WO2024078203A1

WO2024078203A1 - Computing power network end-to-end coordination method, computing power network, electronic device, and storage medium

Info

Publication number: WO2024078203A1
Application number: PCT/CN2023/117066
Authority: WO
Inventors: 骆旭剑
Original assignee: 中兴通讯股份有限公司
Priority date: 2022-10-09
Filing date: 2023-09-05
Publication date: 2024-04-18
Also published as: CN117896253A

Abstract

The present application discloses a computing power network end-to-end coordination method, a computing power network, an electronic device, and a storage medium. The computing power network end-to-end coordination method comprises: receiving a computing power demand sent by a serving end, and then, according to the computing power demand and a preset allocation policy, determining a target computing power resource in a preset computing power pool; sending a computing power occupation request to a target terminal corresponding to the target computing power resource, receiving an occupation confirmation response fed back by the target terminal on the basis of the computing power occupation request, and then constructing a virtual computing power environment according to the target computing power resource, so as to run in the virtual computing power environment service data corresponding to the computing power demand.

Description

Computing network end-to-end collaboration method, computing network, electronic device and storage medium

Related Applications

This application claims priority to Chinese patent application No. 202211231379.8 filed on October 9, 2022, the entire contents of which are incorporated by reference into this application.

Technical Field

The present application relates to the field of network communication technology, and in particular to a computing power network end-to-end collaboration method, a computing power network, an electronic device and a storage medium.

Background technique

The computing network is a new type of information infrastructure that allocates and flexibly schedules computing resources, storage resources, and network resources between the cloud, edge, and end based on business needs. The essence of the computing network is a computing resource service. In the future, corporate customers or individual users will not only need networks and clouds, but also need to flexibly schedule computing tasks to the right place.

As the global digital economy continues to grow steadily, the demand for computing power networks will continue to increase, so the computing power network infrastructure also needs to be built. On the one hand, the construction and operation and maintenance of the central cloud and edge computing nodes in the computing power network infrastructure require high cost investment. On the other hand, the types and numbers of terminals are numerous and mobile, which makes the management and use of terminal computing power relatively complex, resulting in increased costs. Therefore, how to increase the service capabilities of the computing power network while saving costs has become a technical problem that needs to be solved urgently.

Summary of the invention

The main purpose of this application is to provide a computing power network end-to-end collaboration method, a computing power network, an electronic device and a storage medium, aiming to solve the technical problem of how to increase the service capacity of the computing power network while saving costs.

To achieve the above objectives, the present application provides a computing power network end-to-end collaboration method, including:

After receiving the computing power demand sent by the server, determine the target computing power resources in the preset computing power pool according to the computing power demand and the preset allocation strategy;

A computing power occupation request is sent to the target terminal corresponding to the target computing power resource, and after receiving an occupation confirmation response from the target terminal based on the computing power occupation request, a virtual computing power environment is constructed according to the target computing power resource, so as to run the business data corresponding to the computing power demand in the virtual computing power environment.

In addition, to achieve the above purpose, the present application also provides a computing network, which runs the above computing network end-to-end collaboration method, including a server, a computing network brain and multiple terminals;

The server is used to send the computing power requirement corresponding to the business request to the computing network brain after receiving the business request;

The computing network brain is used to determine the target computing power resources according to the computing power requirements and the preset allocation strategy after receiving the computing power requirements sent by the server, determine the target terminal of the target computing power resources among the multiple terminals, and send a computing power occupation request to the target terminal;

The target terminal is used to generate an occupation confirmation response after receiving the computing power occupation request and determining the computing power occupation according to the computing power occupation request, and send the occupation confirmation response to the computing network brain;

The computing network brain is used to build a virtual computing environment based on the target computing resources so as to run the business data corresponding to the computing demand in the virtual computing environment.

In addition, to achieve the above-mentioned purpose, the present application also provides an electronic device, including: a processor; and a memory arranged to store computer-executable instructions, wherein when the executable instructions are executed, the processor executes the steps of the computing power network end-to-end collaboration method as described above.

In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, wherein the storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device Execute the steps of the computing power network end-to-end collaboration method as described above.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a terminal\device structure of a hardware operating environment involved in an embodiment of the present application;

FIG2 is a flow chart of a first embodiment of a computing power network end-to-end collaboration method of the present application;

FIG3 is a schematic diagram of the overall architecture of the computing power network of the present application;

FIG4 is a schematic diagram of the process of terminal computing power processing business needs in the computing power network end-to-end collaboration method of the present application;

FIG5 is a schematic diagram of a processing flow of network collaboration by terminal computing power in the computing power network end-to-end collaboration method of the present application;

FIG6 is a schematic diagram of the process of terminal computing power performing calculation + network collaboration in the computing power network end-to-end collaboration method of the present application;

FIG7 is a schematic diagram of a process flow of computing coordination of terminal computing power in the computing power network end-to-end coordination method of the present application;

FIG8 is a schematic diagram of the process of terminal registration of computing power in the computing power network end-to-end collaboration method of the present application.

The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

Detailed ways

It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

As shown in FIG. 1 , FIG. 1 is a schematic diagram of the terminal structure of the hardware operating environment involved in the embodiment of the present application.

The terminal in the embodiment of the present application is a computing power network end-to-end collaborative device.

As shown in FIG1 , the terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

In one embodiment, the terminal may also include a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among them, the sensors include light sensors, motion sensors, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and/or backlight when the terminal device is moved to the ear. Of course, the terminal device may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

Those skilled in the art will appreciate that the terminal structure shown in FIG. 1 does not limit the terminal and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

As shown in FIG. 1 , the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a computing power network end-to-end collaboration program.

In the terminal shown in FIG1 , the network interface 1004 is mainly used to connect to the background server and communicate data with the background server; the user interface 1003 is mainly used to connect to the client (user end) and communicate data with the client; and the processor 1001 can be used to call the computing power network end-to-end collaborative program stored in the memory 1005 and perform the following operations:

The present application proposes a computing power network end-to-end collaboration. In the first embodiment of the computing power network end-to-end collaboration of the present application, referring to FIG. 2 , the computing power network end-to-end collaboration method includes:

Step S10, after receiving the computing power demand sent by the server, determine the target computing power resources in the preset computing power pool according to the computing power demand and the preset allocation strategy;

Step S20, sending a computing power occupation request to the target terminal corresponding to the target computing power resources, and after receiving an occupation confirmation response from the target terminal based on the computing power occupation request, building a virtual computing power environment according to the target computing power resources, so as to run the business data corresponding to the computing power demand in the virtual computing power environment.

The architecture diagram of the computing network infrastructure in this embodiment can be shown in Figure 1. The computing network includes three computing nodes: the central cloud, the edge computing node, and the terminal. The computing network connects the geographically distributed computing center nodes through new network technology, dynamically and in real time perceives the computing resource status, and then coordinates and schedules computing tasks, transmits data, and forms a network that perceives, allocates, and schedules computing power in a global scope. On this basis, it gathers and shares computing power and data. Data and application resources. Among them, the central cloud and edge computing nodes (such as 5G and MEC (Mobile Edge Computing)) are generally built and managed by operators, while terminals are generally user assets (individuals or units). Among them, the terminal in this embodiment can be a device with a microprocessor that can perform data calculations, storage, and network connections, such as mobile phones, computers, and in-vehicle intelligent systems. Computing power can include computing, storage, and network capabilities.

In this embodiment, when the terminal participates in the computing power network, the computing network brain schedules the computing power resources of the terminal according to the preferred strategy to achieve the effect of fully utilizing the computing power resources of the terminal. The present application also provides a method of end-to-end computing power collaboration, in which multiple terminals complete business needs through collaboration to ensure the SLA level of computing power services. And in this embodiment, in order to achieve cost savings while increasing the service capabilities of the computing power network. Therefore, it is to let users and operators sign contracts to participate in the business of the computing power network, agree on the way the terminal provides computing power, the benefits enjoyed, etc. When the terminal accesses the computing power network, it reports its own computing power information through computing power registration, including but not limited to: the terminal's unique computing power identification, computing/storage/network capabilities (such as processors, storage, network types and quantities), and location information (such as longitude and latitude, altitude). The computing network brain in the computing power network generates resource topology information based on the terminal registration information, and uniformly manages resources for subsequent scheduling and scheduling. Among them, the resource topology information includes information and connection information of each computing power node (cloud, edge, end), and can also include connectable information of adjacent computing power terminals.

When a business (computing power demand) occurs, the computing network brain selects terminal computing power resources suitable for participating in the business operation according to business information, computing power resource information and configuration strategy, and performs computing, storage and other tasks at the terminal. Among them, the task can be general computing, such as voice and image analysis and processing, and the processing results are sent to users or other servers; it can also be local computing, such as the terminal is also a business consumer, and the original computing power network is to process the calculation, and then process or partially process it at the terminal (such as smart cameras for image preprocessing), reducing network traffic; it can also be data storage, caching data in the terminal memory or storage device (such as video downloads) to provide it to other nearby terminals for use, reducing the data access pressure of the cloud server. It can also provide sensor data to provide various useful real-time information such as the location, speed, temperature, etc. of the terminal for specific services.

When there are multiple terminal services, the computing network brain allocates appropriate terminal computing power according to business information, computing power resource information and configuration strategies, and sends the coordination information of computing power terminals (including terminal identification, protocol interface, data type, etc.) in combination with business characteristics and terminal distribution, and dispatches multiple terminals to work together to complete tasks, providing SLA (Service Level Agreement) guarantees (latency, bandwidth, etc.). Among them, the end-to-end collaborative networking forms include: point-to-point: For example, in a computing power network, adjacent terminals can directly communicate point-to-point for communication services; bridging: The terminal can also provide data transfer services for other terminals in the network, so that most of the information does not need to pass through the access network and server when communicating between terminals; star type: The terminal can also undertake the distribution/aggregation services of other computing power terminals in the adjacent network (cluster brain), reducing the access and routing pressure of the computing power network. Among them, the content of end-to-end collaboration can include direct message transmission to reduce transit: such as point-to-point communication between adjacent terminals to reduce the number of messages passing through other network nodes; sharing of information to reduce measurement and calculation: such as coordination of surrounding vehicle movements; public computing to reduce duplication: for example, when multiple adjacent terminals watch live broadcasts together, one terminal can forward the received and decoded audio and video streams to other terminals.

That is, in this embodiment, the user signs a contract with the terminal to participate in the business of the computing power network, and the terminal has the software and hardware capabilities to participate in the computing power network (for example, to load or run a certificate or program for a secure connection); under certain conditions, the terminal registers its own computing power information with the computing power network (and subsequently updates or de-registers the computing power information as appropriate); the computing power brain receives the registration information of the terminal, updates the resource topology, and incorporates the terminal computing power into the computing power pool; when there is a demand for computing power, the computing power brain allocates the terminal's computing power resources according to business needs and resource conditions and according to the computing power scheduling strategy (such as giving priority to adjacent computing power); if the end-to-end collaboration strategy is met, the collaboration information is issued at the same time; the computing power terminal performs business processing; if there is a need for collaboration, end-to-end collaboration (such as point-to-point communication) is performed based on the collaboration information; the computing power brain records the information on the terminal's participation in computing power tasks, and the computing power trading system calculates the user's deserved rights and interests based on this information.

In addition, the computing network brain in this embodiment records the computing power information provided by the terminal, including but not limited to: unique identification of computing power resources, computing power type (such as CPU), actual consumption computing power information (such as TOPS), service time, business type, SLA information, and sends this information to the computing power trading system, which calculates the rights and interests that users deserve according to the agreed rules. In this way, the terminal can participate in the computing power network, thereby increasing the service capacity of the computing power network, maximizing resource utilization, and optimizing the service quality through end-to-end collaboration, achieving a win-win effect for users and operators.

The network architecture of the computing network is shown in Figure 3, which may include computing network operation, computing network brain and computing network base. The computing network operation may be the operation service layer, such as the server, including the computing power transaction module and the business service module. The computing network brain may be the orchestration management layer, including the computing power registration module, the orchestration scheduling module and the computing power coordination module. The computing network base may be the infrastructure layer, including the central cloud, network, edge computing nodes and terminals.

Therefore, in this embodiment, when the server receives a service request from a user, it determines the service demand corresponding to the service request, and sends the service demand to the computing network brain in the computing network. After receiving the computing demand, the computing network brain will search for idle computing resources according to the preset allocation strategy set in advance to determine the target computing resources, and then determine the target terminal with the target computing resources, and send a computing power occupation request to the target terminal. After receiving the occupation confirmation response fed back by the target terminal, the virtual computing environment can be directly constructed according to the target computing resources. For example, as shown in Figure 4, the terminal has completed the computing power registration. When there is a computing power demand, the computing network brain schedules and allocates the terminal computing power. Specifically, the user sends a service request to the business service end (i.e., the server end in this embodiment), and the business service end sends a computing power request to the computing network brain. After receiving the computing power request, the computing network brain searches for idle computing power resources from the computing power pool according to the resource information (CPU/storage/bandwidth, etc.) required by the business, business information (SLA information, etc.), and user information (location information, etc.) according to the policy, and after finding the idle computing power resources, sends a computing power occupation request to the terminal corresponding to the found idle computing power resources, and the terminal replies with a computing power occupation confirmation response to the computing network brain. After receiving the terminal response, the computing network brain arranges the terminal computing power resources on demand, creates an environment (such as creating a virtual computing power environment), and the computing network brain responds to the business service end for computing power allocation, and then the business service end runs the business service on the environment and feeds back the business response to the user. In addition, when the user needs to stop using the service during business use, that is, when the user needs to go offline, he can send a business stop request to the business service end, and the business service end will stop the business service and send a computing power stop request to the computing network brain. The computing network brain returns a computing power stop response to the business end, and releases the computing power resources (that is, releases the computing power resources back to the computing power pool), sends a computing power release notification to the terminal, and the terminal replies to the computing power brain with a computing power release response. The computing power brain sends the terminal computing power billing information to the computing power trading system, including but not limited to: unique identification of computing power resources, computing power type (such as CPU), actual consumption computing power information (such as TOPS), service time, business type, and SLA information. The computing power trading system then calculates the terminal income based on the computing power billing information and the preset billing rules, and pays it to the user, such as paying the income to the terminal equity user in real time or regularly through certain means. Among them, when the computing network brain sends a computing power occupation request to the terminal corresponding to the idle computing power resources found, it may send computing power requests to multiple computing power terminals/nodes according to the actual computing power needs. The computing network can create the same business environment for other computing nodes to achieve disaster recovery and digital twin functions. The billing information can be sent in full at the end of the business, or in multiple segments at the beginning, middle, and end of the business. There are other network nodes between the terminal and the computing network brain, such as edge computing nodes and cloud centers.

In this embodiment, after receiving the computing power demand sent by the server, the target computing power resources are determined according to the computing power demand and the preset allocation strategy, and after the target terminal corresponding to the target computing power resource is allowed to occupy the computing power, a virtual computing power environment is constructed to run the business data corresponding to the computing power demand, so that the computing power of the terminal can be fully used in the computing power network, the service capability of the computing power network can be improved, the construction and operation and maintenance costs of the operator can be reduced, and the waste of terminal computing power can be reduced. In addition, the computing power network path can be optimized through the end-to-end coordination mechanism, the backbone network load can be reduced, the SLA level of the service can be guaranteed, and the intelligent capability of the business can be improved; then, through certain rights and interests feedback, the idle resources of the terminal can generate income, thereby achieving a win-win effect for users and operators.

Based on the first embodiment of the present application, a second embodiment of the computing power network end-to-end collaboration method of the present application is proposed. In this embodiment, step S20 of the above embodiment determines the target computing power resources in the preset computing power pool according to the computing power demand and the preset allocation strategy, including:

Step a, determining the allocation dimension corresponding to the computing power demand according to a preset allocation strategy, and determining the terminals corresponding to all idle computing power resources in a preset computing power pool;

Step b: determine the priority corresponding to each of the terminals according to the allocation dimension, select the terminal with the highest priority as the target terminal, and use the idle computing power resources corresponding to the target terminal as the target computing power resources.

In this embodiment, after obtaining the computing power demand, the pre-set allocation strategy is first determined, and the allocation dimensions corresponding to the computing power demand are determined according to the allocation strategy (which may include but are not limited to computing power resource requirements, business information, and user location information), and it is necessary to search for idle computing power resources in the computing power pool, that is, unallocated computing power resources. The terminal corresponding to the idle computing resources, and then determine the priority of each terminal according to the allocation dimension, select the terminal with the highest priority as the target terminal, and use the idle computing resources in the target terminal as the target computing resources. The computing pool stores the computing resources of each terminal after registration. The idle computing resources can be computing resources in an idle state.

In this embodiment, the allocation dimension is determined according to a preset allocation strategy, the priority of the terminal corresponding to all idle computing resources in the computing power pool is determined according to the allocation dimension, and then the idle computing resources of the terminal with the highest priority are used as the target terminal as the target computing power resources, thereby ensuring the accuracy and effectiveness of the acquired target computing power resources.

The allocation dimensions include computing resource requirements, business information, and user location information.

The determining the priority corresponding to each of the terminals according to the allocation dimension includes:

Step b1, determining the location priority corresponding to each of the terminals according to the user location information;

Step b2, determining the service level agreement priority corresponding to each terminal according to the service information;

Step b3, determining the computing power requirement priority corresponding to each terminal according to the computing power resource requirement;

Step b4, calculating the priority corresponding to each of the terminals according to the location priority, the service level agreement priority and the computing power requirement priority.

In this embodiment, the priorities of all terminals under different allocation dimensions can be determined first, such as location priority, service level agreement priority, and computing power requirement priority. Among them, the method of determining the location priority, service level agreement priority, and computing power requirement priority can be performed according to the conventional settings, so as to determine the final priority corresponding to each terminal, for example, giving priority to the location priority, then considering the service level agreement priority, and finally considering the computing power requirement priority. Among them, the user location information can be the user location distance. The business information can be the business SLA requirements. The computing power resource requirements may include (computing/storage/bandwidth). After determining the priority corresponding to each terminal, the terminal with the highest priority can be selected as the target terminal. For example, a terminal with high location priority, service level agreement priority, and computing power requirement priority is selected as the target terminal. In addition, different weights may be set for the location priority, service level agreement priority, and computing power requirement priority, such as the first weight, the second weight, and the third weight, and the first product between the location priority and the first weight, the second product between the service level agreement priority and the second weight, and the third product between the computing power requirement priority and the third weight may be calculated, and the sum of the first product, the second product, and the third product may be used as the priority of the terminal, and then the priority of each terminal may be compared in turn to select the terminal with the highest priority.

In addition, if the terminal has completed the computing power registration, when there is a business demand in the adjacent terminal, the computing network brain dispatches and allocates the terminal computing power to provide network collaboration. For example, as shown in Figure 5, adjacent terminal users A and B request to use the service (such as video call), and the service service end sends the computing power demand to the computing network brain; the computing network brain receives the computing power request, and according to the resource information (CPU/storage/bandwidth, etc.), service information (SLA information, etc.), user information (location information, etc.) required by the business, it searches for the computing power resources of the adjacent terminal from the computing power pool according to the strategy, and after finding the terminal computing power resources (such as A, B), it sends a computing power occupation request to the terminal, which carries the computing power occupation demand and network collaboration information (terminal direct connection protocol, address, port information); wherein, the preset allocation strategy can be to select resources according to the priority of the resource selection factors in Table 1 below, such as giving priority to one or more computing power nodes that are close to the user, secondly meet the business SLA requirements, and finally meet the computing power demand (computing/storage/network capacity).

Table 1

The terminal replies with an occupation confirmation response to the computing network brain; after receiving the terminal response, the computing network brain manages the terminal computing power as needed, creates virtual computing power resources, and the business service runs business services on the resources and returns a business response to the user; Terminal A and Terminal B start using the business and use the network collaboration function (based on the network collaboration information, point-to-point communication is achieved); when the user stops using the business, the business service stops the business service and sends a request to stop computing power to the computing network brain; the computing network brain returns a confirmation response to the business service; the computing network brain releases the computing power resources back to the computing power pool and sends a computing power release message to the terminal; the terminal replies with a computing power release response to the computing network brain; the computing network brain sends a message to the computing power trading system stating that the terminal has completed the computing power release. The billing information includes, but is not limited to: unique identification of computing power resources, computing power type (such as network bandwidth), actual consumption computing power information (such as network traffic), service time, business type, and SLA information; the computing power trading system calculates the terminal's income based on the above computing power billing information and billing rules, and pays the income to the user to whom the terminal belongs in real time or regularly through certain means. Among them, according to the actual computing power needs, it is possible to send computing power requests to multiple computing power terminals/nodes (such as terminals A and B). Here, only one terminal A is written for simplicity. Terminal direct connection can be through general protocols such as Bluetooth, wifi, and microwave. The sending of billing information can send complete billing information at the end of the business, or it can send multiple segmented billing information at the beginning, middle, and end of the business. Here, for simplicity, only sending complete billing information once is used as an example. In this embodiment, network collaboration takes point-to-point communication as an example. Similarly, this embodiment can also be expanded to a multi-party direct connection and transit implementation method. By directly connecting the network between terminals, the network resource consumption of the cloud edge network can be greatly reduced, and the SLA capability of the business can be improved. There are other network nodes between the terminal and the computing network brain, such as edge computing nodes and cloud centers.

In this embodiment, after allocating dimensions including computing resource requirements, business information and user location information, the location priority, service level agreement priority and computing power requirement priority corresponding to each terminal can be calculated in turn, and the priority corresponding to each terminal can be determined based on the location priority, service level agreement priority and computing power requirement priority, thereby ensuring the accuracy and effectiveness of the priority corresponding to each terminal.

Constructing a virtual computing environment according to the target computing resources includes:

Step c: if there are multiple target terminals corresponding to the target computing power resources, one target terminal is selected from each of the target terminals as the cluster brain, and a computing power cluster including each of the target terminals is constructed with the cluster brain, and a virtual computing power environment is constructed according to all the target computing power resources corresponding to the computing power cluster;

In this embodiment, when the computing power brain finds that there are multiple target computing power resources through query, it can build a computing power cluster and create a virtual computing power environment through the computing power cluster, and run business data through various target terminals in the virtual computing power environment. For example, when multiple adjacent terminals have completed computing power registration and there is a demand for computing power, the computing network brain dispatches and allocates computing power of adjacent terminals to provide computing + network collaboration (computing power cluster). For example, as shown in FIG6 , a user requests to use a service (such as big data computing), and the service server (such as a big data center) sends a service request to the computing network brain; the computing network brain receives the service request, and searches for computing resources from the computing power pool according to the policy based on the resource information (CPU/storage/bandwidth, etc.), service information (SLA information, etc.), and user information (location information, etc.) required by the service, and forms a computing power cluster with several adjacent terminals, and selects one of the terminals as the cluster brain (responsible for the distribution and aggregation of services), and sends a computing power occupation request to the cluster terminal, which carries the computing power occupation demand and computing power coordination information (cluster brain ID, surrounding terminal ID, location information, and coordination interface); the terminal replies to the computing network brain with an occupation confirmation response; after receiving the terminal response, the computing network brain manages the terminal computing power as needed, creates a virtual computing power cluster, and the service server runs the service on the resources and returns the service start response to the user; the cluster terminal starts to process the service, and the cluster brain establishes a link with other terminals. Distribute computing power requests to other terminals; after receiving computing power requests, other terminals process business data and return processing results to the cluster brain; the cluster brain summarizes the processing results and returns them to the business server and user as needed; when the user stops using the business (such as data processing is completed), the business server stops the business service and sends a request to stop computing power to the computing network brain; the computing network brain returns a confirmation response to the business server; the computing network brain releases computing power resources back to the computing power pool and sends a computing power release message to the terminal; the terminal replies to the computing network brain with a computing power release response; the computing network brain sends the terminal computing power billing information to the computing power trading system, including but not limited to: unique identification of computing power resources, computing power type (such as CPU, memory, bandwidth), actual consumption computing power information (such as TOPS, network traffic), service time, business type, SLA information; the computing power trading system calculates the terminal's deserved income based on the above computing power billing information and billing rules, and pays the income to the user to whom the terminal belongs in real time or regularly through certain means. Among them, there are many international terminals, and only two terminals (1 cluster brain + 1 adjacent terminal) are written here for simplicity. The actual implementation may also include dynamic information updates, such as when a terminal joins or leaves the cluster. The specific steps are not given in detail. The billing information can be sent by sending complete billing information at the end of the service, or by sending multiple segmented billing information at the beginning, middle, and end of the service. For simplicity, only sending complete billing information once is used as an example. There are other network nodes between the terminal and the computing network brain, such as edge computing nodes and cloud centers.

Step d: If there is a target terminal corresponding to the target computing power resource, a virtual computing power environment is constructed according to the target computing power resource.

When there is only one target terminal corresponding to the target computing power resource, the virtual Virtual computing power environment. For example, the terminal has completed computing power registration. When a nearby terminal has computing power demand, the computing network brain dispatches and allocates terminal computing power to provide computing collaboration (information sharing and co-computing). For example, as shown in Figure 7, the terminal (A) requests to use a service (such as autonomous driving), and the service server (such as the autonomous driving platform) sends the computing power demand to the computing network brain; the computing network brain receives the computing power request, and searches for the computing power resources of the nearby terminal from the computing power pool according to the resource information (CPU/storage/bandwidth, etc.), service information (SLA information, etc.), and user information (location information, etc.) required by the service according to the policy. After finding the computing power resources, it sends a computing power occupation request to the computing power node (such as terminal A), and the request carries the computing power occupation demand and computing power collaboration information (near terminal information: surrounding terminal labels, location information, collaborative interface, etc.); the terminal replies to the computing network brain with an occupation confirmation response; after receiving the terminal response, the computing network brain manages the terminal computing power as needed, creates virtual computing power resources, and the service server runs the service on the resources and returns the service response to the user; the terminal starts using the service and performs computing collaboration (such as Send the movement trend information calculated by the control system to the nearby terminal to reduce the detection and calculation of the nearby terminal; it can also receive the information of the nearby terminal and use it to control its own system); when the user stops using the service, the service service stops the service and sends a request to stop the computing power to the computing network brain; the computing network brain returns a confirmation response to the service service; the computing network brain releases the computing power resources back to the computing power pool and sends a computing power release message to the terminal; the terminal replies to the computing network brain with a computing power release response; the computing network brain sends the terminal computing power billing information to the computing power trading system, including but not limited to: unique identification of computing power resources, computing power type (such as CPU), actual consumption computing power information (such as TOPS), service time, business type, SLA information; the computing power trading system calculates the terminal's deserved income based on the above computing power billing information and billing rules, and pays the income to the user to whom the terminal belongs in real time or regularly through certain means. Among them, according to the actual computing power needs, it is possible to send computing power requests to multiple computing power terminals/nodes. Here, only one terminal is written for simplicity. The actual implementation can also include dynamic information updates, such as changes in peripheral terminal information. The specific steps are not given in detail. There may be multiple adjacent terminals, but only one adjacent terminal is shown for simplicity. The billing information can be sent by sending complete billing information at the end of the service, or by sending multiple segmented billing information at the beginning, middle, and end of the service. For simplicity, only sending complete billing information once is used as an example. There are other network nodes between the terminal and the computing network brain, such as edge computing nodes and cloud centers, which are omitted for simplicity.

In this embodiment, when there are multiple target terminals, a computing power cluster including multiple target terminals is constructed, and a virtual computing power environment is constructed through all target computing power resources in the computing power cluster, or when there is only one target terminal, a virtual computing power environment is directly constructed with the target computing power resources, thereby ensuring the effectiveness of the construction of the virtual computing power environment.

After constructing a virtual computing power environment according to the target computing power resources so as to run the business data corresponding to the computing power demand in the virtual computing power environment, the method further includes:

Step e, after receiving the stop computing power request sent by the server, stop running the business data according to the stop computing power request, release the target computing power resources in the virtual computing power environment, generate billing data for applying the target computing power resources, and send the billing data to the target terminal.

In this embodiment, when the business data is running in the virtual computing environment, if a stop computing request is received from the server, the business data will be stopped in the virtual computing environment, and the server will be responded to, and then the target computing resources in the virtual computing environment will be released to the computing pool. In addition, the corresponding billing data will be generated, and the final income will be calculated through the computing power trading system and fed back to the target terminal.

In this embodiment, after receiving a request to stop computing power, the business data is stopped, the target computing power resources are released, billing data is generated, and the billing data is sent to the target terminal, thereby achieving a win-win effect for the user and the target terminal.

After receiving the computing power demand sent by the server, before determining the target computing power resources in the preset computing power pool according to the computing power demand and the preset allocation strategy, it includes:

Step f, after receiving the registration request sent by the terminal, storing the computing power information in the registration request as the idle computing power resources of the terminal in a preset computing power pool.

In this embodiment, before performing end-to-end collaboration of the computing power network, the terminal needs to be registered in the computing power network. Therefore, after receiving the registration request sent by the terminal, the computing power network can directly store the computing power information carried in the registration request as computing power resources in the computing power pool. And the default initial state during storage is the idle state.

In this embodiment, after receiving the registration request, the idle computing resources of the terminal are stored in the computing pool. This will enable end-to-end collaboration of the computing network to be achieved in the future.

After storing the computing power information in the registration request as the idle computing power resources corresponding to the terminal in a preset computing power pool, the method further includes:

Step g, after receiving the computing power change request sent by the terminal, if the idle computing power resources corresponding to the terminal have been allocated, determine whether to perform computing power migration on the idle computing power resources, and update the idle computing power resources corresponding to the terminal according to the determination result;

Step h, after receiving the deregistration request sent by the terminal, if the idle computing power resources corresponding to the terminal have been allocated, the idle computing power resources are migrated, and after the computing power migration is completed, the idle computing power resources corresponding to the terminal in the computing power pool are deleted.

In this embodiment, for example, the user has signed an agreement with the operator for the terminal to participate in the computing power network, and the user terminal registers the computing power after it is turned on and connected to the network. As shown in Figure 8, the terminal initiates a registration request to the computing network brain according to the user's settings when certain conditions are met (such as entering the standby state), and the request carries computing power information. The computing power information includes but is not limited to: the terminal's unique computing power identification, computing/storage/network capabilities (such as the number of processor types, the number of storage types, the number of network types), and location information (such as longitude and latitude, altitude); the computing network brain receives the registration information, saves the computing power information in the computing power pool, and updates the computing power network resource topology (adds terminal computing power nodes, identifies the connection information between the terminal and other computing power nodes), and replies to the terminal with a successful registration response;

The computing network brain periodically sends keep-alive messages to the terminal according to the predetermined strategy; the terminal receives the keep-alive message and replies with a keep-alive response; when the terminal reaches a certain change condition (for example, the terminal enters the working mode, resulting in a change in the shared computing power information, or a change in the location), it sends a computing power information change request to the computing network brain; the computing network brain receives the computing power change information, and if the computing power has been allocated, the computing network brain determines whether computing power migration is required; then it updates the computing power information and updates the computing power network resource topology, and replies to the terminal with a computing power information change response; when the terminal reaches the condition of not providing computing power (for example, power off), the terminal initiates a deregistration request to the computing network brain; the computing network brain receives the deregistration information, and if the computing power has been allocated, the computing network brain performs computing power migration; then it deletes the computing power information from the computing power pool, updates the computing power network resource topology, and replies to the terminal with a computing power information deregistration response. Among them, there are other network nodes between the terminal and the computing network brain, such as edge computing nodes and cloud centers. If the registration information or response information is lost, the terminal needs to resend the registration message. If there is no response to the keep-alive request for more than a certain number of times, the computing network brain determines that the terminal is disconnected, performs the deregistration process, deletes the terminal computing power information, and updates the computing power resource topology.

In this embodiment, the validity of the computing power pool can be guaranteed by updating the idle computing power resources after receiving a computing power change request and deleting the idle computing power resources after receiving a deregistration request.

In addition, the present application also provides a computing network, which runs the computing network end-to-end collaboration method in any of the above embodiments, including a server, a computing network brain, and multiple terminals;

The specific implementation methods of the computing power network of the present application are basically the same as the various embodiments of the above-mentioned computing power network end-to-end collaboration method, and will not be repeated here.

In addition, the present application also provides a computing power network end-to-end collaborative device, which includes: a memory, a processor, and a computing power network end-to-end collaborative program stored on the memory; the processor is used to execute the computing power network end-to-end collaborative program to implement the steps of each embodiment of the above-mentioned computing power network end-to-end collaborative method.

The present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a One or more programs, and the one or more programs can also be executed by one or more processors to implement the steps of each embodiment of the above-mentioned computing power network end-to-end collaboration method.

The specific implementation methods of the computer-readable storage medium of the present application are basically the same as the embodiments of the above-mentioned computing power network end-to-end collaboration method, and will not be repeated here.

In this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of more restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or system that includes the element.

The serial numbers of the embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

The above are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

A computing power network end-to-end collaboration method, comprising:

After receiving the computing power demand sent by the server, determine the target computing power resources in the preset computing power pool according to the computing power demand and the preset allocation strategy;

A computing power occupation request is sent to the target terminal corresponding to the target computing power resource, and after receiving an occupation confirmation response from the target terminal based on the computing power occupation request, a virtual computing power environment is constructed according to the target computing power resource, so as to run the business data corresponding to the computing power demand in the virtual computing power environment.
The computing power network end-to-end collaboration method according to claim 1, wherein determining the target computing power resources in a preset computing power pool according to the computing power demand and the preset allocation strategy comprises:

Determine the allocation dimension corresponding to the computing power demand according to the preset allocation strategy, and determine the terminals corresponding to all idle computing power resources in the preset computing power pool;

The priority corresponding to each of the terminals is determined according to the allocation dimension, and the terminal with the highest priority is selected as the target terminal, and the idle computing power resources corresponding to the target terminal are used as the target computing power resources.
The computing power network end-to-end collaboration method according to claim 2, wherein the allocation dimensions include computing power resource requirements, business information, and user location information,

The determining the priority corresponding to each of the terminals according to the allocation dimension includes:

Determine the location priority corresponding to each of the terminals according to the user location information;

Determine the service level agreement priority corresponding to each of the terminals according to the service information;

Determine the computing power requirement priority corresponding to each terminal according to the computing power resource requirement;

The priority corresponding to each of the terminals is calculated according to the location priority, the service level agreement priority and the computing power requirement priority.
The computing power network end-to-end collaboration method according to claim 1, wherein the step of constructing a virtual computing power environment according to the target computing power resources comprises:

If there are multiple target terminals corresponding to the target computing power resources, one target terminal is selected from each of the target terminals as the cluster brain, and a computing power cluster including each of the target terminals is constructed with the cluster brain, and a virtual computing power environment is constructed according to all the target computing power resources corresponding to the computing power cluster;

If there is a target terminal corresponding to the target computing power resource, a virtual computing power environment is constructed according to the target computing power resource.
The computing power network end-to-end collaboration method according to claim 1, wherein the step of constructing a virtual computing power environment according to the target computing power resources so as to run the business data corresponding to the computing power demand in the virtual computing power environment comprises:

After receiving the stop computing power request sent by the server, the business data is stopped according to the stop computing power request, and the target computing power resources in the virtual computing power environment are released, and the billing data of the target computing power resources is generated, and the billing data is sent to the target terminal.
The computing power network end-to-end collaboration method according to claim 1, wherein after receiving the computing power demand sent by the server, before determining the target computing power resource in the preset computing power pool according to the computing power demand and the preset allocation strategy, it includes:

After receiving the registration request sent by the terminal, the computing power information in the registration request is stored in a preset computing power pool as the idle computing power resources corresponding to the terminal.
The computing power network end-to-end collaboration method according to claim 6, wherein after storing the computing power information in the registration request as the idle computing power resources corresponding to the terminal in a preset computing power pool, the method further comprises:

After receiving the computing power change request sent by the terminal, if the idle computing power resources corresponding to the terminal have been allocated, determine whether to perform computing power migration on the idle computing power resources, and update the idle computing power resources corresponding to the terminal according to the determination result;

After receiving the deregistration request sent by the terminal, if the idle computing resources corresponding to the terminal have been allocated, the idle computing resources are migrated, and after the computing migration is completed, the computing resources in the computing pool are deleted. Idle computing resources corresponding to the terminal.
A computing power network, wherein the computing power network end-to-end collaboration method as described in any one of claims 1 to 7 is executed, comprising a server, a computing network brain and a plurality of terminals;

The server is configured to send the computing power requirement corresponding to the business request to the computing network brain after receiving the business request;

The computing network brain is configured to determine the target computing power resources in the preset computing power pool according to the computing power requirements and the preset allocation strategy after receiving the computing power requirements sent by the server, determine the target terminal of the target computing power resources among the multiple terminals, and send a computing power occupation request to the target terminal;

The target terminal is configured to generate an occupation confirmation response after receiving the computing power occupation request and determining the computing power occupation according to the computing power occupation request, and send the occupation confirmation response to the computing network brain;

The computing network brain is configured to construct a virtual computing environment based on the target computing resources so as to run the business data corresponding to the computing demand in the virtual computing environment.
An electronic device comprises: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the steps of the computing power network end-to-end collaboration method described in any one of claims 1 to 7.
A storage medium storing one or more programs, which, when executed by an electronic device including multiple application programs, enables the electronic device to perform the steps of the computing power network end-to-end collaboration method described in any one of claims 1 to 7.