another method for determining candidate computing nodes. According to the latency requirement sent by the computing service consumer, a computing node whose latency for transmitting the computing result to the computing service consumer meets the latency requirement is selected as a candidate computing node. For example, a computing node whose latency is less than a preset latency threshold is selected as a candidate computing node. The utilization efficiency of computing resources can be improved, wherein the network data analysis unit can be a network data analysis function (NWDAF).

[0157] It should be noted that the above two methods for determining candidate computing nodes can be used simultaneously.

[0158] Optionally, the computing resource status request includes a computing resource status subscription request;

[0159] The sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information, includes:

[0160] A computing resource status subscription request is sent to the candidate computing node, where the computing resource status subscription request is used to subscribe to the computing resource status information of the candidate computing node within a time window T, where the computing resource status information is predicted by the candidate computing node based on historical computing resource information and / or determined based on the prediction results and according to a preset or configured computing resource sharing strategy.

[0161] In an embodiment of the present invention, computing resource status information within a time window T is subscribed to a candidate computing node, and the candidate node performs calculations and predictions based on the computing resource status measurement configuration information included in the computing resource status request, combined with historical computing resource information and / or preset or configured computing resource sharing strategies, to generate computing resource status information.

[0162] Optionally, determining at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes includes:

[0163] According to the computing resource status information fed back by the candidate computing nodes, a quality prediction result of the computing task executed by the candidate computing nodes is obtained through a data analysis function and / or an AI function; according to the prediction result and the computing task requirements, at least one target computing node and a corresponding computing task are determined among the candidate computing nodes;

[0164] Alternatively, based on the computing resource status information fed back by the candidate computing node, the mobility prediction result of the candidate computing node is obtained through the computing service consumer; based on the mobility prediction result and the computing task requirements, at least one target computing node and the corresponding computing task are determined among the candidate computing nodes.

[0165] In an embodiment of the present invention, a method for determining at least one target computing node and a corresponding computing task among the candidate computing nodes is specifically described, and the most suitable target computing node is selected to allocate computing tasks to improve efficiency, wherein the quality prediction result includes a quality of experience (Quality of Experience, QoE) prediction result and / or a quality of service (Quality of Service, QoS) prediction result.

[0166] like Figure 2 As shown, an embodiment of the present invention further provides a computing resource and task scheduling method, which is applied to a base station, including:

[0167] Step S201, in the case of receiving a computing resource status request sent by a computing control function, obtaining computing resource status information according to the computing resource status request; wherein the base station that receives the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered base stations and the computing task request sent by the computing service consumer;

[0168] Step S202, sending the computing resource status information to the computing control function;

[0169] Step S203, upon receiving a computing task sent by a computing control function, generating a computing result according to the computing task; wherein the terminal receiving the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and computing task requirements.

[0170] Optionally, the method further comprises:

[0171] The calculation result is sent to the calculation control function.

[0172] In an embodiment of the present invention, a computing control function performs a rough match of tasks among a large number of base stations with registered computing capabilities to determine candidate computing nodes; a base station determined as a candidate computing node obtains computing resource status information based on a computing resource status request issued, and sends the computing resource status information to the computing control function, so that the computing control function can determine a target computing node among the candidate computing nodes based on the computing resource status information; a base station determined as a target computing node performs calculations according to the issued computing tasks.

[0173] By adopting the embodiment of the present invention, the base station can perform unified scheduling of computing resources and tasks after comprehensively considering the computing control functions, thereby realizing global and efficient utilization of computing resources.

[0174] Optionally, the method further comprises:

[0175] When initially establishing a communication connection with the computing control function, sending base station computing capability information to the computing control function;

[0176] Alternatively, the base station computing capability information is sent to the computing control function via an NG interface, where the base station computing capability information is used to register the base station computing capability.

[0177] In an embodiment of the present invention, a method for registering computing capacity of a base station to a computing control function is described, wherein the base station can actively send base station computing capacity information through an NG interface, and can also feedback the base station computing capacity information through the NG interface after receiving a base station computing capacity request from the computing control function.

[0178] Optionally, the sending the base station computing capability information to the computing control function through the NG interface includes:

[0179] According to the base station computing capacity request sent by the computing control function, the base station computing capacity information is fed back to the computing control function through the NG interface; wherein the base station computing capacity request is used to request base station computing capacity information, and the base station computing capacity request includes a base station computing capacity subscription request, and the base station capacity subscription request is based on periodic triggering or based on preset event triggering.

[0180] In an embodiment of the present invention, a computing control function sends a base station computing capacity request to a base station, and the base station computing capacity request includes a computing capacity query request or a base station computing capacity subscription request, wherein the triggering conditions of the base station computing capacity subscription request based on preset event triggering include but are not limited to: changes in the computing power availability of the base station, for example, from non-shareable computing power to shareable computing power, or the size of the computing power resources that can be shared by the base station is greater than a first threshold, or the type of computing power that the base station can provide changes.

[0181] It should be noted that the computing capacity information sent by the base station for the first time is used to register the computing capacity of the base station, and the computing capacity information sent thereafter is used to update the computing capacity.

[0182] Optionally, the computing resource status request includes a computing resource status subscription request;

[0183] The obtaining computing resource status information according to the computing resource status request includes:

[0184] The computing resource status information corresponding to the candidate computing node within a time window T is calculated based on the prediction results obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, the computing resource status information corresponding to the candidate computing node within a time window T is calculated according to a preset or configured computing resource sharing strategy.

[0185] In an embodiment of the present invention, a method for obtaining computing resource status information is described. The computing resource status subscription request includes computing resource status measurement configuration information. The base station obtains a predicted computing resource status measurement result based on the computing resource status measurement configuration information and historical computing resource information, and calculates the computing resource status information within a time window T based on the computing resource status measurement result; or based on the predicted result of the above-mentioned computing resource status measurement, the computing resource status information within a time window T is determined according to a preset or configured computing resource sharing strategy.

[0186] Optionally, the computing resource status information includes computing resource size;

[0187] The step of calculating the computing resource status information corresponding to the candidate computing node within a time window T based on the prediction result and according to a preset or configured computing resource sharing strategy includes:

[0188] Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0189] The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

[0190] In the embodiment of the present invention, a method for calculating the size of computing resources in computing resource status information is described. The following is a more detailed example: N time samples T in the prediction time window T 1 to T N The remaining computing resources size is X 1 To X N, the ratio of resources used for communication computing resource protection is pre-set or configured to be 10%, then the size of computing resource status information that can be provided is 0.9X 1 Up to 0.9X N The third-party network elements include but are not limited to wireless network management functions or wireless computing power management and orchestration functions or wireless computing network management and orchestration functions or policy control units. It should be noted that the names of the above third-party network elements are used as examples. Any network element with similar functions and the same effect can be used. The present invention is not limited to this. Figure 3 As shown, an embodiment of the present invention further provides a computing resource and task scheduling method, which is applied to a terminal and includes:

[0191] Step S301, in the case of receiving a computing resource status request sent by a computing control function, computing resource status information according to the computing resource status request; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered terminal and the computing task request sent by the computing service consumer;

[0192] Step S302, sending the computing resource status information to the computing control function;

[0193] Step S303, upon receiving a computing task sent by a computing control function, generating a computing result according to the computing task; wherein the terminal receiving the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and computing task requirements.

[0194] Optionally, the method further comprises:

[0195] The calculation result is sent to the calculation control function.

[0196] In an embodiment of the present invention, a computing control function performs a rough match of tasks among a large number of terminals with registered computing capabilities to determine candidate computing nodes; a terminal determined as a candidate computing node sends computing resource status information based on a computing resource status request issued, so that the computing control function can determine a target computing node among the candidate computing nodes based on the computing resource status information; a terminal determined as a target computing node performs computing based on the computing task issued.

[0197] By adopting the embodiment of the present invention, the terminal can perform unified scheduling of computing resources and tasks after comprehensively considering the computing control functions, thereby realizing global and efficient utilization of computing resources.

[0198] Optionally, the method further comprises:

[0199] The terminal computing capability information is sent to the computing control function through a signaling process of a non-access layer NAS signaling, where the terminal computing capability is used to register the terminal computing capability.

[0200] In an embodiment of the present invention, the terminal computing capability registration can be achieved by carrying the terminal computing capability information in the service request (ServiceRequest) process to the computing control function in the NAS signaling, or by introducing new NAS signaling, such as the terminal computing capability (UE Computing Capability) to send the terminal computing capability information to the computing control function.

[0201] Optionally, the method further comprises:

[0202] Send terminal device information to the computing control function, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity code IMEI, the terminal device model and / or the International Mobile Equipment Identity code IMEI are used to query and obtain the corresponding terminal computing capability information, and the terminal computing capability information is used to register the terminal computing capability.

[0203] In the embodiment of the present invention, the computing control function obtains the terminal computing capability information by searching the mapping relationship table between the terminal device model or IMEI and the terminal computing capability stored in the data storage function according to the IMEI and / or the model information of the device manufacturer and the device model sent by the terminal. The preset condition of this embodiment is that the computing capability information of the terminal device has been stored in the data storage function.

[0204] Optionally, the computing resource status request includes a computing resource status subscription request;

[0205] The obtaining computing resource status information according to the computing resource status request includes:

[0206] The computing resource status information corresponding to the candidate computing node within a time window T is calculated based on the prediction results obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, the computing resource status information corresponding to the candidate computing node within a time window T is calculated according to a preset or configured computing resource sharing strategy.

[0207] In an embodiment of the present invention, a method for obtaining computing resource status information is described. The computing resource status subscription request includes computing resource status measurement configuration information. The terminal obtains a predicted computing resource status measurement result based on the computing resource status measurement configuration information and historical computing resource information, and calculates the computing resource status information within a time window T based on the computing resource status measurement result; or based on the predicted result of the above computing resource status measurement, the computing resource status information within a time window T is determined according to a preset or configured computing resource sharing strategy.

[0208] Optionally, the computing resource status information includes computing resource size;

[0209] The step of calculating the computing resource status information corresponding to the candidate computing node within a time window T based on the prediction result and according to a preset or configured computing resource sharing strategy includes:

[0210] Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0211] The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

[0212] In the embodiment of the present invention, a method for calculating the size of computing resources in computing resource status information is described. The following is a more detailed example: N time samples T in the prediction time window T 1 to T N The remaining computing resources size is X 1 To X N , the ratio of resources used for communication computing resource protection is pre-set or configured to be 10%, then the size of computing resource status information that can be provided is 0.9X 1 Up to 0.9X N . Third-party network elements include but are not limited to wireless network management functions or wireless computing power management and orchestration functions or wireless computing network management and orchestration functions or policy control units. It should be noted that the names of the above-mentioned third-party network elements are used as examples. As long as the network elements / functions have similar functions and can achieve the same effects, the present invention is not limited to this.

[0213] like Figure 4 As shown, the core network CCF interacts with the radio access network (RAN) through an end-to-end connection with the radio computing control function (RAN Computing Control Function, RCCF).

[0214] like Figure 5 and Figure 6 As shown, in Figure 4 On the basis of the two embodiments of using the method of the present invention to implement the core network CCF to uniformly schedule base station and terminal resources and tasks, the two embodiments are described respectively:

[0215] Example 1: Reference Figure 5 , the specific steps are as follows:

[0216] Step 1a: The terminal (UE) sends terminal equipment information to the core network CCF to register the computing capability of the terminal. The terminal equipment information includes information such as the IMEI and terminal equipment model of the terminal; and optionally, it may also include the available time preference information of the terminal computing resources;

[0217] Step 1b: The core network CCF queries the core network data storage function for the terminal computing capability information based on the terminal device information. It should be noted that the preset condition for executing this step is that the core network data storage function stores the terminal computing capability information table corresponding to different terminal device information;

[0218] Step 1c: The core network data storage function sends the corresponding terminal computing capability information to the core network CCF;

[0219] Step 1d: The RCCF of the base station sends the base station computing capability information to the core network CCF to register the computing capability of the base station;

[0220] Step 2: The core network CCF registers the computing capabilities of the terminal and the base station according to the computing capability information of the terminal and the base station obtained in step 1;

[0221] Step 3: After the computing capabilities of the terminal and the base station are successfully registered, the core network CCF sends the computing capability information of the terminal and the base station to the core network data storage function for storage;

[0222] Step 4a (optional): The core network CCF sends a computing capability information subscription request to the terminal;

[0223] Step 4b (optional): the core network CCF sends a request for subscription to the computing capability information of the base station RCCF;

[0224] Step 5a (optional): When the computing capability information changes, the terminal sends the computing capability information to the core network CCF to update the registered terminal computing capability;

[0225] Step 5b (optional): When the computing capability information changes, the base station RCCF sends the computing capability information to the core network CCF to update the registered base station computing capability;

[0226] Step 6: The third-party service application sends a computing task request to the core network NEF;

[0227] Step 7: The core network NEF sends a computing task request to the core network CCF;

[0228] Step 8: The core network CCF determines at least one candidate computing node from the registered terminals and base stations according to the computing task request and the computing capabilities of the registered base stations and terminals;

[0229] Step 9a: The core network CCF sends a computing resource subscription request to the candidate terminal computing node, where the computing resource subscription request includes computing resource state measurement configuration information;

[0230] Step 9b: The core network CCF sends a computing resource subscription request to the candidate base station computing node, where the computing resource subscription request includes computing resource state measurement configuration information; wherein the core network CCF can directly interact with the RCCF function of the base station through a service-oriented interface;

[0231] Step 10a: The candidate terminal computing node generates computing resource status information according to the computing resource status measurement configuration information, and sends the computing resource status information to the core network CCF according to the computing resource subscription request.

[0232] Step 10b: The base station terminal computing node generates computing resource status information according to the computing resource status measurement configuration information, and sends the computing resource status information to the core network CCF according to the computing resource subscription request.

[0233] Step 11: The core network CCF determines at least one target computing node from the candidate computing nodes according to the computing resource status information, and determines the computing task corresponding to the target computing node according to the computing task requirement;

[0234] Step 12a: The core network CCF sends the computing task to the target terminal computing node;

[0235] Step 12b: The core network CCF sends the computing task to the target base station computing node;

[0236] Step 13: The target terminal computing node and the target base station computing node respectively execute the received computing tasks;

[0237] Step 14a: The target terminal computing node feeds back the computing result to the core network CCF.

[0238] Step 14b: The target base station computing node feeds back the calculation result to the core network CCF.

[0239] Optionally, when executing Figure 5Before the step, the step also includes: the RCCF of the base station registers the wireless computing services that the RCCF can provide with the core network network storage function (NF Repository Function, NRF) or other network registration functions.

[0240] Optionally, the step further includes: the CCF obtains the wireless computing service provided by the RCCF of the base station through the NRF.

[0241] Example 2: Reference Figure 6 The specific steps are similar to those in the first embodiment, except that the computing service consumer that sends the computing task request is different, and the third-party service is changed to a base station and a terminal. The specific description is as follows:

[0242] Step 6 in Embodiment 1: the third-party service application sends a computing task request to the core network NEF;

[0243] Step 7: The core network NEF sends a computing task request to the core network CCF;

[0244] Replaced with step 6a: the terminal sends a computing task request to the core network CCF;

[0245] and step 6b: the base station sends a computing task request to the core network CCF;

[0246] The content of the following steps remains unchanged, but the step numbers change sequentially.

[0247] like Figure 7 As shown, an embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to computing control functions, and the device includes:

[0248] A first determination module 701 is used to determine at least one candidate computing node providing computing services according to the computing capabilities of the registered computing nodes and the computing task request sent by the computing service consumer, wherein the computing node includes at least one of a terminal, a base station, an edge computing node and a core network computing node;

[0249] A first sending module 702 is used to send a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information;

[0250] A second determination module 703 is used to determine at least one target computing node and a corresponding computing task from among the candidate computing nodes according to the computing task requirements and the computing resource status information fed back by the candidate computing nodes;

[0251] The second sending module 704 is used to send the corresponding computing task to the target computing node.

[0252] Optionally, the device further comprises:

[0253] A first receiving module, configured to receive base station computing capability information sent by the base station when initially establishing a communication connection with the base station;

[0254] Alternatively, receiving the base station computing capability information sent by the base station through the NG interface;

[0255] The first registration module is used to register the base station computing capability according to the base station computing capability information.

[0256] Optionally, the device further comprises:

[0257] The fifth sending module is used to send a base station computing capability request to the base station through the NG interface, wherein the base station computing capability request is used to request base station computing capability information, and the base station computing capability request includes a base station computing capability subscription request, and the base station capability subscription request is based on periodic triggering or based on preset event triggering.

[0258] Optionally, the device further comprises:

[0259] A second receiving module is used to receive the terminal computing capability information sent by the terminal through a signaling process of a non-access layer NAS signaling;

[0260] The second registration module is used to register the terminal computing capability according to the terminal computing capability information.

[0261] Optionally, the device further comprises:

[0262] A third receiving module is used to receive terminal device information sent by the terminal, wherein the terminal device information includes a terminal device model and / or an International Mobile Equipment Identity (IMEI);

[0263] A first acquisition module is used to query a pre-stored mapping relationship table from a data storage function according to the terminal device model and / or the international mobile equipment identity code IMEI to obtain corresponding terminal computing capability information, wherein the mapping relationship table is a mapping relationship table between the terminal device model and / or the international mobile equipment identity code IMEI and the terminal computing capability information;

[0264] The third registration module is used to register the terminal computing capability according to the terminal computing capability information.

[0265] Optionally, the first determining module 701 includes:

[0266] The first computing node determination unit is used to determine computing nodes that are located within the computing task deployment location range and meet the computing resource type and / or computing resource size as candidate computing nodes based on the computing capabilities of the computing nodes and the computing task requests sent by the computing service consumers.

[0267] Optionally, the first determining module 701 further includes:

[0268] A second acquisition unit is used to acquire latency information through a network data analysis unit or an AI function according to the computing capability of the computing node, wherein the latency information includes a latency for the computing node to transmit a computing result to the computing service consumer obtained by statistics or prediction by the network data analysis unit or the AI ​​function;

[0269] The second computing node determination unit is used to select at least one computing node whose latency meets the latency requirement as a candidate computing node according to the latency requirement sent by the computing service consumer.

[0270] Optionally, the first sending module 702 includes:

[0271] A resource status subscription unit is used to send a computing resource status subscription request to the candidate computing node, wherein the computing resource status subscription request is used to subscribe to the computing resource status information of the candidate computing node within a time window T, and the computing resource status information is obtained by the candidate computing node based on historical computing resource information prediction and / or based on the prediction result, determined according to a preset or configured computing resource sharing strategy.

[0272] Optionally, the second determining module 703 includes:

[0273] A third acquisition unit is used to obtain the quality prediction result of the candidate computing node executing the computing task through a data analysis function and / or an AI function according to the computing resource status information fed back by the candidate computing node;

[0274] A third computing node determination unit, which determines at least one target computing node and a corresponding computing task from among the candidate computing nodes according to the prediction result and the computing task requirement;

[0275] A fourth acquisition unit, configured to acquire the mobility prediction result of the candidate computing node through a computing service consumer according to the computing resource status information fed back by the candidate computing node;

[0276] The fourth computing node determination unit determines at least one target computing node and a corresponding computing task from among the candidate computing nodes according to the mobility prediction result and the computing task requirement.

[0277] It should be noted that the embodiment of the device is a device corresponding to the embodiment of the above method, and all implementation methods in the embodiment of the above method are applicable to the embodiment of the device and can achieve the same technical effect.

[0278] like Figure 8As shown, an embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to a base station, and the device includes:

[0279] The first computing module 801 is configured to compute resource status information according to a computing resource status request received from a computing control function; wherein the base station receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capabilities of the registered base stations and the computing task request sent by the computing service consumer;

[0280] A third sending module 802 is used to send the computing resource status information to the computing control function;

[0281] The second computing module 803 is used to generate a computing result according to the computing task when receiving the computing task sent by the computing control function; wherein the base station that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

[0282] Optionally, the device further comprises:

[0283] The sixth sending module is used to send base station computing capability information to the computing control function when initially establishing a communication connection with the computing control function; or to send base station computing capability information to the computing control function through an NG interface; wherein the base station computing capability information is used to register the base station computing capability.

[0284] Optionally, the sixth sending module includes:

[0285] A base station computing power information sending unit, configured to feed back the base station computing power information to the computing control function through an NG interface according to the base station computing power request sent by the computing control function; wherein the base station computing power request is used to request the base station computing power information, the base station computing power request includes a base station computing power subscription request, and the base station power subscription request is based on periodic triggering or based on preset event triggering;

[0286] Optionally, the first calculation module 801 includes:

[0287] The first computing unit is used to obtain the computing resource status information corresponding to the candidate computing node within a time window T according to the prediction results obtained by the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, obtain the computing resource status information corresponding to the candidate computing node within a time window T according to a preset or configured computing resource sharing strategy.

[0288] Optionally, the first computing unit includes:

[0289] A fifth acquisition unit, used to acquire a plurality of time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0290] The first determination unit is used to determine the computing resource size according to a preset or configured computing resource protection ratio and the time remaining computing resource size, and the computing resource protection ratio is configured through a third-party network element.

[0291] It should be noted that the embodiment of the device is a device corresponding to the embodiment of the above method, and all implementation methods in the embodiment of the above method are applicable to the embodiment of the device and can achieve the same technical effect.

[0292] like Fig. 9 As shown, an embodiment of the present invention further provides a computing resource and task scheduling device, which is applied to a terminal, and the device includes:

[0293] The third computing module 901 is used to calculate resource status information according to the computing resource status request when receiving the computing resource status request sent by the computing control function; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered terminal and the computing task request sent by the computing service consumer;

[0294] A fourth sending module 902, configured to send the computing resource status information to the computing control function;

[0295] The fourth computing module 903 is used to generate a computing result based on the computing task when receiving the computing task sent by the computing control function; wherein, the terminal that receives the computing task is the target computing node, and the target computing node is determined by the computing control function based on the computing resource status information and the computing task requirements.

[0296] Optionally, the device further comprises:

[0297] The seventh sending module is used to send the terminal computing capability information to the computing control function through the signaling process of the non-access layer NAS signaling, and the terminal computing capability is used to register the terminal computing capability.

[0298] Optionally, the device further comprises:

[0299] The eighth sending module is used to send terminal equipment information to the computing control function, and the terminal equipment information includes the terminal equipment model and / or the international mobile equipment identity code IMEI, and the terminal equipment model and / or the international mobile equipment identity code IMEI are used to query and obtain the corresponding terminal computing capability information, and the terminal computing capability information is used to register the terminal computing capability.

[0300] Optionally, the third calculation module 901 includes:

[0301] The second computing unit is used to obtain the computing resource status information corresponding to the candidate computing node within a time window T according to the prediction results obtained by the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction results, obtain the computing resource status information corresponding to the candidate computing node within a time window T according to a preset or configured computing resource sharing strategy.

[0302] Optionally, the second computing unit includes:

[0303] A sixth acquisition unit, used to acquire a plurality of time samples within a time window T in the prediction result and the corresponding time remaining computing resource size;

[0304] The second determination unit is used to determine the computing resource size according to a preset or configured computing resource protection ratio and the time remaining computing resource size, and the computing resource protection ratio is configured through a third-party network element.

[0305] It should be noted that the embodiment of the device is a device corresponding to the embodiment of the above method, and all implementation methods in the embodiment of the above method are applicable to the embodiment of the device and can achieve the same technical effect.

[0306] An embodiment of the present invention also provides a computing resource and task scheduling device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the computing resource and task scheduling method as described in any one of the above items is implemented.

[0307] An embodiment of the present invention further provides a readable storage medium, comprising: a program is stored on the readable storage medium, and when the program is executed by a processor, the computing resource and task scheduling method as described in any one of the above items is implemented.

[0308] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0309] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A computing resource and task scheduling method, characterized in that: Applied to computing control functions, including: Determine at least one candidate computing node providing computing services according to the computing capabilities of the registered computing nodes and the computing task requests sent by the computing service consumers, wherein the computing nodes include at least one of a terminal, a base station, an edge computing node, and a core network computing node; Sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information; Determine at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes; Send the corresponding computing task to the target computing node.

2. The computing resource and task scheduling method according to claim 1, characterized in that: The method further comprises: When initially establishing a communication connection with the base station, receiving base station computing capability information sent by the base station; Alternatively, receiving the base station computing capability information sent by the base station through the NG interface; The base station computing capability is registered according to the base station computing capability information.

3. The computing resource and task scheduling method according to claim 2, characterized in that: Before receiving the base station computing capability information sent by the base station through the NG interface, the method further includes: A base station computing capability request is sent to the base station through the NG interface, where the base station computing capability request is used to request base station computing capability information, where the base station computing capability request includes a base station computing capability subscription request, and the base station capability subscription request is based on periodic triggering or based on preset event triggering.

4. The computing resource and task scheduling method according to claim 1, characterized in that: The method further comprises: Receiving terminal computing capability information sent by the terminal through a signaling process of non-access layer NAS signaling; The terminal computing capability is registered according to the terminal computing capability information.

5. The computing resource and task scheduling method according to claim 1, characterized in that: The method further comprises: Receiving terminal device information sent by the terminal, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity (IMEI); According to the terminal device model and / or the International Mobile Equipment Identity code IMEI, the data storage function is queried for a pre-stored mapping relationship table to obtain the corresponding terminal computing capability information, wherein the mapping relationship table is a mapping relationship table between the terminal device model and / or the International Mobile Equipment Identity code IMEI and the terminal computing capability information; The terminal computing capability is registered according to the terminal computing capability information.

6. The computing resource and task scheduling method according to claim 1, characterized in that: The computing task request includes one or more of a computing task deployment location range, a computing resource type, and a computing resource size; The computing capacity includes computing resource type and computing resource size; The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer includes: According to the computing capabilities of the computing nodes and the computing task requests sent by the computing service consumers, computing nodes that are within the computing task deployment location range and meet the computing resource type and / or computing resource size are determined as candidate computing nodes.

7. The computing resource and task scheduling method according to claim 1, characterized in that: The computing task request includes a latency requirement, where the latency requirement is a latency requirement for a computing node to transmit a computing task result to a computing service consumer; The step of determining at least one candidate computing node providing computing services according to the computing capabilities of the computing nodes and the computing task request sent by the computing service consumer further includes: According to the computing capability of the computing node, latency information is acquired through a network data analysis unit or an AI function, where the latency information includes a latency for the computing node to transmit a computing result to the computing service consumer obtained through statistics or prediction by the network data analysis unit or the AI ​​function; According to the latency requirement sent by the computing service consumer, at least one computing node whose latency meets the latency requirement is selected as a candidate computing node.

8. The computing resource and task scheduling method according to claim 1, characterized in that: The computing resource status request includes a computing resource status subscription request; The sending a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information, includes: A computing resource status subscription request is sent to the candidate computing node, where the computing resource status subscription request is used to subscribe to the computing resource status information of the candidate computing node within a time window T, where the computing resource status information is predicted by the candidate computing node based on historical computing resource information and / or determined based on the prediction results and according to a preset or configured computing resource sharing strategy.

9. The computing resource and task scheduling method according to claim 1, characterized in that: The step of determining at least one target computing node and a corresponding computing task from among the candidate computing nodes according to the computing task requirements and the computing resource status information fed back by the candidate computing nodes includes: According to the computing resource status information fed back by the candidate computing nodes, a quality prediction result of the computing task executed by the candidate computing nodes is obtained through a data analysis function and / or an AI function; according to the quality prediction result and the computing task requirement, at least one target computing node and a corresponding computing task are determined among the candidate computing nodes; Alternatively, based on the computing resource status information fed back by the candidate computing node, the mobility prediction result of the candidate computing node is obtained through the computing service consumer; based on the mobility prediction result and the computing task requirements, at least one target computing node and the corresponding computing task are determined among the candidate computing nodes.

10. A method for scheduling computing resources and tasks, characterized in that: Applied to base stations, including: In the case of receiving a computing resource status request sent by a computing control function, obtaining computing resource status information according to the computing resource status request; wherein the base station that receives the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered base stations and the computing task request sent by the computing service consumer; sending the computing resource status information to the computing control function; When receiving a computing task sent by a computing control function, a computing result is generated according to the computing task; wherein the base station that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

11. The computing resource and task scheduling method according to claim 10, characterized in that: The method further comprises: When initially establishing a communication connection with the computing control function, sending base station computing capability information to the computing control function; Alternatively, sending the base station computing capability information to the computing control function via the NG interface; The base station computing capability information is used to register the base station computing capability.

12. The computing resource and task scheduling method according to claim 11, characterized in that: The sending the base station computing capability information to the computing control function through the NG interface includes: According to the base station computing capacity request sent by the computing control function, the base station computing capacity information is fed back to the computing control function through the NG interface; wherein the base station computing capacity request is used to request base station computing capacity information, and the base station computing capacity request includes a base station computing capacity subscription request, and the base station capacity subscription request is based on periodic triggering or based on preset event triggering.

13. The computing resource and task scheduling method according to claim 10, characterized in that: The computing resource status request includes a computing resource status subscription request; The obtaining computing resource status information according to the computing resource status request includes: The computing resource status information corresponding to the candidate computing node within a time window T is obtained according to the prediction result obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction result, the computing resource status information corresponding to the candidate computing node within a time window T is obtained according to a preset or configured computing resource sharing strategy.

14. The computing resource and task scheduling method according to claim 13, characterized in that: The computing resource status information includes the computing resource size; The obtaining, based on the prediction result and according to a preset or configured computing resource sharing strategy, the computing resource status information corresponding to the candidate computing node within a time window T includes: Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size; The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

15. A method for scheduling computing resources and tasks, characterized in that: Applied to terminals, including: In the case of receiving a computing resource status request sent by a computing control function, obtaining computing resource status information according to the computing resource status request; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capacity of the registered terminal and the computing task request sent by the computing service consumer; sending the computing resource status information to the computing control function; When receiving a computing task sent by a computing control function, a computing result is generated according to the computing task; wherein the terminal receiving the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and computing task requirements.

16. The computing resource and task scheduling method according to claim 15, characterized in that: The method further comprises: The terminal computing capability information is sent to the computing control function through a signaling process of a non-access layer NAS signaling, where the terminal computing capability is used to register the terminal computing capability.

17. The computing resource and task scheduling method according to claim 15, characterized in that: The method further comprises: Send terminal device information to the computing control function, the terminal device information including the terminal device model and / or the International Mobile Equipment Identity code IMEI, the terminal device model and / or the International Mobile Equipment Identity code IMEI are used to query and obtain the corresponding terminal computing capability information, and the terminal computing capability information is used to register the terminal computing capability.

18. The computing resource and task scheduling method according to claim 15, characterized in that: The computing resource status request includes a computing resource status subscription request; The obtaining computing resource status information according to the computing resource status request includes: The computing resource status information corresponding to the candidate computing node within a time window T is obtained according to the prediction result obtained from the computing resource status subscription request and the historical computing resource information prediction, and / or based on the prediction result, the computing resource status information corresponding to the candidate computing node within a time window T is obtained according to a preset or configured computing resource sharing strategy.

19. The computing resource and task scheduling method according to claim 18, characterized in that: The computing resource status information includes the computing resource size; The step of calculating the computing resource status information corresponding to the candidate computing node within a time window T based on the prediction result and according to a preset or configured computing resource sharing strategy includes: Obtain multiple time samples within a time window T in the prediction result and the corresponding time remaining computing resource size; The computing resource size is determined according to a preset or configured computing resource protection ratio and the computing resource size remaining in the time, and the computing resource protection ratio is configured through a third-party network element.

20. A computing resource and task scheduling device, characterized in that: Applied to computing control functions, the device comprises: A first determination module is used to determine at least one candidate computing node that provides computing services according to the computing capabilities of the registered computing nodes and the computing task request sent by the computing service consumer, wherein the computing node includes at least one of a terminal, a base station, an edge computing node, and a core network computing node; A first sending module, used to send a computing resource status request to the candidate computing node, wherein the computing resource status request is used to request computing resource status information; A second determination module is used to determine at least one target computing node and a corresponding computing task from among the candidate computing nodes according to computing task requirements and computing resource status information fed back by the candidate computing nodes; The second sending module is used to send the corresponding computing task to the target computing node.

21. A computing resource and task scheduling device, characterized in that: Applied to a base station, the device comprises: A first computing module, configured to, upon receiving a computing resource status request sent by a computing control function, obtain computing resource status information according to the computing resource status request; wherein the base station receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capabilities of the registered base stations and the computing task request sent by the computing service consumer; A third sending module, used to send the computing resource status information to the computing control function; The second computing module is used to generate a computing result according to the computing task when receiving the computing task sent by the computing control function; wherein the base station that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

22. A computing resource and task scheduling device, characterized in that: Applied to a terminal, the device comprises: a third computing module, configured to, upon receiving a computing resource status request sent by a computing control function, obtain computing resource status information according to the computing resource status request; wherein the terminal receiving the computing resource status request is a candidate computing node, and the candidate computing node is determined by the computing control function according to the computing capability of the registered terminal and the computing task request sent by the computing service consumer; A fourth sending module, used to send the computing resource status information to the computing control function; The fourth computing module is used to generate a computing result according to the computing task when receiving the computing task sent by the computing control function; wherein, the terminal that receives the computing task is the target computing node, and the target computing node is determined by the computing control function according to the computing resource status information and the computing task requirements.

23. A computing resource and task scheduling device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the program implements the computing resource and task scheduling method as described in any one of claims 1 to 9, or implements the computing resource and task scheduling method as described in any one of claims 10 to 14, or implements the computing resource and task scheduling method as described in any one of claims 15 to 19.

24. A readable storage medium, characterized in that: include: The readable storage medium stores a program, which, when executed by the processor, implements the computing resource and task scheduling method as described in any one of claims 1 to 9, or implements the computing resource and task scheduling method as described in any one of claims 10 to 14, or implements the computing resource and task scheduling method as described in any one of claims 15 to 19.

Images