CN118802935A

CN118802935A - Edge node-based power calculation scheduling method and device and terminal equipment

Info

Publication number: CN118802935A
Application number: CN202410715857.5A
Authority: CN
Inventors: 蒋海明; 曲鑫
Original assignee: Shenzhen Onething Technology Co Ltd
Current assignee: Shenzhen Onething Technology Co Ltd
Priority date: 2024-06-04
Filing date: 2024-06-04
Publication date: 2024-10-18

Abstract

The application provides a computing power scheduling method, a computing power scheduling device and terminal equipment based on edge nodes, wherein the computing power scheduling method comprises the following steps: acquiring service requests of other applications in the user terminal; the service request is a request for computing service; determining a target edge node matched with the service request from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; after establishing point-to-point communication connection with the target edge node, sending a service request to the target edge node; after the target edge node determines the corresponding request result according to the service request, the access layer proxy module acquires the request result and sends the request result to other applications of the user terminal. The method can not only avoid carrying out virtualization operation on the edge nodes, but also fully utilize the computing resources, so that the method can save the processing resources.

Description

Edge node-based power calculation scheduling method and device and terminal equipment

Technical Field

The present application relates to the field of information processing technologies, and in particular, to a computing power scheduling method, device and terminal equipment based on edge nodes.

Background

In recent years, with the rapid development of artificial intelligence (AI, artificial Intelligence) technology, particularly, in recent years, the development of large language models (LLM, large Language Model) represented by GPT (GENERATIVE PRE-Trained Transformer, generative pre-training transducer models) has been further advanced to a new stage. However, the problem is that the demand for computing power expands very rapidly, so how to provide a set of efficient and low-cost computing power services is a current competitive problem.

The existing main flow computing force cloud platform provides services to the outside based on centralized computing force resources, a user accesses through computing force API (Application Programming Interface, application program interface) interfaces provided by the cloud platform, an AI reasoning result is returned after the interface processing is finished, an AI related resource is provided by a computing force area, the AI related resource comprises a computing force cluster (GPU (Graphics Processing Unit, a graphic processing unit), NPU (Neural-network Processing Unit, an embedded neural network processor), TPU (Tensor Processing Unit, a tensor processing unit), FPGA (Field-Programmable GATE ARRAY, field Programmable gate array) and other core systems, an AI training model, an AI data set and other core systems, and all AI reasoning related contents are completed on the AI computing force cluster and the processing result is returned to a DCN area core service module. In order to improve the calculation effect of the calculation force area, the current technical scheme is that a calculation force node (edge node) is subjected to virtualization operation, and a user terminal sends a service request to the virtualized calculation force node; the virtualized computing node provides a corresponding service based on the service request. However, the current technical solution requires a large amount of processing resources for the virtualization operation.

Therefore, how to save processing resources based on determining service results by using edge nodes is a technical problem that needs to be solved by those skilled in the art.

Disclosure of Invention

The application aims to provide an edge node-based computing power scheduling method, an edge node-based computing power scheduling device, terminal equipment and a computer-readable storage medium, which can save processing resources on the basis of determining a service result by utilizing an edge node.

In a first aspect, the present application provides an edge node-based computing power scheduling method, which is applied to an access layer proxy module, where the access layer proxy module is disposed in a user terminal, and the method includes:

acquiring service requests of other applications in the user terminal; the service request is a request for computing service;

Determining a target edge node matched with the service request from candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal;

after establishing point-to-point communication connection with a target edge node, sending the service request to the target edge node;

After the target edge node determines a corresponding request result according to the service request, the access layer proxy module acquires the request result and sends the request result to the other application of the user terminal.

In one embodiment, the determining, from the candidate edge nodes, the target edge node that matches the service request includes:

the service request is sent to a power computing scheduling server, and a service node list fed back by the power computing scheduling server based on the service request is received;

And determining a target edge node from the service node list.

In one embodiment, the process of establishing a point-to-point communication connection with the target edge node includes:

a point-to-point communication connection is established with the target edge node through a hole punching server.

In one embodiment, the establishing a point-to-point communication connection with the target edge node through the hole punching server includes:

Sending a hole punching service request to a hole punching server, and receiving target node information of the target edge node fed back by the hole punching server based on the hole punching service request;

and establishing point-to-point communication connection with the target edge node according to the target node information.

In one embodiment, each candidate edge node in the service node list determines a service priority according to a distance from a service location of the user terminal;

And/or each candidate edge node determines service priority according to the matching degree of the request type of the service request;

And/or each candidate edge node determines a service priority according to the matching degree with the calculation power requirement of the service request.

In a second aspect, the present application further provides an edge node-based computing power scheduling method, which is applied to a target edge node, and the method includes:

After establishing point-to-point communication connection with an access layer proxy module, acquiring a service request sent by the access layer proxy module; the service request is a request for computing service; the access layer proxy module acquires service requests of other applications in the user terminal, and determines a target edge node matched with the service requests from candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; the access layer proxy module performs the steps of the method as described above;

and determining a corresponding request result according to the service request, and sending the request result to the access layer proxy module so that the access layer proxy module sends the request result to other applications of the user terminal.

In one embodiment, the process of determining the candidate edge node includes:

if the current time belongs to the preset time period, determining each edge node as a candidate edge node;

And/or, if the usage rate of the edge node is lower than a preset usage rate threshold, determining the edge node as the candidate edge node.

In a third aspect, the present application further provides an edge node-based computing power scheduling device, which is applied to an access layer proxy module, where the access layer proxy module is disposed in a user terminal, and the device includes:

A service request acquisition module, configured to acquire service requests of other applications in the user terminal; the service request is a request for computing service;

a target edge node determining module, configured to determine a target edge node that matches the service request from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal;

a service request sending module, configured to send a service request to a target edge node after establishing a peer-to-peer communication connection with the target edge node;

And the request result acquisition module is used for acquiring the request result and transmitting the request result to the other application of the user terminal by the access layer proxy module after the target edge node determines the corresponding request result according to the service request.

In a fourth aspect, the application further provides a terminal device. The terminal device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, which processor, when executing the computer program, implements the steps of the method as described above.

In a fifth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of the method as described above.

The embodiment of the application provides a computing power dispatching method based on an edge node, which is applied to an access layer proxy module, wherein the access layer proxy module is arranged in a user terminal, the access layer proxy module directly performs point-to-point communication with the target edge node, processes the service request through the target edge node, determines a corresponding request result, and feeds back the request result to other applications of the user terminal; the method can not only avoid carrying out virtualization operation on the edge nodes, but also fully utilize computing resources, so that the method can save processing resources.

It can be appreciated that the computing power scheduling device, the terminal device and the computer readable storage medium based on the edge node provided by the embodiment of the application have the same beneficial effects as the computing power scheduling method based on the edge node, and are not described herein.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a block diagram of an edge node-based computing power scheduling system according to an embodiment of the present application;

FIG. 2 is a flowchart of an edge node-based computing power scheduling method according to an embodiment of the present application;

FIG. 3 is a flowchart of another computing power scheduling method based on edge nodes according to an embodiment of the present application;

FIG. 4 is a timing diagram of another method for edge node-based power scheduling according to an embodiment of the present application;

Fig. 5 is a schematic structural diagram of an edge node-based computing power scheduling device according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of another computing power dispatching device based on edge nodes according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a terminal device according to an embodiment of the present application.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It should also be understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

As used in the present description and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".

Furthermore, the terms "first," "second," "third," and the like in the description of the present specification and in the appended claims, are used for distinguishing between descriptions and not necessarily for indicating or implying a relative importance.

Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise. "plurality" means "two or more".

The computing power scheduling method based on the edge node is applied to an access layer proxy module, and the access layer proxy module is arranged in a user terminal. Fig. 1 is a schematic diagram of an edge node-based computing power dispatching system according to an embodiment of the present application. As shown in fig. 1, an edge node-based computing power scheduling system 100 includes a user terminal 110 and candidate edge nodes 120; the user terminal 110 is provided with an access layer proxy module 111; the candidate edge node 120 is provided with an agent module 121 of the computing power node, and the candidate edge node 120 provides computing power resources for the user terminal 110 by using the agent module 121 of the computing power node; the computing node agent module 121 may be specifically a GPU (Graphics Processing Unit, graphics processor), NPU (embedded neural network processor), TPU (Tensor Processing Unit ), FPGA (Field-Programmable GATE ARRAY, field Programmable gate array), or the like.

Fig. 2 is a flowchart of an edge node-based computing power scheduling method according to an embodiment of the present application, and for convenience of explanation, only a portion related to the embodiment is shown, where the method provided in the embodiment includes the following steps:

s210: acquiring service requests of other applications in the user terminal; the service request is a request for a computing service.

The service request is a request for performing computing service, that is, a request for providing services such as AI (ARTIFICIAL INTELLIGENCE ) computing, image processing, data processing and the like by a target edge node, and the type of the service request is determined according to other application types in the user terminal.

In this embodiment, the user terminal includes an access layer proxy module and other applications, and when the other applications have a computing power requirement, a corresponding service request is generated; then the service request is sent to an access layer proxy module; correspondingly, the access layer proxy module acquires service requests sent by other applications in the user terminal.

In practical applications, other applications in the user terminal may send the service request to the access layer proxy module through an http protocol (HyperText Transfer Protocol ) or other custom protocols, which is not limited in this embodiment, and only needs other applications in the user terminal to match with the protocol of the access layer proxy module.

S220: determining a target edge node matched with the service request from the candidate edge nodes; the target edge node is configured to provide computing power resources for the user terminal.

The candidate edge nodes are edge nodes which can be used for participating in computational power scheduling in the computational power scheduling system based on the edge nodes; in addition, the edge node-based computing power dispatching system generally has a plurality of candidate edge nodes, and the number of the candidate edge nodes in the edge node-based computing power dispatching system is not limited in this embodiment.

The target edge node is a candidate edge node determined from the candidate edge nodes and used for responding to the service request. It can be understood that the computing power resources required for different service requests are different, and the computing power resources that can be provided by different candidate edge nodes are also different, so that it is necessary to determine a matched target edge node from the candidate edge nodes according to the service requests.

In practical applications, only one candidate edge node may be needed to respond to the service request, or a plurality of candidate edge nodes may be needed to respond to the service request in a coordinated manner, so that the number of target edge nodes may be one or more, a corresponding number of target edge nodes need to be determined from the candidate edge nodes according to actual requirements, and computing power resources are provided for the user terminal by using the determined one or more target edge nodes.

S230: after establishing a point-to-point communication connection with the target edge node, a service request is sent to the target edge node.

Wherein, peer-to-Peer (p 2p for short) communication refers to a communication mode of sharing computer resources and services through direct exchange; in this embodiment, point-to-point refers to that a communication connection is established between a user terminal and a target edge node to directly exchange information, so that an intermediate server is not required to be relied on to forward data.

Specifically, after the access layer proxy module establishes a point-to-point communication connection with the target edge node, it means that the user terminal establishes a point-to-point communication connection with the target edge node; the access layer proxy module sends a service request to the target edge node based on the established point-to-point communication connection.

S240: after the target edge node determines the corresponding request result according to the service request, the access layer proxy module acquires the request result and sends the request result to other applications of the user terminal.

Specifically, after the access layer proxy module sends the service request to the target edge node, the target edge node may provide computing resources for the user terminal. Specifically, the target edge node responds to the service request by using the own computing power node proxy module to obtain a corresponding request result, and then sends the request result to the access layer proxy module based on point-to-point communication; correspondingly, the access layer proxy module acquires the request result and sends the request result to other applications of the user terminal.

It should be noted that, the request result may be data content such as a picture, a video, a voice, etc. corresponding to the service request, and the specific content of the request result is not limited in this embodiment.

On the basis of the above embodiment, the present embodiment further describes and optimizes a technical solution, and specifically, in this embodiment, determining, from candidate edge nodes, a target edge node that matches with a service request includes:

sending a service request to a power dispatching server, and receiving a service node list fed back by the power dispatching server based on the service request;

and determining the target edge node from the service node list.

The power computing scheduling server refers to a server for performing power computing scheduling, and the power computing scheduling server can be implemented by an independent server or a server cluster formed by a plurality of servers.

Specifically, the candidate edge nodes are provided with an algorithm force node proxy module, the algorithm force node proxy module deploys a corresponding model, after deployment is completed, the algorithm force scheduling server registers the model, and then after other applications of the subsequent user terminals generate algorithm force requests, the algorithm force scheduling server can determine a corresponding service node list according to the service requests. The candidate edge nodes can update the computing power node proxy modules deployed by the candidate edge nodes or update the registration content to the computing power scheduling server according to a preset period. In this embodiment, after receiving the service request, the access layer proxy module sends the service request to the power calculation scheduling server; the computing power dispatching server screens each candidate edge node based on the service request, determines the candidate edge node which can be used for responding to the service request, and determines a corresponding service node list according to the screened candidate edge node. That is, the service node list includes candidate edge nodes capable of providing computational resources for the service request of the user terminal.

It is understood that the service node list includes at least one candidate edge node; after the service node list is determined, the target edge node can be determined from the service node list according to a preset rule, or the user performs a clicking operation based on the user terminal, and the target edge node is determined from the candidate edge nodes in the service node list.

It should be noted that, if the number of target edge nodes to be determined is 1 and only one candidate edge node exists in the service node list, determining the candidate edge node as the target edge node; if the number of the target edge nodes to be determined is 1 and a plurality of candidate edge nodes exist in the service node list, determining one target edge node from the plurality of candidate edge nodes in the service node list.

In addition, if a plurality of target edge nodes need to be determined and the number of candidate edge nodes in the service node list is smaller than the number of the required target edge nodes, abnormal prompt information is sent to the user terminal; if a plurality of target edge nodes need to be determined and the number of candidate edge nodes in the service node list is greater than or equal to the number of target edge nodes required, determining the required target edge nodes from the plurality of candidate edge nodes in the service node list.

Therefore, according to the method of the embodiment, the target edge node can be determined efficiently and conveniently, so that the efficiency of computing power scheduling based on the edge node is improved.

On the basis of the above embodiment, the technical solution is further described and optimized in this embodiment, and specifically, in this embodiment, a process of establishing a peer-to-peer communication connection with a target edge node includes:

A point-to-point communication connection is established with the target edge node through the hole server.

It should be noted that, the hole punching server in this embodiment refers to a server for implementing the direct point-to-point communication connection between the access layer proxy module and the target edge node, and the specific type of the hole punching server in this embodiment is not limited, and may be implemented by an independent server or a server cluster formed by multiple servers.

Specifically, the hole server may hole through UDP (User Datagram Protocol ) or TCP (Transmission Control Protocol, transmission control protocol) so that the access layer proxy module establishes a point-to-point communication connection with the target edge node. In a specific embodiment, the hole punching server may be embodied as a STUN server.

It should be noted that, the point-to-point communication connection established by the access layer proxy module and the target edge node is generally a long connection,

According to the method of the embodiment, the access layer proxy module can efficiently and conveniently establish point-to-point communication connection with the target edge node, so that the efficiency of computing power scheduling based on the edge node is improved.

Based on the above embodiment, the technical solution is further described and optimized in this embodiment, and specifically in this embodiment, the establishing, by the hole punching server, a peer-to-peer communication connection with the target edge node includes:

sending a hole punching service request to a hole punching server, and receiving target node information of a target edge node fed back by the hole punching server based on the hole punching service request;

Wherein, the hole punching service request refers to a request for requesting the hole punching server to perform hole punching so that the access layer proxy module establishes a point-to-point communication connection with the target edge node. It should be noted that the hole punching service request at least includes identification information of the hole punching object, that is, identification information of the target edge node.

In this embodiment, the hole punching server establishes communication connections with each candidate edge node in advance, and the communication connections may be long connections; after the access layer proxy module sends a hole punching service request to the hole punching server, the hole punching server acquires target node information of a target edge node according to the identification information of the target edge node based on the pre-established communication connection; wherein the destination node information includes, but is not limited to, a port number and an IP address (Internet Protocol Address ) of the destination edge node; then, the hole punching server sends the target node information of the target edge node to the access layer proxy module, namely the access layer proxy module acquires the target node information of the target edge node; the access layer proxy module may establish a point-to-point communication connection with the target edge node based on the target node information.

Therefore, according to the method of the embodiment, the access layer proxy module can efficiently and conveniently establish the point-to-point communication connection with the target edge node, so that the efficiency of computing power scheduling based on the edge node is improved.

On the basis of the above embodiment, the present embodiment further describes and optimizes the technical solution, and specifically, in this embodiment, each candidate edge node in the service node list determines the service priority according to the distance from the service location of the user terminal;

and/or each candidate edge node determines the service priority according to the matching degree with the request type of the service request;

And/or each candidate edge node determines the service priority according to the matching degree with the calculation power requirement of the service request.

The service priority represents the matching degree of the candidate edge node and the service request; in general, the higher the service priority, the higher the correspondence indicates that the candidate edge node matches the service request, and the greater the probability of determining the candidate edge node as the target edge node.

In this embodiment, the power scheduling server may determine the service priority of the candidate edge node according to the following three manners, and feed back the corresponding service node list to the access layer proxy module.

Mode 1: and determining the service priority of the candidate edge node according to the distance between the candidate edge node and the service position of the user terminal.

It can be understood that the further the candidate edge node is from the user terminal, the longer the time required for data transmission between the candidate edge node and the user terminal (access layer proxy module) is, and the greater the probability of failure abnormality in data transmission, so that the service priority is inversely related to the distance, that is, the user is served near each candidate edge node, and the request result corresponding to the service request can be more conveniently and timely obtained.

In practical application, acquiring service positions of a user terminal (an access layer proxy module), acquiring position information corresponding to each candidate edge node, determining distances between each candidate edge node and the user terminal (the access layer proxy module) according to the service positions and the position information, and determining service priorities corresponding to each candidate edge node according to ascending order of the distances; if the distance between the candidate edge node and the user terminal (access layer proxy module) is smaller, the service priority corresponding to the candidate edge node is higher; the greater the distance between the candidate edge node and the user terminal (access layer proxy module), the lower the service priority corresponding to the candidate edge node.

Mode 2: and determining the service priority according to the matching degree of the candidate edge node and the request type of the service request.

It should be noted that, the candidate edge nodes are provided with the agent modules of the computing power nodes, and the types of services that can be provided by the agent modules of different computing power nodes are different, and the types of services include AI calculation, image processing, data processing and the like. The request type of the service request is the type of service that the target edge node is required to provide. For the request type of the service request, the service type of the service request is responded by using the computing power node proxy module comprising the corresponding service type, so that the higher the matching degree of the service type of the computing power node proxy module in the candidate edge node and the request type of the service request is, the higher the resource utilization rate of the candidate edge node is, and the higher the service priority corresponding to the candidate edge node is, namely, the service priority is positively correlated with the matching degree of the request type.

Mode 3: and determining the service priority according to the matching degree of the candidate edge node and the calculation force demand of the service request.

It should be noted that, the resource configurations corresponding to the candidate edge nodes are different, the deployable models are different, that is, the corresponding configured computing power node proxy modules are different. For example, assuming that the user needs to conduct an AI dialogue, the candidate edge node's inode proxy module should be configured with a large language model (LLM-7B,Large Language Model-7B), and assuming that a more intelligent AI dialogue is needed, the inode proxy module should be configured with a large language model with a higher level of inode (LLM-13B,Large Language Model-13B). Or assuming that the user needs to draw a graph according to the text description, the computing node proxy module of the candidate edge node should configure a multi-mode AI model, such as an SD model (Stable diffion model).

It should be further noted that multiple different models may be configured in the same candidate edge node.

Generally, the higher the resource configuration corresponding to the candidate edge node, the higher the computational power level of the computational power node proxy module which can be deployed, and the higher the computational power requirement which can be satisfied. For example, A100 may deploy a large language model (LLM-13B), 4090 may only deploy a large language model (LLM-7B); therefore, candidate edge nodes of the power calculation node proxy modules with different power calculation levels are deployed, so that the power calculation demands of the different power calculation levels can be met; correspondingly, for the computational effort requirement of the service request, a candidate edge node corresponding to the computational effort level needs to be selected. In this embodiment, the service priority and the matching degree of the computing power demand are positively correlated, that is, the higher the matching degree of the computing power level of the computing power node proxy module in the candidate edge node and the computing power demand of the service request is, the higher the resource utilization rate of the candidate edge node is, and the higher the service priority corresponding to the candidate edge node is.

In a specific embodiment, the service priority of each candidate edge node may be determined according to the distance between the candidate edge node and the service location of the user terminal, the matching degree between the request type of the service request and the service type of the candidate edge node, and the matching degree between the computing power requirement of the service request and the computing power level of the candidate edge node.

According to the method of the embodiment, the candidate edge nodes matched with the service request can be determined, so that the efficiency of computing power scheduling based on the edge nodes is improved.

FIG. 3 is a flowchart of another computing power scheduling method based on edge nodes according to an embodiment of the present application; the computing power scheduling method based on the edge node, provided by the embodiment of the application, is applied to the target edge node, and comprises the following steps:

S310: after establishing point-to-point communication connection with the access layer proxy module, acquiring a service request sent by the access layer proxy module; the service request is a request for computing service; the access layer proxy module acquires service requests of other applications in the user terminal, and determines a target edge node matched with the service requests from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; the access layer proxy module executes the steps in the method embodiments;

S320: and determining a corresponding request result according to the service request, and sending the request result to the access layer proxy module so that the access layer proxy module sends the request result to other applications of the user terminal.

For the specific implementation process of each step in this embodiment, reference may be made to the descriptions in the foregoing embodiments, which are not repeated here.

On the basis of the above embodiment, the present embodiment further describes and optimizes a technical solution, and specifically, in this embodiment, a process of determining candidate edge nodes includes:

And/or determining the edge node as a candidate edge node if the utilization rate of the edge node is lower than a preset utilization rate threshold value.

Specifically, a preset time period is generally determined according to the idle time period of each edge node; in practical application, the preset time period can be determined according to the use habit of the user or big data analysis; the preset time period may be specifically determined according to a time period in which the edge node is in an idle state or a time period in which the usage rate of the edge node is lower than a preset usage rate threshold.

When the candidate edge nodes need to be determined, acquiring the current time and judging whether the current time belongs to a preset time period or not; if the current time belongs to the preset time period, determining each edge node as a candidate edge node; if the current time does not belong to the preset time period, the candidate edge nodes do not need to be determined.

Specifically, the usage of the edge node indicates the ratio of the processing resources used by the edge node to the total processing resources. In this embodiment, the usage rate of the edge node is determined, and a preset usage rate threshold of the edge node is obtained, if the usage rate of the edge node is lower than the preset usage rate threshold, it indicates that the edge node has a spare processing resource available for responding to a service request of the user terminal, that is, the edge node is available for participating in computational scheduling, so that the edge node is determined as a candidate edge node.

According to the method, candidate edge nodes are screened out from the edge nodes through node redundancy and a disaster recovery scheduling design mode, the candidate edge nodes are used for participating in computational scheduling, the availability of the candidate edge nodes participating in computational scheduling can be relatively improved, and the computational scheduling efficiency based on the edge nodes is improved.

In order to enable those skilled in the art to better understand the technical scheme of the present application, the technical scheme in the embodiment of the present application is described in detail below in conjunction with practical application scenarios. Fig. 4 is a timing chart of another edge node-based power scheduling method according to an embodiment of the present application, as shown in fig. 4, in an embodiment of the present application, a specific step of the edge node-based power scheduling method is as follows:

Other applications (Application program APP) in the user terminal generate a service request and send the service request to an access layer proxy module (local access layer proxy);

after receiving the service request, the access layer proxy module (local access layer proxy) sends the service request to the power calculation scheduling server;

The power dispatching server determines a service node list based on the service request and sends the service node list to an access layer proxy module (local access layer proxy);

the access layer proxy module (local access layer proxy) determines a target edge node from the service node list;

An access layer proxy module (local access layer proxy) sends a hole punching service request to a hole punching server;

the hole punching server acquires target node information of a target edge node based on the hole punching service request and sends the target node information to an access layer proxy module (local access layer proxy);

The access layer proxy module (local access layer proxy) establishes point-to-point communication connection with the target edge node according to the target node information; i.e. the user terminal establishes a point-to-point communication connection (p 2p communication connection) with the target edge node;

the access layer proxy module (local access layer proxy) encapsulates (protocol conversion) the service request according to a preset communication protocol to obtain an encapsulated service request, and sends the encapsulated service request to a target edge node through point-to-point communication connection;

the target edge node unpacks the packed service request (content reduction), and sends the unpacked service request to the own computing power node proxy module;

The target edge node responds to the service request by utilizing the self power computing node proxy module, a corresponding request result is determined, and the self power computing node proxy module sends the request result to the target edge node;

The target edge node sends a request result to an access layer proxy module (local access layer proxy) based on a pre-established point-to-point communication connection;

After the request result is encapsulated (protocol conversion) by the access layer proxy module (local access layer proxy), the encapsulated request result is sent to other applications (application program APP) of the user terminal;

Other applications (application program APP) unpack the packed request result (content restoration) to obtain a request result corresponding to the service request; the whole power calculation scheduling process is completed.

In addition, the embodiment realizes computational power scheduling and result distribution based on p2p communication connection through the access layer proxy module (local access layer proxy), and does not need additional adjustment of an application program APP side; candidate edge nodes are screened from the edge nodes through node redundancy and disaster recovery scheduling design modes, and the candidate edge nodes are used for participating in computational power scheduling, so that the availability of each candidate edge node participating in computational power scheduling can be relatively improved, and the computational power scheduling efficiency based on the edge nodes is improved; in addition, the computing power of the application program APP of the user terminal is transferred to the edge node with low cost through p2p communication, namely the edge node and the user terminal are utilized to share the computing power, so that the computing power cost can be reduced; in addition, the corresponding request result (such as pictures, videos, voices and the like) can be directly distributed to the user terminal without processing through a centralized IDC (data center), and the generation of extra bandwidth cost can be further avoided.

It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present application.

Fig. 5 is a schematic structural diagram of an edge node-based computing power scheduling device according to an embodiment of the present application. As shown in fig. 5, the edge node-based computing power scheduling apparatus of this embodiment is applied to an access layer proxy module, which is disposed in a user terminal and includes a service request acquisition module 510, a target edge node determination module 520, a service request transmission module 530, and a request result acquisition module 540; wherein,

A service request obtaining module 510, configured to obtain service requests of other applications in the user terminal; the service request is a request for computing service;

A target edge node determining module 520, configured to determine a target edge node matching the service request from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal;

A service request sending module 530, configured to send the service request to a target edge node after establishing a point-to-point communication connection with the target edge node;

and the request result obtaining module 540 is configured to, after the target edge node determines a corresponding request result according to the service request, obtain the request result by using the access layer proxy module and send the request result to the other applications of the user terminal.

The edge node-based computational power scheduling device provided by the embodiment of the application has the same beneficial effects as the edge node-based computational power scheduling method.

In one embodiment, the target edge node determining module includes:

the service node list acquisition sub-module is used for sending the service request to the power calculation scheduling server and receiving a service node list fed back by the power calculation scheduling server based on the service request;

and the target edge node determining submodule is used for determining the target edge node from the service node list.

In one embodiment, the process of establishing a point-to-point communication connection between the service request sending module and the target edge node includes:

and the communication connection establishment sub-module is used for establishing point-to-point communication connection with the target edge node through the hole punching server.

In one embodiment, the communication connection establishment submodule includes:

a target node information acquisition unit, configured to send a hole drilling service request to a hole drilling server, and receive target node information of the target edge node fed back by the hole drilling server based on the hole drilling service request;

And the communication connection establishment unit is used for establishing point-to-point communication connection with the target edge node according to the target node information.

Fig. 6 is a schematic structural diagram of another computing power dispatching device based on edge nodes according to an embodiment of the present application. As shown in fig. 6, the edge node-based computational power scheduling apparatus of this embodiment is applied to a target edge node, and includes a service request determining module 610 and a request result feedback module 620, wherein,

The service request determining module 610 is configured to obtain a service request sent by the access layer proxy module after establishing a point-to-point communication connection with the access layer proxy module; the service request is a request for computing service; the access layer proxy module acquires service requests of other applications in the user terminal, and determines a target edge node matched with the service requests from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; the access layer proxy module executes the steps in the method embodiments;

the request result feedback module 620 is configured to determine a corresponding request result according to the service request, and send the request result to the access layer proxy module, so that the access layer proxy module sends the request result to other applications of the user terminal.

In one embodiment, the service request acquisition module includes:

The first candidate edge node determining submodule is used for determining each edge node as a candidate edge node if the current time belongs to a preset time period;

And/or a second candidate edge node determining submodule, configured to determine the edge node as the candidate edge node if the usage rate of the edge node is lower than a preset usage rate threshold.

It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present application, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.

Fig. 7 is a schematic structural diagram of a terminal device according to an embodiment of the present application. As shown in fig. 7, the terminal device 700 of this embodiment includes a memory 701, a processor 702, and a computer program 703 stored in the memory 701 and executable on the processor 702; the processor 702, when executing the computer program 703, implements the steps of the various edge node-based power scheduling method embodiments described above; or the processor 702 when executing the computer program 703 performs the functions of the modules/units in the above-described apparatus embodiments.

By way of example, the computer program 703 may be divided into one or more modules/units stored in the memory 701 and executed by the processor 702 to implement the methods of embodiments of the present application. One or more of the modules/units may be a series of computer program instruction segments capable of performing a specific function for describing the execution of the computer program 703 in the terminal device 700. For example, the computer program 703 may be divided into a service request acquisition module, a target edge node determination module, a service request transmission module, and a request result acquisition module, each of which specifically functions as follows:

Or the computer program 703 may be divided into a service request acquisition module and a request result feedback module, where each module specifically functions as follows:

The service request acquisition module is used for acquiring a service request sent by the access layer proxy module after establishing point-to-point communication connection with the access layer proxy module; the service request is a request for computing service; the access layer proxy module acquires service requests of other applications in the user terminal, and determines a target edge node matched with the service requests from the candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; the access layer proxy module executes the steps in the method embodiments;

and the request result feedback module is used for determining a corresponding request result according to the service request and sending the request result to the access layer proxy module so that the access layer proxy module sends the request result to other applications of the user terminal.

In application, the terminal device 700 may be a computing device such as a desktop computer, a notebook computer, a palm computer, and a cloud server. The terminal device 700 may include, but is not limited to, a memory 701 and a processor 702. It will be appreciated by those skilled in the art that fig. 7 is merely an example of a terminal device and is not meant to be limiting, and may include more or fewer components than shown, or may combine certain components, or different components, e.g., a terminal device may also include an input-output device, a network access device, a bus, etc.; the input and output equipment can comprise a camera, an audio acquisition/play device, a display screen and the like; the network access device may include a communication module for wireless communication with an external device.

In applications, the Processor may be a central processing unit (Central Processing Unit, CPU), or other general purpose Processor, digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), off-the-shelf Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

In an application, the memory may be an internal storage unit of the terminal device, such as a hard disk or a memory of the terminal device; external storage devices of the terminal device, such as a plug-in hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), etc., which are provided on the terminal device; it may also comprise both an internal memory unit of the terminal device and an external memory device. The memory is used to store an operating system, application programs, boot Loader (Boot Loader), data, and other programs, etc., such as program code for a computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.

The embodiments of the present application also provide a computer readable storage medium storing a computer program, which when executed by a processor, implements the steps of the above-described method embodiments.

The computer readable storage medium provided by the embodiment of the application has the same beneficial effects as the computing power scheduling method based on the edge node.

The present application may be implemented in whole or in part by a computer program which, when executed by a processor, performs the steps of the method embodiments described above, and which may be embodied in a computer readable storage medium. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, executable files or in some intermediate form, etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a terminal device, a recording medium, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a U-disk, removable hard disk, magnetic or optical disk, etc.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the various illustrative apparatus and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the embodiments of the apparatus described above are illustrative only, and the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, the apparatus may be indirectly coupled or in communication connection, whether in electrical, mechanical or other form.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims

1. The computing power scheduling method based on the edge node is characterized by being applied to an access layer proxy module, wherein the access layer proxy module is arranged in a user terminal, and the method comprises the following steps:

2. The method of claim 1, wherein the determining a target edge node from the candidate edge nodes that matches the service request comprises:

And determining a target edge node from the service node list.

3. The method of claim 1, wherein the process of establishing a point-to-point communication connection with the target edge node comprises:

4. A method according to claim 3, wherein said establishing a point-to-point communication connection with said target edge node via a hole server comprises:

5. The method according to claim 2, wherein each of said candidate edge nodes in said list of service nodes determines a service priority in accordance with a distance from a service location of said user terminal;

6. An edge node-based computational power scheduling method, which is applied to a target edge node, the method comprising:

After establishing point-to-point communication connection with an access layer proxy module, acquiring a service request sent by the access layer proxy module; the service request is a request for computing service; the access layer proxy module acquires service requests of other applications in the user terminal, and determines a target edge node matched with the service requests from candidate edge nodes; the target edge node is used for providing computing power resources for the user terminal; the access layer proxy module performing the method of any one of claims 1 to 5;

7. The method of claim 6, wherein determining the candidate edge node comprises:

8. An edge node-based computational power scheduling device, which is applied to an access layer proxy module, wherein the access layer proxy module is arranged in a user terminal, and the device comprises:

9. Terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 5 or 6 to 7 when the computer program is executed.

10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1 to 5 or 6 to 7.