WO2021023042A1 - 边缘计算服务器的寻找方法及相关设备 - Google Patents

边缘计算服务器的寻找方法及相关设备 Download PDF

Info

Publication number
WO2021023042A1
WO2021023042A1 PCT/CN2020/104835 CN2020104835W WO2021023042A1 WO 2021023042 A1 WO2021023042 A1 WO 2021023042A1 CN 2020104835 W CN2020104835 W CN 2020104835W WO 2021023042 A1 WO2021023042 A1 WO 2021023042A1
Authority
WO
WIPO (PCT)
Prior art keywords
server
task
edge
terminal
edge server
Prior art date
Application number
PCT/CN2020/104835
Other languages
English (en)
French (fr)
Inventor
王楠楠
齐飞
黄梦龙
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2021023042A1 publication Critical patent/WO2021023042A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/40—Support for services or applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/1021—Server selection for load balancing based on client or server locations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/1023—Server selection for load balancing based on a hash applied to IP addresses or costs
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services

Definitions

  • the invention relates to the technical field of mobile edge computing, in particular to a method for finding an edge computing server and related equipment.
  • the existing mobile edge computing architecture is mainly a centralized resource configuration and scheduling architecture from the central cloud to the edge cloud to the terminal. All resources and scheduling data in this architecture are distributed and configured by the central cloud. Just do task execution. This leads to inefficient decision-making and inability to respond to terminal demands with the lowest possible delay. In addition, the request of the terminal needs to be sent to the central cloud to make a decision, and then the decision is returned, which increases network overhead and causes a waste of resources.
  • the embodiment of the application discloses a method for finding an edge computing server and related equipment, which can respond to the needs of the terminal with the lowest delay and save network overhead.
  • an embodiment of the application discloses a method for finding an edge computing server.
  • the above method is applied to an edge computing system.
  • the edge computing system includes a terminal, a task offloading device, a central server, and multiple edge servers; the above method includes:
  • the task offloading device receives a resource search request sent by the terminal, the resource searching request carries information about computing resources needed to complete the task, and the task offloading device is deployed in a wireless access point accessed by the terminal;
  • the task offloading device determines, according to the resource search request, that there is a central server and/or at least one edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the foregoing task offloading device selects a server for performing the foregoing task from a central server and/or at least one edge server that meets the requirements of the foregoing computing resources.
  • the above-mentioned task offloading device may be a mobile edge computing (mobile edge computing, MEC) server or may be another device that conforms to the same communication protocol standard as the MEC server and can implement related functions.
  • MEC mobile edge computing
  • a service uninstallation agent is deployed on the task uninstallation device, the service uninstallation agent may be a software or an application, and the task uninstallation device runs the service uninstallation agent to implement the operations performed by the task uninstallation device in the embodiment of this application. .
  • the edge cloud only serves as a task execution architecture.
  • the present invention uses task offloading deployed in wireless access points.
  • the equipment completes the scheduling of resources and data, that is, the nearest server that meets the terminal needs is selected to calculate and complete the tasks that the terminal needs to calculate, which achieves the lowest delay in responding to the needs of the terminal and saves network overhead.
  • the task offloading device selects a server for performing the above task from a central server and/or at least one edge server that meets the requirements of the computing resources, including:
  • the task offloading device sends the resource search request to a central server and/or at least one edge server that meets the requirements of the computing resources;
  • the foregoing task offloading device receives the address information sent by the central server and/or at least one edge server that meets the requirements of the foregoing computing resources according to the foregoing resource search request;
  • the task offloading device selects a server for executing the task from a central server and/or at least one edge server that meets the requirements of the computing resources according to the order of the received address information.
  • the aforementioned address information includes the Internet Protocol IP address or media storage control address (MAC) of the central server and/or at least one edge server that meets the requirements of the aforementioned computing resources, etc., which can route packets to the server. .
  • the address information may be used to instruct the task offloading device to send the task to the server corresponding to the address information for execution.
  • the task offloading device selects a server for performing the above task from a central server and/or at least one edge server that meets the requirements of the computing resources, including:
  • the task offloading device sends the resource search request to a central server and/or at least one edge server that meets the requirements of the computing resources;
  • the foregoing task offloading device receives the address information and location information sent by the central server and/or at least one edge server that meets the requirements of the foregoing computing resources according to the foregoing resource search request;
  • the task offloading device calculates the distance between the central server and/or at least one edge server required by the computing resource and the terminal according to the received location information and the location information of the terminal;
  • the task offloading device selects the server closest to the terminal from the central server and/or at least one edge server that meets the requirements of the computing resources according to the distance obtained by the calculation.
  • the above-sent request can instruct the server that meets the above-mentioned computing resource requirements among the servers that receive the request to send its own location information to the task offloading device, so that the closest to the terminal can be found to meet the requirements.
  • the server calculates and completes the tasks that the terminal needs to calculate, so as to further reduce the delay and network overhead.
  • the task offloading device selects a server for performing the above task from a central server and/or at least one edge server that meets the requirements of the computing resources, including:
  • the foregoing task offloading device sends the foregoing resource search request and the location information of the foregoing terminal to a central server and/or at least one edge server that meets the requirements of the foregoing computing resources;
  • the task offloading device receives the address information sent by the central server and/or at least one edge server that meets the requirements of the computing resources according to the resource search request, and the central server and/or at least one edge server that meets the requirements of the computing resources and the foregoing The distance between terminals;
  • the task offloading device selects the distance from the central server and/or at least one edge server that meets the requirements of the computing resources according to the distance between the central server and/or at least one edge server that meets the requirements of the computing resources and the terminal.
  • the nearest server of the aforementioned terminal is used to perform the aforementioned tasks.
  • the task offloading device selects a server for performing the task from a central server and/or at least one edge server that meets the requirements of the computing resources according to the order of the received address information.
  • the above method also includes:
  • the task offloading device sends first response information in response to the resource search request to the terminal;
  • the task uninstallation device receives the task sent by the terminal according to the first response information
  • the task uninstallation device sends the task to the selected server.
  • the task offloading device selects the server for executing the task for the terminal
  • the task is sent to the selected server for execution by interacting with the terminal, which improves the efficiency of task offloading.
  • the task offloading device determines, according to the resource search request, that there is no central server or edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the task offloading device sends the resource search request to the preset edge server
  • the task offloading device receives the second response information sent by the preset edge server, and the second response information is that among the servers that have received the resource search request broadcast by the preset edge server, the server that meets the requirements of the computing resources sends the Sent by a preset edge server, the second response information includes address information and location information of the server that meets the requirements of the computing resources;
  • the task offloading device calculates the distance between the server that meets the computing resource requirement and the terminal according to the location information of the server that meets the computing resource requirement and the location information of the terminal;
  • the task offloading device selects the server closest to the terminal from among the servers that meet the requirements of the computing resources according to the distance obtained by the calculation.
  • the preset edge cloud server in the embodiment of this application is a preset central cloud server or edge cloud server that does not meet the requirements of the computing resources within the space covered by the wireless access point accessed by the terminal. To find other cloud servers that meet the needs to perform the above tasks.
  • the task offloading device will send the request of the terminal to the preset edge server to expand the search range.
  • the request can also be used to instruct the server that meets the demand to send its own location information to the task offloading device, so that it can find the server that meets the demand closest to the terminal to calculate and complete the task that the terminal needs to calculate, so as to further reduce time Delay and network overhead.
  • the task offloading device determines, according to the resource search request, that there is no central server or edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the task offloading device sends the resource search request and the location information of the terminal to a preset edge server;
  • the task offloading device receives the third response information sent by the preset edge server, where the third response information is the server that receives the resource search request and the location information of the terminal broadcast by the preset edge server, and satisfies the computing resource
  • the third response information includes the address information of the server that meets the requirements of the computing resources and the distance between the server that meets the requirements of the computing resources and the terminal ;
  • the task offloading device selects the server closest to the terminal from the servers that meet the computing resource requirements based on the distance between the server that meets the computing resource requirements and the terminal to perform the task.
  • the task offloading device searches for the nearest server that meets the requirements for the terminal based on the location information of the terminal to calculate and complete the tasks that the terminal needs to calculate.
  • the task offloading device determines, according to the resource search request, that there is no central server or edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the task offloading device sends the resource search request to the preset edge server
  • the task offloading device receives the fourth response information sent by the preset edge server, and the fourth response information is that among the servers that have received the resource search request broadcast by the preset edge server, the server that meets the requirements of the computing resources sends the Sent by a preset edge server, the fourth response information includes address information of the server that meets the requirements of the computing resources;
  • the task offloading device selects the server that executes the task according to the order of receiving the address information of the server that meets the requirements of the computing resource.
  • the embodiment of the present application discloses a method for finding an edge computing server.
  • the above method is applied to an edge computing system.
  • the edge computing system includes a terminal, a task offloading device, a central server, and multiple edge servers.
  • the uninstallation device is deployed in the wireless access point accessed by the terminal; the method includes:
  • the preset edge server receives the resource search request sent by the task offloading device, the resource search request carries information about computing resources needed to complete the task, and the preset edge server is a preset wireless access point that is accessed by the terminal. In the case that there is no central server or edge server that meets the requirements of the above computing resources in the space range of, it is used to find other servers that meet the requirements to perform the above tasks;
  • the above-mentioned preset edge server determines that the above-mentioned preset edge server does not meet the requirement of the above-mentioned computing resources according to the above-mentioned resource search request;
  • the preset edge server broadcasts the resource search request to other edge servers and/or central servers.
  • the method further includes:
  • the preset edge server receives the second response information sent by the server that meets the computing resource demand among the other edge servers and/or the central server, and the second response information includes the address information of the server that meets the computing resource demand and location information;
  • the preset edge server sends the second response information to the task offloading device.
  • the method before the preset edge server determines that the preset edge server does not meet the requirements of the computing resources according to the resource search request, the method further includes:
  • the foregoing preset edge server receives the location information of the foregoing terminal sent by the foregoing task offloading device;
  • the method further includes:
  • the preset edge server broadcasts the location information of the terminal to the other edge servers and/or the central server.
  • the method further includes:
  • the preset edge server receives the third response information sent by the server that meets the computing resource demand among the other edge servers and/or the central server, and the third response information includes the address information of the server that meets the computing resource demand; And including the distance between the foregoing server that meets the foregoing computing resource requirements and the foregoing terminal;
  • the preset edge server sends the third response information to the task offloading device.
  • the method further includes:
  • the foregoing preset edge server receives fourth response information sent by a server that meets the requirements of the computing resources among the other edge servers and/or central servers, where the fourth response information includes address information of the server that meets the requirements of the computing resources;
  • the preset edge server sends the fourth response information to the task offloading device.
  • the method further includes:
  • the foregoing preset edge server determines, according to the foregoing resource search request, that the foregoing preset edge server meets the requirements of the foregoing computing resources;
  • the preset edge server sends the address information of the preset edge server to the task offloading device.
  • the embodiments of the present application disclose a method for finding an edge computing server.
  • the method is applied to an edge computing system.
  • the edge computing system includes a terminal, a task offloading device, a central server, and multiple edge servers; the above method includes:
  • the first edge server receives a resource search request broadcast by the preset edge server, the resource search request carries information about computing resources needed to complete the task; the resource search request is that the preset edge server receives the resource search request sent by the task offloading device Then, it is broadcast when it is judged that it does not meet the requirements of the above computing resources; the above preset edge server is a preset center that does not meet the requirements of the above computing resources within the space covered by the wireless access point accessed by the terminal In the case of a server or an edge server, it is a server used to find other servers that meet the requirements to perform the above tasks; the above task offloading device is deployed in the wireless access point accessed by the above terminal; the above first edge server includes the exceptions in the above system Preset servers other than edge servers;
  • the first edge server determines, according to the resource search request, that the first edge server meets the requirements of the computing resources
  • the first edge server sends second response information in response to the resource search request to the task offloading device, and the second response information includes address information and location information of the first edge server.
  • the method before the first edge server determines that the first edge server meets the requirements of the computing resources according to the resource search request, the method further includes:
  • the first edge server receives the location information of the foregoing terminal broadcast by the preset edge server;
  • the method further includes:
  • the first edge server calculates the distance between the first edge server and the terminal according to the location information of the terminal;
  • the first edge server sends third response information in response to the resource search request to the task offloading device.
  • the third response information includes the address information of the first edge server, and includes information between the first edge server and the terminal. distance.
  • the method includes: the first edge server sends a response to the resource to the task offloading device The fourth response information of the search request, where the fourth response information includes the address information of the first edge server.
  • an embodiment of the present invention provides a mobile edge computing task offloading device, the above task offloading device is a device in an edge computing system, the above edge computing system further includes a terminal, a central server, and multiple edge servers; the above task offloading device include:
  • a receiving unit configured to receive a resource search request sent by the terminal, the resource search request carries information about computing resources required to complete the task, and the task offloading device is deployed in the wireless access point accessed by the terminal;
  • the determining unit is configured to determine, according to the resource search request, that there is a central server and/or at least one edge server that meets the demand for computing resources in the space covered by the wireless access point accessed by the terminal;
  • the selection unit is configured to select a server for performing the above-mentioned task from a central server and/or at least one edge server that meets the requirements of the above-mentioned computing resources.
  • the above selection unit is specifically used for:
  • a server for performing the above task is selected from a central server and/or at least one edge server that meets the requirements of the computing resource.
  • the above selection unit is specifically used for:
  • the server closest to the foregoing terminal is selected to perform the foregoing task.
  • the above selection unit is specifically used for:
  • the central server and/or at least one edge server that meets the requirements of the above computing resources and the above-mentioned terminal from the central server and/or at least one edge server that meets the requirements of the above-mentioned computing resources, select the one closest to the terminal.
  • the server is used to perform the above tasks.
  • the above-mentioned task offloading device further includes a sending unit, which is configured to send from a central server and/or at least one edge that meets the requirements of the computing resources according to the order of the received address information by the above-mentioned selection unit. After selecting the server for performing the above task among the servers, sending the first response information in response to the above resource search request to the above terminal;
  • the receiving unit is further configured to receive the task sent by the terminal according to the first response information
  • the sending unit is further configured to send the task to the selected server.
  • the above-mentioned task offloading device further includes a sending unit and a computing unit;
  • the foregoing determining unit is further configured to determine, according to the foregoing resource search request, that there is no central server or edge server that meets the requirements of the foregoing computing resources in the space covered by the wireless access point accessed by the foregoing terminal;
  • the sending unit is configured to send the resource search request to the preset edge server
  • the receiving unit is further configured to receive second response information sent by the preset edge server, where the second response information is a server that has received the resource search request broadcast by the preset edge server and meets the requirements of the computing resources. Sent by the server to the foregoing preset edge server, the foregoing second response information includes address information and location information of the foregoing server that meets the foregoing computing resource requirements;
  • the calculation unit is configured to calculate the distance between the server that meets the computing resource demand and the terminal based on the location information of the server that meets the computing resource demand and the location information of the terminal;
  • the selection unit is further configured to select the server closest to the terminal from among the servers that meet the requirements of the computing resources according to the distance obtained by the calculation.
  • the above-mentioned task offloading device further includes a sending unit
  • the foregoing determining unit is further configured to determine, according to the foregoing resource search request, that there is no central server or edge server that meets the requirements of the foregoing computing resources in the space covered by the wireless access point accessed by the foregoing terminal;
  • the sending unit is configured to send the resource search request and the location information of the terminal to a preset edge server;
  • the receiving unit is further configured to receive third response information sent by the preset edge server, where the third response information is a server that receives the resource search request and the location information of the terminal broadcast by the preset edge server, The server that meets the computing resource requirements sends the third response information to the preset edge server, and the third response information includes the address information of the server that meets the computing resource requirements, and the relationship between the server that meets the computing resource requirements and the terminal.
  • the distance between the third response information sent by the preset edge server where the third response information is a server that receives the resource search request and the location information of the terminal broadcast by the preset edge server, The server that meets the computing resource requirements sends the third response information to the preset edge server, and the third response information includes the address information of the server that meets the computing resource requirements, and the relationship between the server that meets the computing resource requirements and the terminal.
  • the selection unit is further configured to select the server closest to the terminal from the servers that meet the requirements of the computing resources according to the distance between the server that meets the requirements of the computing resources and the terminal to perform the above tasks.
  • an embodiment of the present invention provides an edge server.
  • the edge server is a server in an edge computing system.
  • the edge computing system further includes a terminal, a task offloading device, a central server, and multiple edge servers other than the edge server.
  • the receiving unit is configured to receive a resource search request sent by the task offloading device, the resource search request carries information about computing resources required to complete the task, and the edge server is a preset wireless access point covered by the terminal If there is no central server or edge server that meets the requirements of the above computing resources in the space, the server is used to find other servers that meet the requirements to perform the above tasks;
  • a determining unit configured to determine, according to the resource search request, that the edge server does not meet the requirements of the computing resources
  • the broadcasting unit is used to broadcast the resource search request to other edge servers and/or central servers.
  • the foregoing edge server further includes a sending unit
  • the receiving unit is further configured to receive the request from a server that meets the computing resource requirements of the other edge servers and/or central servers after the broadcasting unit broadcasts the resource search request to the other edge servers and/or central servers.
  • Second response information where the second response information includes address information and location information of the server that meets the requirements of the computing resources;
  • the sending unit is configured to send the second response information to the task offloading device.
  • the receiving unit is further configured to receive the position of the terminal sent by the task offloading device before the determining unit determines that the edge server does not meet the requirements of the computing resources according to the resource search request information;
  • the broadcasting unit is further configured to broadcast the location information of the terminal to the other edge servers and/or central servers after the determining unit determines that the edge server does not meet the requirements of the computing resources according to the resource search request.
  • the receiving unit is further configured to receive the other edge server after the broadcasting unit broadcasts the resource search request and the location information of the terminal to the other edge server and/or central server. And/or the third response information sent by the server that meets the requirements of the computing resources in the central server, the third response information includes the address information of the servers that meet the requirements of the computing resources, and the information that meets the requirements of the computing resources.
  • the sending unit is configured to send the third response information to the task offloading device.
  • the receiving unit is further configured to receive the request from the other edge servers and/or central server after the broadcasting unit broadcasts the resource search request to other edge servers and/or central servers.
  • Fourth response information sent by the server that meets the demand for computing resources where the fourth response information includes address information of the server that meets the demand for computing resources;
  • the sending unit is configured to send the fourth response information to the task offloading device.
  • the foregoing edge server further includes a sending unit
  • the determining unit is further configured to, after the receiving unit receives the resource search request sent by the task offloading device, determine, according to the resource search request, that the edge server meets the computing resource requirement;
  • the sending unit is configured to send the address information of the edge server to the task offloading device.
  • an embodiment of the present invention provides an edge server.
  • the edge server is a server in an edge computing system.
  • the edge computing system further includes a terminal, a task offloading device, an edge server other than the edge server, and a central server;
  • Edge servers include:
  • the receiving unit is configured to receive a resource search request broadcast by a preset edge server.
  • the resource search request carries information about computing resources required to complete the task;
  • the resource search request is the resource search sent by the task offloading device received by the preset edge server After the request, it is broadcast when it is judged that it does not meet the requirements of the above computing resources;
  • the above preset edge server is a preset that does not meet the requirements of the above computing resources within the space covered by the wireless access point accessed by the terminal In the case of a central server or an edge server, a server used to find other servers that meet the requirements to perform the above tasks;
  • the above task offloading device is deployed in the wireless access point accessed by the above terminal;
  • the above first edge server includes the above system Servers other than the preset edge server;
  • a determining unit configured to determine, according to the resource search request, that the first edge server meets the requirements of the computing resources
  • the sending unit is configured to send second response information in response to the resource search request to the task offloading device, where the second response information includes address information and location information of the first edge server.
  • the aforementioned edge server further includes a computing unit
  • the receiving unit is further configured to receive the location information of the terminal broadcast by the preset edge server before the determining unit determines that the first edge server meets the requirements of the computing resources according to the resource search request;
  • the calculation unit is configured to calculate the distance between the first edge server and the terminal according to the location information of the terminal after the determining unit determines that the first edge server meets the requirements of the computing resources according to the resource search request;
  • the sending unit is further configured to send third response information in response to the resource search request to the task offloading device, where the third response information includes the address information of the first edge server and the relationship between the first edge server and the terminal. The distance between.
  • the sending unit is further configured to send a response to the resource search to the task offloading device after the determining unit determines that the first edge server meets the demand of the computing resource according to the resource search request.
  • the requested fourth response information where the fourth response information includes the address information of the first edge server.
  • an embodiment of the present invention provides a mobile edge computing task offloading device.
  • the server may include a memory, a processor coupled with the memory, and a communication interface.
  • the communication interface may be used to support the server to execute the steps of sending and receiving information in the method for finding an edge computing server provided in the first aspect.
  • the communication interface may include a transmitter and a receiver.
  • the processor is configured to support the task offloading device to perform other processing steps in the edge computing server search method provided in the first aspect above except for sending information and receiving information.
  • the memory is used to store the computer program for implementing the edge computing server search method described in the first aspect
  • the processor is used to execute the program stored in the memory, that is, to perform the search for the edge computing server provided in the first aspect. Method, or any one of the possible implementation manners provided in the first aspect.
  • the memory and the processor can be integrated together or coupled through a coupler.
  • an embodiment of the present invention provides an edge server.
  • the server may include a memory, a processor coupled with the memory, and a communication interface.
  • the communication interface may be used to support the edge server to execute the steps of sending and receiving information in the method for finding an edge computing server provided in the second aspect.
  • the communication interface may include a transmitter and a receiver.
  • the processor is configured to support the edge server to perform other processing steps in the edge computing server search method provided in the second aspect above except for sending information and receiving information.
  • the memory is used to store the computer program for implementing the edge computing server search method described in the second aspect
  • the processor is used to execute the program stored in the memory, that is, perform the search for the edge computing server provided in the second aspect.
  • the memory and the processor can be integrated together or coupled through a coupler.
  • an embodiment of the present invention provides an edge server.
  • the server may include a memory, a processor coupled with the memory, and a communication interface.
  • the communication interface may be used to support the edge server to execute the steps of sending and receiving information in the method for finding an edge computing server provided in the third aspect.
  • the communication interface may include a transmitter and a receiver.
  • the processor is used to support the edge server to perform other processing steps except for sending information and receiving information in the method for finding an edge computing server provided in the third aspect.
  • the memory is used to store the computer program for implementing the edge computing server search method described in the third aspect
  • the processor is used to execute the program stored in the memory, that is, perform the search for the edge computing server provided in the third aspect. Method, or any one of the possible implementation manners provided in the third aspect.
  • the memory and the processor can be integrated together or coupled through a coupler.
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method of any one of the above-mentioned first aspects .
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any one of the foregoing second aspect method.
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any one of the foregoing third aspect method.
  • an embodiment of the present invention provides a computer program product, which when running on a computer, causes the computer to execute the method described in any one of the first aspects.
  • an embodiment of the present invention provides a computer program product, which when it runs on a computer, causes the computer to execute the method described in any one of the above-mentioned second aspects.
  • the embodiments of the present invention provide a computer program product, which when running on a computer, causes the computer to execute the method described in any one of the foregoing third aspects.
  • the edge cloud only implements the task execution architecture.
  • the present invention is implemented by the task offloading device deployed in the wireless access point.
  • Resource and data scheduling that is, the nearest server that meets the needs of the terminal is selected to calculate and complete the tasks that the terminal needs to calculate, which achieves the lowest delay in responding to the needs of the terminal and saves network overhead.
  • FIG. 1 is a schematic diagram of a system architecture of a method for finding an edge computing server provided by an embodiment of the solution;
  • FIG. 2 is a schematic diagram of the interaction flow of the method for finding an edge computing server provided by an embodiment of the solution
  • FIG. 3 is a schematic diagram of another interaction flow of the method for finding an edge computing server provided by an embodiment of the solution
  • FIG. 4 is a schematic diagram of another system architecture of a method for finding an edge computing server provided by an embodiment of the solution
  • Figure 5 is a schematic diagram of the logical structure of a task offloading device provided by an embodiment of the solution
  • FIG. 6 is a schematic diagram of the logical structure of an edge server provided by an embodiment of the solution.
  • FIG. 7 is a schematic diagram of the logical structure of another edge server provided by an embodiment of the solution.
  • FIG. 8 is a schematic diagram of the hardware structure of a task offloading device provided by an embodiment of the solution.
  • FIG. 9 is a schematic diagram of the hardware structure of an edge server provided by an embodiment of the solution.
  • FIG. 10 is a schematic diagram of the hardware structure of another edge server provided by an embodiment of the solution.
  • FIG. 1 is a schematic diagram of a system architecture to which the method for finding an edge computing server according to an embodiment of the present invention is applicable.
  • the system architecture may include four layers, where the first layer is the terminal layer, the second layer is the wireless access layer, the third layer is the edge cloud layer, and the fourth layer is the central cloud layer.
  • the terminal layer includes one or more terminals 101; the wireless access layer includes one or more wireless access points 102, and one or more task offloading devices 103, that is, the task offloading device 103 is deployed at the wireless access layer, and each wireless The access point may correspondingly deploy one or more task offloading devices 103; the edge cloud layer includes one or more edge servers 104; the central cloud layer includes one or more central servers 105 (multiple servers may form a server cluster). It should be noted that the edge server 104 may be an edge cloud server, and the central server 105 may be a central cloud server.
  • the terminal 101 may install and run related application software or APP, and the related application software or APP is an application software or APP suitable for the user terminal and adapted to the relevant needs of the user.
  • Application software is a collection of computer programs and instructions that provide services to customers.
  • the service may include, but is not limited to: sending a resource search request to the wireless access point 102 and receiving an Internet Protocol (IP) address or other address (such as a media storage control address (cedia access) of a server that can complete the corresponding task. control address, MAC)) and so on.
  • IP Internet Protocol
  • MAC media storage control address
  • the terminal 101 can include any kind of handheld electronic product based on a smart operating system, which can interact with the user through input devices such as a keyboard, a virtual keyboard, a touch panel, a touch screen, and a voice control device, such as smart phones, tablet computers, Handheld computers, wearable electronic devices, etc.
  • the intelligent operating system includes, but is not limited to, any operating system that enriches device functions by providing various applications to the device, such as Android, IOS, Windows, and MAC.
  • the terminal 101 may also include artificial intelligence devices such as vehicle-mounted equipment, vehicles, and robots.
  • the wireless access point 102 is the access point for users using the terminal 101 to enter the wired network. It is mainly used for broadband homes, buildings, campuses, parks, warehouses, factories and other places that require wireless monitoring. Typical distances cover tens of meters. Up to hundreds of meters, some can be used for long-distance transmission, and the farthest one can reach about 30KM.
  • the main technology is IEEE802.11 series. Most wireless access points also have an access point client mode, which can connect wirelessly with other access points to extend the network coverage.
  • the wireless access point 102 may include a routing switch access integrated device and a pure access point device.
  • the integrated device performs access and routing work.
  • the pure access device is only responsible for the access of the wireless terminal, and the pure access device is usually used as a wireless network extension Use, connect with other access points or the main access point to expand the wireless coverage, and the integrated device is generally the core of the wireless network.
  • each wireless access point 102 is covered by one or more edge servers 104, and may also be covered by one or more central servers 105.
  • the specific coverage is determined according to actual conditions. This program does not impose restrictions on this.
  • the wireless access point 102 can be used to receive a resource search request sent by the terminal 101 through a wireless network such as 2G, 3G, 4G, or 5G, and then send the request to the wireless access point deployed
  • the task offloading device 103 on the side of 102, the task offloading device 103 processes the request (for example, according to the request, it is determined whether there is a server that meets the computing resource requirements in the coverage area of the wireless access point 102, and the task of the terminal is sent to Execute on the server that meets the demand, etc.).
  • Mobile edge computing can use the wireless access network to provide nearby user information technology (IT) services and cloud computing functions, creating a carrier-grade service environment with high performance, low latency and high bandwidth, and accelerating the network
  • IT information technology
  • cloud computing functions creating a carrier-grade service environment with high performance, low latency and high bandwidth, and accelerating the network
  • the fast download of various content, services and applications allows consumers to enjoy an uninterrupted high-quality network experience.
  • the edge server 104 is a server deployed on the edge of the central cloud, and is used to provide various required computing services for user terminals nearby. For example, if the central cloud is deployed in Shanghai and user terminals are scattered in areas such as Guangzhou, Shenzhen, and Beijing, then edge servers can be deployed in Guangzhou, Shenzhen, and Beijing, so that local user terminals can be accessed from nearby edge servers. Obtain the required services and computing functions without having to seek a response from the central server over a long distance, thereby saving network resources and improving response efficiency.
  • the central server 105 is mainly used to provide necessary response when the edge server cannot respond to the request of the terminal.
  • the task offloading device 103 may be a mobile edge computing (mobile edge computing, MEC) server or may be another device that conforms to the same communication protocol standard as the MEC server and can implement related functions.
  • MEC mobile edge computing
  • a service uninstallation agent is deployed on the task uninstallation device, the service uninstallation agent may be a software or an application, and the task uninstallation device runs the service uninstallation agent to implement the operations performed by the task uninstallation device in the embodiment of this application. .
  • task offloading devices are usually general-purpose and small-scale servers, they provide limited computing resources and cannot handle a large number of virtual reality (VR) scenarios, augmented reality (AR) scenarios, or Computing for heavy scenes in cloud games.
  • the edge server 104 in the embodiment of the present application is a server that is deployed nearby the terminal, and has sufficient computing resources to complete calculations required for scenarios such as VR, AR, and cloud games.
  • system architecture to which the method for finding edge computing servers provided in the embodiments of the present application is applicable is not limited to the system architecture shown in FIG. 1.
  • the following describes a method for finding edge computing servers, which can be applied to the system architecture shown in Figure 1.
  • the method includes but is not limited to the following steps:
  • Step 201 The terminal sends a resource search request to the task offloading device.
  • the request carries information about computing resources required to complete the task.
  • Step 202 The task offloading device receives the resource search request sent by the terminal.
  • the task offloading device is deployed in a wireless access point accessed by the terminal.
  • the task offloading device can communicate with the wireless access point. Specifically, after the wireless access point receives the resource search request, the task offloading device obtains the request from the wireless access point.
  • the wireless access point can obtain information such as the resource availability of edge servers and/or central servers within its coverage area.
  • the task offloading device can obtain this information through the wireless access point, for example, through network slicing. Technology to obtain this information.
  • the resource idle condition of the edge server and/or the central server may include the memory idle condition of the processor, the idle condition of graphics processing unit (GPU) resources, the idle condition of hard disk space, and input/output (Input/Output). /Output, I/O) interface or equipment idle status and other information.
  • the foregoing task may be a task of performing image processing such as rendering, encoding, etc.
  • the foregoing resource search request may be, for example, a request for finding computing resources to perform the image processing task.
  • the computing resource may include information such as processor memory, GPU, hard disk, and I/O.
  • the computing resource information carried in the resource search request may include the memory space required to complete the task, the type and/or quantity of GPU, the size of hard disk space, I/O, and time.
  • the computing resource information may be information such as the memory of the processor in the server performing the task is not less than 1G, 2 GPUs are required, the size of the hard disk space is not less than 4G, and it takes 1 second to complete. This is only to facilitate the understanding of the exemplary description, and the specific implementation is not limited to the description herein.
  • the foregoing terminal may be the terminal 101 shown in FIG. 1
  • the foregoing wireless access point may be the wireless access point 102 shown in FIG. 1
  • the foregoing task offloading device may be the task offloading device 103 shown in FIG.
  • the foregoing edge server may be the edge server 104 shown in FIG. 1.
  • Step 203 The task offloading device determines, according to the resource search request, that there is a central server and/or at least one edge server that meets the requirements of the computing resources within a preset range.
  • Step 204 The above-mentioned task offloading device selects a server for executing the task from a central server and/or at least one edge server that meets the requirements of the computing resources.
  • the foregoing preset range includes a spatial range covered by the foregoing wireless access point, and the spatial range may include an edge server and/or a central server.
  • the central server here may be the central server 105 described in FIG. 1.
  • the task offloading device After the task offloading device receives the above request, it parses the request, obtains the information of the computing resource required to complete the above task included in the request, and then compares the required computing resource information with the wireless access point. Information such as the resource availability of edge servers and/or central servers in the coverage area is compared one by one to determine whether there is a server that meets the above task requirements.
  • the task offloading device selects the one that meets the requirements as the server The server that performs the above tasks. Then, the task offloading device sends the above resource search request to the server that meets the demand. After the server that satisfies the demand receives the request, it sends the address information of the server to the task offloading device according to the request.
  • the address information includes address information such as an Internet Protocol IP address or a media storage control address (cedia access control address, MAC) of the server that meets the requirements, which can route the message to the server.
  • address information may be used to instruct the task offloading device to send the task to the server corresponding to the address information for execution.
  • the task offloading device will select one server from the multiple servers that meet the requirements as the server for executing the task.
  • the following describes how the task offloading device selects a server from the above multiple servers that meet the requirements as the server to perform the above tasks: the task offloading device synchronously sends the resource search request to the multiple servers that meet the needs; the task offloading device receives The address information sent by the multiple servers that meet the requirements; the task offloading device selects the server that executes the task according to the order of receiving the address information from the multiple servers that meet the requirements.
  • the above-mentioned task offloading device will respectively send the above-mentioned resource search request to the above-mentioned multiple demand-satisfied servers at the same time.
  • the multiple demand-satisfied servers After the multiple demand-satisfied servers receive the request, they respectively send to the task offloading device according to the request.
  • Respective address information and then the task unloading device can select the server corresponding to the address information received first as the server to perform the above task.
  • the task offloading device may also select the server corresponding to the received second or third address information as the server for executing the above-mentioned task.
  • the specific selected server is determined according to the specific situation, and this solution does not limit this.
  • the foregoing task offloading device selects a server for executing the task from the central server and/or at least one edge server that meets the requirements of the computing resource. It can be implemented in one of the following ways:
  • the server that meets the demand sends the address information and location information of the server to the task offloading device according to the resource search request.
  • the location information may be actual location information.
  • the location information may be information about the latitude and longitude of the location where the server that meets the demand is located, or it may be the location mark information of the server that meets the demand that is preset by the system.
  • the task offloading device calculates the distance between the server that meets the demand and the terminal according to the received location information of the server that meets the demand and the location information of the terminal, and then selects the server that is closest to the terminal among the servers that meet the demand. As a server to perform the above tasks.
  • the foregoing task offloading device selects a server for executing the task from the central server and/or at least one edge server that meets the requirements of the computing resource. It can be implemented in one of the following ways:
  • the task offloading device when the task offloading device sends the resource search request to the server that meets the demand, it also sends the location information of the terminal to the server that meets the demand.
  • the task offloading device may send a request in which the location information of the terminal is added to the resource search request to the server that meets the demand.
  • the location information of the terminal may be actual location information.
  • the location information may be information about the latitude and longitude of the location where the terminal is located, or may also be information about a location mark preset by the system.
  • the wireless access point accessed by the terminal may obtain the location information of the terminal, and then the task offloading device may obtain the location information through the wireless access point, for example, obtain the location information through network slicing technology .
  • the task offloading device selects the one that meets the requirement as the server that performs the above task. It is not necessary to send the above-mentioned terminal location information to the above-mentioned one server that meets the requirements, and then for the specific operation, refer to the corresponding description in step 204 in FIG. 2, which will not be repeated here.
  • the task offloading device will select one server from the multiple servers that meet the requirements as the server for executing the task.
  • the task offloading device will send the resource search request and the location information of the terminal to the multiple servers that meet the requirements. After the multiple servers that meet the requirements receive the request and the location information of the terminal, they respectively The location information of the terminal calculates the distance between the server and the terminal, and then sends the address information of the server and the distance between the server and the terminal to the task offloading device. After the task offloading device receives the information, the task offloading device selects the server closest to the terminal as the server for performing the above-mentioned tasks.
  • the task offloading device sends response information to the above resource search request sent by the terminal to the terminal.
  • the above response information is referred to as first response information
  • the first response information includes instruction information for instructing the terminal to send the above-mentioned task that needs to be calculated to the task uninstallation device.
  • the task offloading device sends the above-mentioned first response information to the terminal.
  • the terminal parses the information to obtain the instruction information, and then sends the above-mentioned task to be performed to the task according to the instruction information. Uninstall the device.
  • the task uninstallation device sends the task to the server selected above, and the server executes the task, and sends the execution result to the task uninstallation device, and the task uninstallation device sends the execution result to the terminal.
  • the indication information in the first response message may be the address information of the server selected to perform the task, or other predetermined information.
  • the task offloading device sends the address information of the server that executes the task selected above to the terminal.
  • the terminal After receiving the address information, the terminal directly sends the task to be executed to the selected server based on the address information.
  • the server that executes the above task will execute the task by the server and send the execution result directly to the terminal.
  • the foregoing method embodiment describes the case where there are servers that meet the requirements of the foregoing computing resources within the foregoing preset range.
  • the following describes an embodiment of the case where there are no servers that meet the foregoing requirements of computing resources within the foregoing preset range, see FIG. 3.
  • Step 301 The terminal sends a resource search request to the task offloading device.
  • the request carries information about computing resources required to complete the task.
  • Step 302 The task offloading device receives the resource search request sent by the terminal.
  • step 301 and step 302 correspond to the same as step 201 and step 202 described in FIG. 2 above, and the specific implementation manner can be referred to the corresponding description above, which will not be repeated here.
  • Step 303 The task offloading device determines, according to the resource search request, that no server within the preset range meets the requirements of the computing resources.
  • Step 304 The task offloading device sends the resource search request to the preset edge server.
  • the foregoing preset edge cloud server is a preset central cloud server or edge cloud server that does not meet the requirements of the computing resources within the space covered by the wireless access point accessed by the terminal. To find other cloud servers that meet the needs to perform the tasks.
  • the foregoing preset edge server may be an edge server within the coverage area of the foregoing wireless access point, or may be an edge server within the coverage area of the wireless access point other than the wireless access point accessed by the terminal in the same network. Edge server.
  • Step 305 The preset edge server determines, according to the resource search request, that the preset edge server does not meet the requirements of the computing resources.
  • Step 306 The preset edge server broadcasts the resource search request to other edge servers and/or central servers.
  • edge servers and/or central servers include edge servers and/or central servers in the network where the wireless access point is located except for the preset edge server.
  • the edge server after receiving the resource search request sent by the task offloading device, the edge server parses the request, obtains information about the computing resource in the request, and then compares the information about the computing resource with the server's own computing resource. Compare them one by one to determine whether they meet the requirements of computing resources required by the above tasks. In the case where it is determined that it is not satisfied, the aforementioned preset edge server may broadcast the aforementioned resource search request to the aforementioned other edge servers and/or central servers.
  • the foregoing preset edge server and the foregoing other edge servers and/or central servers may be networked in a form of full mesh.
  • edge servers and/or central servers After the above-mentioned other edge servers and/or central servers receive the broadcast resource search request, they parse the resource search request to obtain the information of the computing resources required to complete the above tasks in the request, and then compare the computing resources with the computing resources in the respective servers. The resources are compared one by one to determine whether the computing resources required by the above tasks are met.
  • the other edge servers and/or central servers meet the requirements, and the other edge servers and/or central servers include the above-mentioned central server, then all servers that do not meet the requirements are sent to the above-mentioned preset edge server Response information that does not meet the requirements, and then the above-mentioned preset edge server sends these response information to the above-mentioned task offloading device, and the task offloading device determines that the resource search failed this time according to the response information, and sends the response that the resource search failed to the above-mentioned terminal.
  • the other edge servers and/or central servers if none of the other edge servers and/or central servers meets the requirements, and the other edge servers and/or central servers do not include the aforementioned central server, then all servers that do not meet the requirements will be sent to the aforementioned preset edge server.
  • Send response information that does not meet the demand, and then the preset edge server sends the resource search request to the central server according to the response information, and the central server searches for a server that meets the demand according to the request.
  • the server that meets the demand sends its own address information to the aforementioned preset edge server, and then presets the address that the edge server will receive The information is sent to the above task uninstall device.
  • the task offloading device determines the server that meets the requirements as the server that performs the task. For the specific implementation after determining the server to perform the above task, refer to the two processing methods listed in the "after the task uninstallation device selects the server to perform the above task", which will not be repeated here.
  • the task offloading device may select a server for performing the above-mentioned task according to the sequence of the received address information of the plurality of servers that meet the requirements. For example, the server corresponding to the address information received first may be selected as the server for performing the above-mentioned task.
  • the task offloading device may also select the server corresponding to the received second or third address information as the server for executing the above-mentioned task.
  • the specific selected server is determined according to the specific situation, and this solution does not limit this. For the specific implementation after selecting the server to perform the above task, refer to the two processing methods listed in "After the task uninstallation device selects the server to perform the above task", the details are not repeated here.
  • the method further includes: the edge server determines the preset edge server according to the resource search request Meet the requirements of the aforementioned computing resources.
  • the foregoing preset edge server sends its own address information to the task offloading device, and after receiving the address information of the preset edge server, the task offloading device determines that the preset edge server is a server that performs the foregoing task.
  • the task offloading device determines that the preset edge server is a server that performs the foregoing task.
  • the foregoing preset edge server broadcasts the foregoing resource search request to other edge servers and/or central servers
  • the following implementation manners may also be included:
  • the other edge servers and/or the central server above After the other edge servers and/or the central server above receive the resource search request, they first parse the request to obtain the computing resource information required to complete the task in the request, and then compare the computing resource with the information in the respective server. The computing resources are compared one by one to determine whether the computing resources required by the above tasks are met.
  • the other edge server and/or the server that meets the demand in the central server sends its own location information and location information to the aforementioned preset edge server according to the request. Address information. Then, the preset edge server sends the received address information and location information to the task offloading device.
  • the task offloading device determines the server that meets the requirements as the server that performs the task. For the specific implementation after determining the server to perform the above task, refer to the two processing methods listed in the "after the task uninstallation device selects the server to perform the above task", which will not be repeated here.
  • the task offloading device first calculates the distance between the terminal and the multiple servers that meet the needs based on the received location information of the multiple servers that meet the requirements and the location information of the above-mentioned terminal, and then the task offloading device selects The server that is closest to the terminal among the plurality of servers that meet the requirements serves as the server that performs the above-mentioned tasks.
  • the server that is closest to the terminal among the plurality of servers that meet the requirements serves as the server that performs the above-mentioned tasks.
  • the task offloading device determines according to the resource search request that no server within the preset range meets the demand for computing resources", in “step 305, the edge server according to the resource Before searching for a request to determine that the aforementioned preset edge server does not meet the requirements of the aforementioned computing resources", it also includes: the aforementioned task offloading device also sends the location information of the aforementioned terminal to the preset edge server.
  • the edge server determines that the preset edge server does not meet the requirements of the computing resources according to the resource search request", in addition to broadcasting the resource search request to other edge servers and/or centers
  • the location information of the aforementioned terminal is also broadcast to the other edge server and/or central server.
  • the server that meets the demand calculates the distance between the server and the terminal according to the received position information of the terminal, and then calculates the distance between the server and the terminal.
  • the obtained distance and its own address information are sent to the aforementioned preset edge server.
  • the preset edge server sends the received distance and address information to the task offloading device.
  • the task offloading device determines the server that meets the requirements as the server that performs the task. For the specific implementation after determining the server to perform the above task, refer to the two processing methods listed in the "after the task uninstallation device selects the server to perform the above task", which will not be repeated here.
  • the task offloading device first selects the server that is the closest to the terminal among the plurality of servers that meet the requirements according to the received distances between the plurality of servers that meet the requirements and the terminal as the server to perform the above tasks.
  • the server to perform the above task refer to the two processing methods listed in "After the task uninstallation device selects the server to perform the above task", the details are not repeated here.
  • the task offloading device may not first determine whether there is a server within the preset range that meets the computing resource demand, but directly sends the resource search request Send to the preset edge server.
  • the preset edge server can determine whether it meets the requirement according to the resource search request, and then sends its own address information to the task offloading device if it is satisfied, and broadcasts the resource search request to the other edge servers and/or the central server if it is not satisfied. Then, for the specific implementation after the broadcast, refer to the corresponding description in the foregoing method embodiment, which is not repeated here.
  • the task offloading device may first send the resource search request without determining whether there is a server within the preset range that meets the demand for computing resources. Give the default edge server.
  • the preset edge server may also not first determine whether it meets the demand according to the resource search request, but directly broadcast the resource search request to the other edge servers and/or the central server. Then, for the specific implementation after the broadcast, refer to the corresponding description in the foregoing method embodiment, which is not repeated here.
  • FIG. 4 exemplarily shows a schematic diagram of the architecture of a terminal and an edge server in the coverage area of a wireless access point in the system described in FIG. 1.
  • a task offloading device 402 is deployed in a wireless access point (one or more task offloading devices 402 can be deployed), and a service offloading agent is deployed on the task offloading device 402.
  • One or more terminals 401 are included in the coverage area of the wireless access point, and a task calculation manager is deployed in the terminal 401.
  • One or more edge servers 403 are included in the coverage area of the wireless access point, and task receivers are deployed in the edge servers 403.
  • the edge server 403 can communicate with the central server 404, and a task coordinator is deployed in the central server 404.
  • the service offloading agent in the task offloading device 402 can obtain the information of the edge server 403 within the preset range through the wireless access point, and provide local scheduling capabilities to quickly find a server that meets the needs of the terminal to respond to the request of the terminal; When there is no server within the preset range to meet the terminal requirements, the request is further encapsulated and sent to the task receiver of the preset edge server to perform a larger range of search operations.
  • the preset edge server may be the same as the preset edge server in the foregoing method embodiment.
  • the preset range is the spatial range covered by the wireless access point.
  • the task receiver in the edge server 403 can provide the wireless access point with an application program interface (application programming interface, API) for receiving task requests downwards, and can provide sending requests broadcast to other edge servers or sent to the center side.
  • API application programming interface
  • the task coordinator of the central server 404 mainly provides its API to the task receiver of the edge server.
  • the task coordinator can perform global scheduling.
  • the terminal 401 may first send the request to the computing task management in the terminal 401 In the device, the computing task manager calculates whether the terminal 401 has resources to complete the task.
  • the terminal 401 will send the resource search request to the wireless access point, and then the task offloading device 402 obtains the request from the wireless access point based on the network slicing technology, and transfers the request to the wireless access point.
  • the request is sent to the business offloading agent.
  • the service offloading agent uses the capability API of the task offloading device, such as Traffic offload, to register the service type or traffic slice, and the task offloading device demodulates the specified flow direction and sends it to the service offloading agent on the task offloading device.
  • the service offloading agent After the service offloading agent obtains the above request, it parses the request and obtains the information of the computing resources required to complete the above task in the request. For example, it may include the memory availability of the processor, the graphics processing unit (GPU) ) Information about the availability of resources, the availability of hard disk space, and the availability of input/output (Input/Output, I/O) interfaces or equipment. At the same time, the service offloading agent obtains the idle state of the computing resources of the edge server 403 within the preset range through the above-mentioned wireless access point.
  • the service offloading agent obtains the idle state of the computing resources of the edge server 403 within the preset range through the above-mentioned wireless access point.
  • the service offloading agent compares the computing resources obtained from the request with the idle computing resources of the edge server 403 within a preset range one by one to determine whether there is a server that meets the demand for computing resources in the request.
  • the service offloading agent repackages the above request and sends it to the task receiver in the edge server 403 that meets the requirements.
  • the task receiver parses the request, and then sends the request to the above task according to the analysis result.
  • the uninstallation device 402 sends the address information of the server. After the task offloading device 402 receives the address information, it sends the address information to the service offloading agent of the server, and the service offloading agent performs subsequent processing.
  • the operations processed by the task offloading device in the foregoing method embodiment can be processed by the service offloading agent in the task offloading device.
  • the subsequent processing please refer to the corresponding description in the foregoing method embodiment, which will not be omitted here. Repeat.
  • the service offloading agent repackages the above request and sends it to the task receiver of the preset edge server, and the task receiver performs subsequent processing.
  • the operations processed by the preset edge server in the foregoing method embodiment can be processed by the task receiver in the preset edge server.
  • the subsequent processing refer to the corresponding description in the foregoing method embodiment. No longer.
  • the service offloading agent repackages the request and sends it to the task receiver of the preset edge server, including adding the location information of the terminal 401 to the request. After being encapsulated, it is sent to the task receiver of the preset edge server, and then the task receiver performs subsequent processing.
  • the operations processed by the preset edge server in the foregoing method embodiment can be processed by the task receiver in the preset edge server.
  • I won’t repeat it here.
  • each network element such as a terminal, a task offloading device, and a server, includes a hardware structure and/or software module corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed by hardware or computer software-driven hardware depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiments of the present application can divide functional modules into terminals, task offloading devices, servers, etc. according to the above method examples.
  • each functional module can be divided corresponding to each function, or two or more functions can be integrated into one module. in.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • FIG. 5 shows a schematic diagram of a possible logical structure of the task offloading device involved in any of the foregoing method embodiments, and the task offloading device 500 includes:
  • the receiving unit 501 is configured to receive a resource search request sent by the aforementioned terminal.
  • the aforementioned resource search request carries information about computing resources needed to complete the task.
  • the task offloading device 500 is deployed in the wireless access point accessed by the aforementioned terminal;
  • the determining unit 502 is configured to determine, according to the foregoing resource search request, that there is a central server and/or at least one edge server that meets the requirements of the foregoing computing resources in the space covered by the wireless access point accessed by the foregoing terminal;
  • the selecting unit 503 is configured to select a server for performing the above-mentioned task from a central server and/or at least one edge server that meets the requirements of the above-mentioned computing resources.
  • the selection unit 503 is specifically configured to:
  • a server for performing the above task is selected from a central server and/or at least one edge server that meets the requirements of the computing resource.
  • the task offloading device 500 further includes a sending unit, configured to select a central server and/or at least one edge server that meets the requirements of the computing resources according to the sequence of the received address information by the selecting unit 503 After selecting the server for performing the above task, send the first response information in response to the above resource search request to the above terminal;
  • the receiving unit 501 is further configured to receive the task sent by the terminal according to the first response information
  • the sending unit is further configured to send the task to the selected server.
  • the task offloading device 500 further includes a sending unit and a computing unit;
  • the determining unit 502 is further configured to determine, according to the resource search request, that there is no central server or edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the sending unit is configured to send the resource search request to the preset edge server
  • the receiving unit 501 is further configured to receive second response information sent by the above-mentioned preset edge server, where the above-mentioned second response information is the server that has received the above-mentioned resource search request broadcast by the above-mentioned preset edge server and meets the requirements of the above-mentioned computing resources.
  • the calculation unit is configured to calculate the distance between the server that meets the computing resource demand and the terminal based on the location information of the server that meets the computing resource demand and the location information of the terminal;
  • the selection unit 503 is further configured to select the server closest to the terminal from the servers that meet the requirements of the computing resources according to the distance obtained by the foregoing calculation to perform the foregoing tasks.
  • the task offloading device 500 further includes a sending unit
  • the determining unit 502 is further configured to determine, according to the resource search request, that there is no central server or edge server that meets the requirements of the computing resources in the space covered by the wireless access point accessed by the terminal;
  • the sending unit is configured to send the resource search request and the location information of the terminal to a preset edge server;
  • the receiving unit 501 is further configured to receive third response information sent by the preset edge server, where the third response information is the server that receives the resource search request and the location information of the terminal broadcasted by the preset edge server.
  • the server that meets the computing resource requirements sends the third response information to the preset edge server, and the third response information includes the address information of the server that meets the computing resource requirements, and the relationship between the server that meets the computing resource requirements and the terminal.
  • the selecting unit 503 is further configured to select the server closest to the terminal from the servers that meet the requirements of the computing resources according to the distance between the server that meets the requirements of the computing resources and the terminal to perform the above tasks.
  • FIG. 6 shows a schematic diagram of a possible logical structure of a preset edge server involved in any of the foregoing method embodiments, and the edge server 600 includes:
  • the receiving unit 601 is configured to receive a resource search request sent by the task offloading device, the resource search request carries information about computing resources required to complete the task, and the edge server 600 covers the preset wireless access point accessed by the terminal In the case that there is no central server or edge server that meets the requirements of the above computing resources in the space range of, it is used to find other servers that meet the requirements to perform the above tasks;
  • the determining unit 602 is configured to determine, according to the foregoing resource search request, that the edge server 600 does not meet the foregoing computing resource requirements;
  • the broadcasting unit 603 is configured to broadcast the above resource search request to other edge servers and/or central servers.
  • the edge server 600 further includes a sending unit
  • the receiving unit 601 is further configured to, after the broadcasting unit 603 broadcasts the resource search request to the other edge servers and/or the central server, receive the information sent by the server that meets the requirements of the computing resources among the other edge servers and/or central servers. Second response information, where the second response information includes address information and location information of the server that meets the requirements of the computing resources;
  • the sending unit is configured to send the second response information to the task offloading device.
  • the receiving unit 601 is further configured to receive the position of the terminal sent by the task offloading device before the determining unit 602 determines that the edge server 600 does not meet the requirements of the computing resources according to the resource search request. information;
  • the broadcasting unit 603 is further configured to broadcast the location information of the terminal to the other edge servers and/or central servers after the determining unit 602 determines that the edge server 600 does not meet the requirements of the computing resources according to the resource search request.
  • the receiving unit 601 is further configured to receive the other edge server after the broadcast unit 603 broadcasts the resource search request and the location information of the terminal to the other edge servers and/or central servers. And/or the third response information sent by the server that meets the requirements of the computing resources in the central server, the third response information includes the address information of the servers that meet the requirements of the computing resources, and the information that meets the requirements of the computing resources.
  • the sending unit is configured to send the third response information to the task offloading device.
  • the receiving unit 601 is further configured to, after the broadcasting unit 603 broadcasts the above-mentioned resource search request to other edge servers and/or central servers, receive the above-mentioned other edge servers and/or central servers.
  • the sending unit is configured to send the fourth response information to the task offloading device.
  • the edge server 600 further includes a sending unit
  • the determining unit 602 is further configured to, after the receiving unit 601 receives the resource search request sent by the task offloading device, determine, according to the resource search request, that the edge server 600 meets the requirements of the aforementioned computing resources;
  • the sending unit is configured to send the address information of the edge server 600 to the task offloading device.
  • FIG. 7 shows a possible logical structure of any one of the other edge servers and/or central servers involved in any of the foregoing method embodiments.
  • the server 700 includes:
  • the receiving unit 701 is configured to receive a resource search request broadcast by a preset edge server.
  • the resource search request carries information about computing resources needed to complete the task;
  • the resource search request is the resource sent by the task offloading device received by the preset edge server After searching for the request, it is broadcast when it is judged that it does not meet the above computing resource requirements;
  • the above preset edge server is preset to fail to meet the above computing resource requirements within the space covered by the wireless access point accessed by the terminal In the case of a central server or an edge server, the server is used to find other servers that meet the requirements to perform the above tasks;
  • the above task offloading device is deployed in the wireless access point accessed by the above terminal;
  • the above first edge server includes the above system Servers other than the preset edge server in
  • the determining unit 702 is configured to determine, according to the resource search request, that the first edge server meets the requirements of the computing resources;
  • the sending unit 703 is configured to send second response information in response to the resource search request to the task offloading device, where the second response information includes address information and location information of the first edge server.
  • the server 700 further includes a computing unit
  • the receiving unit 701 is further configured to receive the location information of the terminal broadcast by the preset edge server before the determining unit 702 determines that the first edge server meets the requirements of the computing resources according to the resource search request;
  • the calculation unit is configured to calculate the distance between the first edge server and the terminal according to the location information of the terminal after the determining unit 702 determines that the first edge server meets the requirements of the computing resources according to the resource search request;
  • the sending unit 703 is further configured to send third response information in response to the resource search request to the task offloading device, where the third response information includes the address information of the first edge server and the relationship between the first edge server and the terminal. The distance between.
  • the sending unit 703 is further configured to send a response to the task offloading device to the task offloading device after the determining unit 702 determines that the first edge server meets the requirements of the computing resources according to the resource finding request.
  • the requested fourth response information where the fourth response information includes the address information of the first edge server.
  • FIG. 8 shows a schematic diagram of a possible hardware structure of a task offloading device involved in an embodiment of a method for finding an edge computing server provided by an embodiment of this application.
  • the task offloading device 800 includes a processor 801, a memory 802, and a communication interface 803.
  • the processor 801, the communication interface 803, and the memory 802 may be connected to each other or through a bus 804.
  • the memory 802 is used to store computer programs and data of the task offloading device 800.
  • the memory 802 may include but is not limited to random access memory (RAM) and read-only memory (ROM). , Erasable programmable read-only memory (erasable programmable read-only memory, EPROM) or portable read-only memory (compact disc read-only memory, CD-ROM), etc.
  • the communication interface 803 is used to support the task offloading device 800 to communicate, such as receiving or sending data.
  • the processor 801 may be a central processing unit, a general-purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
  • the processor may also be a combination of computing functions, for example, a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on.
  • the processor 801 may be configured to read programs stored in the memory 802, and execute operations implemented by the task offloading device in the methods described in FIGS. 2, 3, and possible implementation manners.
  • FIG. 9 shows a schematic diagram of a possible hardware structure of a preset edge server involved in the embodiment of the method for finding an edge computing server provided by an embodiment of this application.
  • the edge server 900 includes a processor 901, a memory 902, and a communication interface 903.
  • the processor 901, the communication interface 903, and the memory 902 may be connected to each other or through a bus 904.
  • the memory 902 is used to store computer programs and data of the edge server 900, and the memory 902 may include but is not limited to RAM, ROM, EPROM, or CD-ROM, etc.
  • the communication interface 903 is used to support the edge server 900 to communicate, such as receiving or sending data.
  • the processor 901 may be a central processing unit, a general-purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
  • the processor may also be a combination of computing functions, for example, a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on.
  • the processor 901 may be used to read the program stored in the memory 902, and execute the operations implemented by the preset edge server in the methods described in FIGS. 2, 3 and possible implementation manners.
  • FIG. 10 shows a schematic diagram of a possible hardware structure of any one of the other edge servers and/or the central server involved in the embodiment of the method for finding the edge computing server provided by the embodiment of this application.
  • the server 1000 includes a processor 1001, a memory 1002, and a communication interface 1003.
  • the processor 1001, the communication interface 1003, and the memory 1002 may be connected to each other or through a bus 1004.
  • the memory 1002 is used to store computer programs and data of the server 1000, and the memory 1002 may include but is not limited to RAM, ROM, EPROM, or CD-ROM, etc.
  • the communication interface 1003 is used to support the server 1000 to communicate, such as receiving or sending data.
  • the processor 1001 may be a central processing unit, a general-purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.
  • the processor may also be a combination of computing functions, for example, a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on.
  • the processor 1001 may be used to read the program stored in the memory 1002, and execute operations implemented by any of the other edge servers and/or the central server in the methods described in FIGS. 2, 3, and possible implementations.
  • the embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the methods described in FIGS. 2, 3, and possible implementations.
  • the operations implemented by the medium task uninstallation device are not limited to:
  • the embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the methods described in FIGS. 2, 3, and possible implementations.
  • the operations implemented by the edge server are preset in the.
  • the embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the methods described in FIGS. 2, 3, and possible implementations.
  • the edge cloud only implements the task execution architecture.
  • the present invention is implemented by the task offloading device deployed in the wireless access point.
  • Resource and data scheduling that is, the nearest server that meets the needs of the terminal is selected to calculate and complete the tasks that the terminal needs to calculate, which achieves the lowest delay in responding to the needs of the terminal and saves network overhead.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本发明实施例提供一种边缘计算服务器的寻找方法及相关设备，该方法应用于包括终端、任务卸载设备、中心服务器以及多个边缘服务器的边缘计算系统；该方法包括：任务卸载设备接收终端发送的资源寻找请求，该资源寻找请求携带有完成任务需要的计算资源的信息，其中，该任务卸载设备部署在该终端接入的无线接入点中；然后，根据该资源寻找请求确定该终端接入的无线接入点覆盖的空间范围内有满足该计算资源的需求的服务器；再从满足该计算资源的需求的服务器中选择出用于执行该任务的服务器。采用本发明实施例，能够最低时延地响应终端的需求且节约网络开销。

Description

边缘计算服务器的寻找方法及相关设备 技术领域
本发明涉及移动边缘计算技术领域，尤其涉及边缘计算服务器的寻找方法及相关设备。
背景技术
现有的移动边缘计算架构中，主要是从中心云到边缘云再到终端的集中式资源配置和调度的架构，这种架构中所有的资源及调度数据都由中心云分发以及配置，边缘云只做任务执行。这就导致了决策效率低下、无法实现最低时延地响应终端的需求。此外，需要将终端的请求发送到中心云来做决策，再将决策返回，增加了网络开销，导致资源浪费。
综上所述，如何实现最低时延的响应终端的需求且节约网络开销是本领域技术人员急需解决的技术问题。
发明内容
本申请实施例公开了一种边缘计算服务器的寻找方法及相关设备，能够实现最低时延地响应终端的需求且节约网络开销。
第一方面，本申请实施例公开了一种边缘计算服务器的寻找方法，上述方法应用于边缘计算系统，上述边缘计算系统包括终端、任务卸载设备、中心服务器以及多个边缘服务器；上述方法包括：
上述任务卸载设备接收上述终端发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，其中，上述任务卸载设备部署在上述终端接入的无线接入点中；
上述任务卸载设备根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内有满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
上述任务卸载设备从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器。
在具体实施例中上述任务卸载设备可以是移动边缘计算(mobile edge computing，MEC)服务器或者可以是其它与MEC服务器一样符合相同通信协议标准且可以实现相关功能的设备。
可选的，上述任务卸载设备上部署有业务卸载代理，该业务卸载代理可以是一个软件或应用程序，上述任务卸载设备运行该业务卸载代理可以实现本申请实施例中任务卸载设备所执行的操作。
在本申请实施例中，相比于现有技术的所有的资源及调度数据都由中心云分发以及配置，边缘云只做任务执行的架构，本发明通过部署在无线接入点中的任务卸载设备完成资源和数据的调度，即就近选择满足终端需求的服务器来计算完成终端需要计算的任务，实现了最低时延地响应终端的需求且节约网络开销。
在其中一种可能的实施方式中，上述任务卸载设备从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器，包括：
上述任务卸载设备将上述资源寻找请求发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
上述任务卸载设备接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息；
上述任务卸载设备根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行上述任务的服务器。
上述地址信息包括上述满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器的互联网协议IP地址或者媒体存储控制地址(cedia access control address，MAC)等可以将报文路由到服务器的地址信息。在本申请实施例中，该地址信息可以用于指示上述任务卸载设备将上述任务发送给上述地址信息对应的服务器上执行。
在其中一种可能的实施方式中，上述任务卸载设备从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器，包括：
上述任务卸载设备将上述资源寻找请求发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
上述任务卸载设备接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息和位置信息；
上述任务卸载设备根据接收的上述位置信息和上述终端的位置信息计算上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离；
上述任务卸载设备根据上述计算得到的距离，从上述满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在本申请实施例中，通过上述发送的请求可以指示接收到该请求的服务器中满足上述计算资源需求的服务器向任务卸载设备发送自身的位置信息，以使得能够寻找到与终端距离最近的满足需求的服务器来计算完成终端需要计算的任务，以进一步降低时延和网络开销。
在其中一种可能的实施方式中，上述任务卸载设备从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器，包括：
上述任务卸载设备将上述资源寻找请求和上述终端的位置信息发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
上述任务卸载设备接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息，以及满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离；
上述任务卸载设备根据满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离，从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在其中一种可能的实施方式中，上述任务卸载设备根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用 于执行上述任务的服务器之后，上述方法还包括：
上述任务卸载设备向上述终端发送响应上述资源寻找请求的第一响应信息；
上述任务卸载设备接收上述终端根据上述第一响应信息发送的上述任务；
上述任务卸载设备将上述任务发送给上述选择出的服务器。
在本申请实施例中，任务卸载设备为终端选择好执行任务的服务器之后，通过与终端交互将任务发送给选择的服务器执行，提高了任务卸载的效率。
在其中一种可能的实施方式中，上述任务卸载设备根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述任务卸载设备将上述资源寻找请求发送给预设边缘服务器；
上述任务卸载设备接收上述预设边缘服务器发送的第二响应信息，上述第二响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第二响应信息包括上述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述任务卸载设备根据上述满足上述计算资源的需求的服务器的位置信息和上述终端的位置信息，计算上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述任务卸载设备根据上述计算得到的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
本申请实施例中的预设边缘云服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心云服务器或边缘云服务器的情况下，用于寻找其它满足需求的云服务器来执行上述任务的服务器。
在本申请实施例中，在无线接入点覆盖的范围内没有满足终端需求的服务器，则任务卸载设备会将终端的请求发送给预设的边缘服务器以扩大寻找的范围。此外，该请求还可以用于指示满足需求的服务器向任务卸载设备发送自身的位置信息，以使得能够寻找到与终端距离最近的满足需求的服务器来计算完成终端需要计算的任务，以进一步降低时延和网络开销。
在其中一种可能的实施方式中，上述任务卸载设备根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述任务卸载设备将上述资源寻找请求和上述终端的位置信息发送给预设边缘服务器；
上述任务卸载设备接收上述预设边缘服务器发送的第三响应信息，上述第三响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求和上述终端的位置信息的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述任务卸载设备根据上述满足上述计算资源的需求的服务器与上述终端之间的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在本申请实施例中，任务卸载设备基于终端的位置信息为终端寻找距离最近的满足需求的服务器来计算完成终端需要计算的任务。
在其中一种可能的实施方式中，上述任务卸载设备根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述任务卸载设备将上述资源寻找请求发送给预设边缘服务器；
上述任务卸载设备接收上述预设边缘服务器发送的第四响应信息，上述第四响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第四响应信息包括上述满足上述计算资源的需求的服务器的地址信息；
上述任务卸载设备根据接收上述满足上述计算资源的需求的服务器的地址信息的先后顺序选择执行上述任务的服务器。
第二方面，本申请实施例公开了一种边缘计算服务器的寻找方法，上述方法应用于边缘计算系统，上述边缘计算系统包括终端、任务卸载设备、中心服务器以及多个边缘服务器，其中，上述任务卸载设备部署在上述终端接入的无线接入点中；上述方法包括：
预设边缘服务器接收上述任务卸载设备发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，上述预设边缘服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；
上述预设边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求；
上述预设边缘服务器将上述资源寻找请求广播给其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，上述预设边缘服务器将上述资源寻找请求广播给其它边缘服务器和/或中心服务器之后，还包括：
上述预设边缘服务器接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第二响应信息，上述第二响应信息包括上述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述预设边缘服务器将上述第二响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述预设边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求之前，还包括：
上述预设边缘服务器接收上述任务卸载设备发送的上述终端的位置信息；
上述预设边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求之后，还包括：
上述预设边缘服务器将上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，上述预设边缘服务器将上述资源寻找请求和上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器之后，还包括：
上述预设边缘服务器接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第三响应信息，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述预设边缘服务器将上述第三响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述预设边缘服务器将上述资源寻找请求广播给其它边缘服务器和/或中心服务器之后，还包括：
上述预设边缘服务器接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第四响应信息，上述第四响应信息包括上述满足上述计算资源的需求的服务器的地址信息；
上述预设边缘服务器将上述第四响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述预设边缘服务器接收上述任务卸载设备发送的资源寻找请求之后，还包括：
上述预设边缘服务器根据上述资源寻找请求确定上述预设边缘服务器满足上述计算资源的需求；
上述预设边缘服务器向上述任务卸载设备发送上述预设边缘服务器的地址信息。
第二方面描述的方法的有益效果可以参见上述第一方面对应的有益效果，此处不再赘述。
第三方面，本申请实施例公开了一种边缘计算服务器的寻找方法，该方法应用于边缘计算系统，上述边缘计算系统包括终端、任务卸载设备、中心服务器以及多个边缘服务器；上述方法包括：
第一边缘服务器接收预设边缘服务器广播的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息；上述资源寻找请求为上述预设边缘服务器接收上述任务卸载设备发送的资源寻找请求后，判断自身不满足上述计算资源的需求的情况下广播的；上述预设边缘服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；上述任务卸载设备部署在上述终端接入的无线接入点中；上述第一边缘服务器包括上述系统中除预设边缘服务器之外的服务器；
上述第一边缘服务器根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求；
上述第一边缘服务器向上述任务卸载设备发送响应上述资源寻找请求的第二响应信息，上述第二响应信息包括上述第一边缘服务器的地址信息和位置信息。
在其中一种可能的实施方式中，上述第一边缘服务器根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之前，还包括：
第一边缘服务器接收预设边缘服务器广播的上述终端的位置信息；
上述第一边缘服务器根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之后，还包括：
上述第一边缘服务器根据上述终端的位置信息计算上述第一边缘服务器与上述终 端之间的距离；
上述第一边缘服务器向上述任务卸载设备发送响应上述资源寻找请求的第三响应信息，上述第三响应信息包括上述第一边缘服务器的地址信息，以及包括上述第一边缘服务器与上述终端之间的距离。
在其中一种可能的实施方式中，上述第一边缘服务器根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之后，包括：上述第一边缘服务器向上述任务卸载设备发送响应上述资源寻找请求的第四响应信息，上述第四响应信息包括上述第一边缘服务器的地址信息。
第三方面描述的方法的有益效果可以参见上述第一方面对应的有益效果，此处不再赘述。
第四方面，本发明实施例提供一种移动边缘计算任务卸载设备，上述任务卸载设备为边缘计算系统中的设备，上述边缘计算系统还包括终端、中心服务器以及多个边缘服务器；上述任务卸载设备包括：
接收单元，用于接收上述终端发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，其中，上述任务卸载设备部署在上述终端接入的无线接入点中；
确定单元，用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内有满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
选择单元，用于从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器。
在其中一种可能的实施方式中，上述选择单元具体用于：
将上述资源寻找请求发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息；
根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行上述任务的服务器。
在其中一种可能的实施方式中，上述选择单元具体用于：
将上述资源寻找请求发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息和位置信息；
根据接收的上述位置信息和上述终端的位置信息计算上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离；
根据上述计算得到的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在其中一种可能的实施方式中，上述选择单元具体用于：
将上述资源寻找请求和上述终端的位置信息发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息，以及满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离；
根据满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器与上述终端之间的距离，从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在其中一种可能的实施方式中，上述任务卸载设备还包括发送单元，用于在上述选择单元根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行上述任务的服务器之后，向上述终端发送响应上述资源寻找请求的第一响应信息；
上述接收单元，还用于接收上述终端根据上述第一响应信息发送的上述任务；
上述发送单元，还用于将上述任务发送给上述选择出的服务器。
在其中一种可能的实施方式中，上述任务卸载设备还包括发送单元和计算单元；
上述确定单元，还用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述发送单元，用于将上述资源寻找请求发送给预设边缘服务器；
上述接收单元，还用于接收上述预设边缘服务器发送的第二响应信息，上述第二响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第二响应信息包括上述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述计算单元，用于根据上述满足上述计算资源的需求的服务器的位置信息和上述终端的位置信息，计算上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述选择单元，还用于根据上述计算得到的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在其中一种可能的实施方式中，上述任务卸载设备还包括发送单元；
上述确定单元，还用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述发送单元，用于将上述资源寻找请求和上述终端的位置信息发送给预设边缘服务器；
上述接收单元，还用于接收上述预设边缘服务器发送的第三响应信息，上述第三响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求和上述终端的位置信息的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述选择单元，还用于根据上述满足上述计算资源的需求的服务器与上述终端之间的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
第五方面，本发明实施例提供一种边缘服务器，上述边缘服务器为边缘计算系统 中的服务器，上述边缘计算系统还包括终端、任务卸载设备、中心服务器以及除了上述边缘服务器之外的多个边缘服务器，其中，上述任务卸载设备部署在上述终端接入的无线接入点中；上述边缘服务器包括：
接收单元，用于接收上述任务卸载设备发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，上述边缘服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；
确定单元，用于根据上述资源寻找请求确定上述边缘服务器不满足上述计算资源的需求；
广播单元，用于将上述资源寻找请求广播给其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，上述边缘服务器还包括发送单元；
上述接收单元，还用于在上述广播单元将上述资源寻找请求广播给上述其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第二响应信息，上述第二响应信息包括上述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述发送单元，用于将上述第二响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述接收单元，还用于在上述确定单元根据上述资源寻找请求确定上述边缘服务器不满足上述计算资源的需求之前，接收上述任务卸载设备发送的上述终端的位置信息；
上述广播单元，还用于在上述确定单元根据上述资源寻找请求确定上述边缘服务器不满足上述计算资源的需求之后，将上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，上述接收单元，还用于在上述广播单元将上述资源寻找请求和上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第三响应信息，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述发送单元，用于将上述第三响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述接收单元，还用于在上述广播单元将上述资源寻找请求广播给其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第四响应信息，上述第四响应信息包括上述满足上述计算资源的需求的服务器的地址信息；
上述发送单元，用于将上述第四响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，上述边缘服务器还包括发送单元；
上述确定单元，还用于在上述接收单元接收上述任务卸载设备发送的上述资源寻找请求之后，根据上述资源寻找请求确定上述边缘服务器满足上述计算资源的需求；
上述发送单元，用于向上述任务卸载设备发送上述边缘服务器的地址信息。
第六方面，本发明实施例提供一种边缘服务器，上述边缘服务器为边缘计算系统中的服务器，上述边缘计算系统还包括终端、任务卸载设备、除上述边缘服务器以外 的边缘服务器以及中心服务器；上述边缘服务器包括：
接收单元，用于接收预设边缘服务器广播的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息；上述资源寻找请求为上述预设边缘服务器接收上述任务卸载设备发送的资源寻找请求后，判断自身不满足上述计算资源的需求的情况下广播的；上述预设边缘服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；上述任务卸载设备部署在上述终端接入的无线接入点中；上述第一边缘服务器包括上述系统中除预设边缘服务器之外的服务器；
确定单元，用于根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求；
发送单元，用于向上述任务卸载设备发送响应上述资源寻找请求的第二响应信息，上述第二响应信息包括上述第一边缘服务器的地址信息和位置信息。
在其中一种可能的实施方式中，上述边缘服务器还包括计算单元；
上述接收单元，还用于在上述确定单元根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之前，接收预设边缘服务器广播的上述终端的位置信息；
上述计算单元，用于在上述确定单元根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之后，根据上述终端的位置信息计算上述第一边缘服务器与上述终端之间的距离；
上述发送单元，还用于向上述任务卸载设备发送响应上述资源寻找请求的第三响应信息，上述第三响应信息包括上述第一边缘服务器的地址信息，以及包括上述第一边缘服务器与上述终端之间的距离。
在其中一种可能的实施方式中，上述发送单元，还用于在上述确定单元根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之后，向上述任务卸载设备发送响应上述资源寻找请求的第四响应信息，上述第四响应信息包括上述第一边缘服务器的地址信息。
第七方面，本发明实施例提供一种移动边缘计算任务卸载设备，该服务器可包括：存储器以及与所述存储器耦合的处理器、通信接口。其中，所述通信接口可以用于支持所述服务器执行上述第一方面所提供的边缘计算服务器的寻找方法中发送信息和接收信息的步骤。其中，通信接口可以包括发射器和接收器。处理器用于支持所述任务卸载设备执行上述第一方面所提供的边缘计算服务器的寻找方法中除发送信息以及接收信息以外的其他处理步骤。
所述存储器用于存储上述第一方面描述的边缘计算服务器的寻找方法的实现计算机程序，所述处理器用于执行所述存储器中存储的程序，即执行第一方面所提供的边缘计算服务器的寻找方法，或者上述第一方面所提供的可能的实施方式中的任意一种方法。存储器和处理器可以集成在一起，也可以通过耦合器耦合。
第八方面，本发明实施例提供一种边缘服务器，该服务器可包括：存储器以及与所述存储器耦合的处理器、通信接口。其中，所述通信接口可以用于支持该边缘服务 器执行上述第二方面所提供的边缘计算服务器的寻找方法中发送信息和接收信息的步骤。其中，通信接口可以包括发射器和接收器。处理器用于支持该边缘服务器执行上述第二方面所提供的边缘计算服务器的寻找方法中除发送信息以及接收信息以外的其他处理步骤。
所述存储器用于存储上述第二方面描述的边缘计算服务器的寻找方法的实现计算机程序，所述处理器用于执行所述存储器中存储的程序，即执行第二方面所提供的边缘计算服务器的寻找方法，或者上述第二方面所提供的可能的实施方式中的任意一种方法。存储器和处理器可以集成在一起，也可以通过耦合器耦合。
第九方面，本发明实施例提供一种边缘服务器，该服务器可包括：存储器以及与所述存储器耦合的处理器、通信接口。其中，所述通信接口可以用于支持该边缘服务器执行上述第三方面所提供的边缘计算服务器的寻找方法中发送信息和接收信息的步骤。其中，通信接口可以包括发射器和接收器。处理器用于支持该边缘服务器执行上述第三方面所提供的边缘计算服务器的寻找方法中除发送信息以及接收信息以外的其他处理步骤。
所述存储器用于存储上述第三方面描述的边缘计算服务器的寻找方法的实现计算机程序，所述处理器用于执行所述存储器中存储的程序，即执行第三方面所提供的边缘计算服务器的寻找方法，或者上述第三方面所提供的可能的实施方式中的任意一种方法。存储器和处理器可以集成在一起，也可以通过耦合器耦合。
第十方面，本申请实施例提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述第一方面任意一项所述的方法。
第十一方面，本申请实施例提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述第二方面任意一项所述的方法。
第十二方面，本申请实施例提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述第三方面任意一项所述的方法。
第十三方面，本发明实施例提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面任意一项所述的方法。
第十四方面，本发明实施例提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行上述第二方面任意一项所述的方法。
第十五方面，本发明实施例提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行上述第三方面任意一项所述的方法。
综上所述，相比于现有技术的所有的资源及调度数据都由中心云分发以及配置，边缘云只做任务执行的架构，本发明通过部署在无线接入点中的任务卸载设备完成资源和数据的调度，即就近选择满足终端需求的服务器来计算完成终端需要计算的任务，实现了最低时延地响应终端的需求且节约网络开销。
附图说明
下面将对本申请实施例中所需要使用的附图作介绍。
图1为本方案实施例提供的边缘计算服务器的寻找方法的系统架构示意图；
图2为本方案实施例提供的边缘计算服务器的寻找方法的交互流程示意图；
图3为本方案实施例提供的边缘计算服务器的寻找方法的另一个交互流程示意图；
图4为本方案实施例提供的边缘计算服务器的寻找方法的另一个系统架构示意图；
图5为本方案实施例提供的一种任务卸载设备的逻辑结构示意图；
图6为本方案实施例提供的一种边缘服务器的逻辑结构示意图；
图7为本方案实施例提供的另一种边缘服务器的逻辑结构示意图；
图8为本方案实施例提供的一种任务卸载设备的硬件结构示意图；
图9为本方案实施例提供的一种边缘服务器的硬件结构示意图；
图10为本方案实施例提供的另一种边缘服务器的硬件结构示意图。
具体实施方式
下面结合附图对本申请实施例中的技术方案进行描述。
为了更好的理解本发明实施例提供的一种边缘计算服务器的寻找方法，下面先对本发明实施例适用的系统架构进行描述。参阅图1，图1是本发明实施例提供的边缘计算服务器的寻找方法适用的系统架构示意图。如图1所示，该系统架构可以包括四层，其中，第一层是终端层，第二层是无线接入层、第三层是边缘云层，第四层是中心云层。终端层包括一个或多个终端101；无线接入层包括一个或多个无线接入点102，以及一个或多个任务卸载设备103，即任务卸载设备103部署在无线接入层，每一个无线接入点可以对应部署一个或多个任务卸载设备103；边缘云层包括一个或多个边缘服务器104；中心云层包括一个或多个中心服务器105(多个服务器可以构成服务器集群)。需要说明的是，边缘服务器104可以是边缘云服务器，中心服务器105可以是中心云服务器。
终端101可以安装并运行相关的应用软件或APP，该相关的应用软件或APP为适用于用户终端的与用户的相关需求相适应的应用软件或APP。应用软件是为客户提供服务的计算机程序和指令的集合。这里，该服务可包括但不限于：向无线接入点102发送资源寻找请求和接收能够完成对应任务的服务器的互联网协议(internet protocol，IP)地址或者其它地址(例如媒体存储控制地址(cedia access control address，MAC))等等。终端101可以包括任何一种基于智能操作系统的手持式电子产品，其可与用户通过键盘、虚拟键盘、触摸板、触摸屏以及声控设备等输入设备来进行人机交互，诸如智能手机、平板电脑、手持计算机、可穿戴电子设备等。其中，智能操作系统包括但不限于任何通过向设备提供各种应用来丰富设备功能的操作系统，诸如安卓Android、IOS、Windows和MAC等系统。终端101还可以包括车载设备、车辆、机器人等人工智能设备。
无线接入点102是使用终端101的用户进入有线网络的接入点，主要用于宽带家庭、大楼内部、校园内部、园区内部以及仓库、工厂等需要无线监控的地方，典型距离覆盖几十米至上百米，也有可以用于远距离传送，目前最远的可以达到30KM左右，主要技术为IEEE802.11系列。大多数无线接入点还带有接入点客户端模式，可以和其 它接入点进行无线连接，延展网络的覆盖范围。
无线接入点102可以包括路由交换接入一体设备和纯接入点设备，一体设备执行接入和路由工作，纯接入设备只负责无线终端的接入，纯接入设备通常作为无线网络扩展使用，与其它接入点或者主接入点连接，以扩大无线覆盖范围，而一体设备一般是无线网络的核心。
在本申请实施例中，每一个无线接入点102的覆盖范围内都覆盖有一个或多个边缘服务器104，还可以覆盖有一个或多个中心服务器105，具体的覆盖情况根据实际情况确定，本方案对此不做限制。
此外，在本申请实施例中，无线接入点102可以用于接收终端101通过2G、3G、4G或5G等无线网络发送的资源寻找请求，然后将该请求发送到部署在该无线接入点102一侧的任务卸载设备103，由该任务卸载设备103对该请求进行处理(例如根据该请求判断该无线接入点102覆盖范围内是否有满足计算资源需求的服务器、将终端的任务发送给满足需求的服务器上执行等)。移动边缘计算可利用无线接入网络就近提供用户信息技术(information technology，IT)所需服务和云端计算功能，而创造出一个具备高性能、低延迟与高带宽的电信级服务环境，加速网络中各项内容、服务及应用的快速下载，让消费者享有不间断的高质量网络体验。
边缘服务器104是部署在中心云边缘侧的服务器，用于就近为用户终端提供各种需要的计算服务。例如，如果中心云部署在上海，而用户终端分散在广州、深圳和北京等地区，那么，可以在广州、深圳和北京等地区部署边缘服务器，从而使得当地的用户终端就近从附近的边缘服务器中获取需要的服务和计算功能，而不用长距离地向中心服务器寻求响应，从而节约了网络资源、提高了响应效率。
中心服务器105主要用于在边缘服务器无法响应终端的请求的时候提供必要的响应。
在本申请实施例中任务卸载设备103可以是移动边缘计算(mobile edge computing，MEC)服务器或者可以是其它与MEC服务器一样符合相同通信协议标准且可以实现相关功能的设备。
可选的，上述任务卸载设备上部署有业务卸载代理，该业务卸载代理可以是一个软件或应用程序，上述任务卸载设备运行该业务卸载代理可以实现本申请实施例中任务卸载设备所执行的操作。
在现有技术中，由于任务卸载设备器通常以通用和小型服务器为主，提供的计算资源有限，无法处理大量的虚拟现实(virtual reality，VR)场景、增强现实(augmented reality，AR)场景或云游戏重型场景的计算。而本申请实施例中的边缘服务器104为就近部署在终端附近的服务器，有足够的计算资源完成VR、AR和云游戏等场景所需要的计算。
需要说明的是，本申请实施例提供的边缘计算服务器的寻找方法适用的系统架构不限于图1所示系统架构。
下面介绍一种边缘计算服务器的寻找方法，可以应用于如图1所示的系统架构。该方法包括但不限于如下步骤：
步骤201、终端向任务卸载设备发送资源寻找请求，上述请求携带有完成任务需 要的计算资源的信息。
步骤202、上述任务卸载设备接收上述终端发送的资源寻找请求。
在具体实施例中，该任务卸载设备部署在该终端接入的无线接入点中。该任务卸载设备可以与该无线接入点通信。具体的，无线接入点接收到上述资源寻找请求之后，任务卸载设备从无线接入点获取到该请求。此外，该无线接入点可以获取到其覆盖范围内的边缘服务器和/或中心服务器的资源空闲情况等信息，任务卸载设备可以通过无线接入点获取到的这些信息，例如可以通过网络切片等技术获取这些信息。可选的，边缘服务器和/或中心服务器的资源空闲情况可以包括处理器的内存空闲情况、图形处理器(graphics processing unit，GPU)资源的空闲情况、硬盘空间的空闲情况以及输入/输出(Input/Output，I/O)接口或设备的空闲情况等信息。
可选的，上述任务可以是进行图像处理例如渲染、编码等处理的任务，那么上述资源寻找请求例如可以是寻找计算资源执行该图像处理任务的请求。
可选的，该计算资源可以包括处理器的内存、GPU、硬盘和I/O等信息。上述资源寻找请求中携带的计算资源的信息可以包括完成该任务所需要的内存空间、GPU的类型和/或数量、硬盘空间的大小、I/O以及时间等信息。例如，计算资源的信息可以是执行任务的服务器中处理器的内存不小于1G、需要2个GPU、硬盘空间的大小不少于4G以及需要1秒钟完成等信息。这里只是为了便于理解示例性说明，具体实施时不限于此处的描述。
可选的，上述终端可以是图1中所示的终端101，上述无线接入点可以是图1所示的无线接入点102，上述任务卸载设备可以是图1所示的任务卸载设备103，上述边缘服务器可以是图1所示的边缘服务器104。
步骤203、上述任务卸载设备根据上述资源寻找请求确定预设范围内有满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器。
步骤204、上述任务卸载设备从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器。
可选的，上述预设范围包括上述无线接入点覆盖的空间范围，该空间范围内可以包括边缘服务器和/或中心服务器。可选的，这里的中心服务器可以是图1所述的中心服务器105。
在具体实施例中，任务卸载设备接收到上述请求后，解析该请求，获取该请求中包括的完成上述任务需要的计算资源的信息，然后，将该需要的计算资源的信息与无线接入点覆盖范围内的边缘服务器和/或中心服务器的资源空闲情况等信息逐一对比，以确定是否有满足上述任务需求的服务器。
为了便于理解上述服务器是否满足需求，下面举例说明，参见表1。
表1
Figure PCTCN2020104835-appb-000001
Figure PCTCN2020104835-appb-000002
假设上述完成上述任务需要的计算资源的信息为需要服务器的内存空闲不小于2G、一个第一类型的GPU以及服务器的硬盘空间空闲不小于50G。那么，将该信息包括的各项情况逐一与表1中的服务器的情况进行比对。可以发现，表1中只有服务器1符合上述任务的需求，而服务器2和服务器3不满足需求。表1只是为了便于理解而示例性示出的一个简单的例子，具体的计算资源需求根据具体情况确定，本方案对此不作限制。
上述在确定预设范围内有服务器满足上述计算资源的需求的情况下，如果上述预设范围内只有一个满足所述计算资源需求的服务器，那么上述任务卸载设备则选择该一个满足需求的服务器作为执行上述任务的服务器。然后，任务卸载设备会将上述资源寻找请求发送给该一个满足需求的服务器。该一个满足需求的服务器接收到该请求后，根据该请求向任务卸载设备发送本服务器的地址信息。
所述地址信息包括所述满足需求的服务器的互联网协议IP地址或者媒体存储控制地址(cedia access control address，MAC)等可以将报文路由到服务器的地址信息。在本申请实施例中，该地址信息可以用于指示所述任务卸载设备将所述任务发送给所述地址信息对应的服务器上执行。
如果上述预设范围内有多个满足所述计算资源需求的服务器，那么上述任务卸载设备则会从该多个满足需求的服务器中选择一个服务器作为执行上述任务的服务器。
下面介绍任务卸载设备如何从上述多个满足需求的服务器中选择一个服务器作为执行上述任务的服务器：上述任务卸载设备将上述资源寻找请求同步发送到上述多个满足需求的服务器；上述任务卸载设备接收上述多个满足需求的服务器发送的地址信息；上述任务卸载设备根据接收上述多个满足需求的服务器发送的地址信息的先后顺序选择执行上述任务的服务器。
在具体实施例中，上述任务卸载设备会分别将上述资源寻找请求同时发送到上述多个满足需求的服务器，该多个满足需求的服务器接收到该请求之后，分别根据该请求向任务卸载设备发送各自的地址信息，然后任务卸载设备可以选择最先接到的地址信息对应的服务器作为执行上述任务的服务器。当然，任务卸载设备也可以选择接收到的第二个或第三个地址信息对应的服务器作为执行上述任务的服务器。具体选择的服务器根据具体的情况决定，本方案对此不做限定。
在其中一种可能的实施例中，上述“步骤204、上述任务卸载设备从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器”也可以是如下一种实现方式：
在这种实现方式中，上述任务卸载设备将上述资源寻找请求发送给上述满足需求的服务器之后，上述满足需求的服务器根据该资源寻找请求向上述任务卸载设备发送本服务器的地址信息和位置信息。该位置信息可以是实际的位置的信息，例如该位置信息可以是上述满足需求的服务器所在位置的经纬度的信息，或者也可以是系统预设 的上述满足需求的服务器的位置标记的信息等等。
上述任务卸载设备根据接收到的上述满足需求的服务器的位置信息与终端的位置信息，计算上述满足需求的服务器与上述终端之间的距离，然后选择上述满足需求的服务器中距离上述终端最近的服务器作为执行上述任务的服务器。
在其中一种可能的实施例中，上述“步骤204、上述任务卸载设备从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器”还可以是如下一种实现方式：
在这种实现方式中任务卸载设备将上述资源寻找请求发送给上述满足需求的服务器的同时，还将上述终端的位置信息发送给上述满足需求的服务器。
可选的，在这种实现方式中任务卸载设备可以是将在上述资源寻找请求中添加了上述终端的位置信息的请求发送给上述满足需求的服务器。
具体的，该终端的位置信息可以是实际的位置的信息，例如该位置信息可以是该终端所在位置的经纬度的信息，或者也可以是系统预设的位置标记的信息等等。
可选的，上述终端接入的无线接入点可以获取到该终端的位置信息，然后上述任务卸载设备可以通过该无线接入点获取到该位置信息，例如通过网络切片技术获取到该位置信息。
在具体实施例中，如果上述预设范围内只有一个满足所述计算资源需求的服务器，那么上述任务卸载设备则选择该一个满足需求的服务器作为执行上述任务的服务器。可以不向上述一个满足需求的服务器发送上述终端位置信息，然后具体的操作参见图2所述步骤204中对应的描述，此处不再赘述。
如果上述预设范围内有多个满足所述计算资源需求的服务器，那么上述任务卸载设备则会从该多个满足需求的服务器中选择一个服务器作为执行上述任务的服务器。
具体的，上述任务卸载设备会分别向上述多个满足需求的服务器发送在上述资源寻找请求和上述终端的位置信息，该多个满足需求的服务器接收到该请求和终端的位置信息之后，分别根据该终端的位置信息计算本服务器与该终端之间的距离，然后向任务卸载设备发送本服务器的地址信息，以及本服务器与上述终端的距离。任务卸载设备接收到这些信息之后，任务卸载设备选择与该终端距离最近的服务器作为执行上述任务的服务器。
上述任务卸载设备选择出执行上述任务的服务器之后，有如下两种处理方式：
第一种处理方式，任务卸载设备将对所述终端发送的上述资源寻找请求的响应信息发送给终端。
在这种处理方式中，将上述响应信息称为第一响应信息，该第一响应信息包括用于指示终端将上述需要计算的任务发送给任务卸载设备的指示信息。
在具体实施例中，任务卸载设备将上述第一响应信息发送给终端，终端接收到该第一响应信息后，解析该信息得到指示信息，然后根据该指示信息将上述需要执行的任务发送给任务卸载设备。任务卸载设备接收到该任务后，将该任务发送给上述选择出的服务器，由该服务器执行该任务，并将执行结果发送给任务卸载设备，由任务卸 载设备将执行结果发送给终端。
可选的，上述第一响应信息中的指示信息可以是上述选择出的执行上述任务的服务器的地址信息，也可以是其它的预定的信息等。
第二种处理方式，任务卸载设备将上述选择出的执行上述任务的服务器的地址信息发送给终端，终端接收到该地址信息之后，直接将需要执行的任务根据该地址信息发送给上述选择出的执行上述任务的服务器，由该服务器执行该任务，并将执行结果直接发送给终端。
上述方法实施例介绍的是在上述预设范围内有满足上述计算资源的需求的服务器的情况，下面介绍在上述预设范围内没有满足上述计算资源的需求的服务器的情况的实施例，参见图3。
步骤301、终端向任务卸载设备发送资源寻找请求，上述请求携带有完成任务需要的计算资源的信息。
步骤302、上述任务卸载设备接收上述终端发送的资源寻找请求。
在具体实施例中，步骤301和步骤302对应与上述图2所述的步骤201和步骤202相同，具体的实现方式可以参见上述对应的描述，此处不再赘述。
步骤303、上述任务卸载设备根据上述资源寻找请求确定上述预设范围内没有服务器满足上述计算资源的需求。
步骤304、上述任务卸载设备将上述资源寻找请求发送给预设边缘服务器。
具体的，上述预设边缘云服务器为预设的在所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心云服务器或边缘云服务器的情况下，用于寻找其它满足需求的云服务器来执行所述任务的服务器。
可选的，上述预设边缘服务器可以是一个或多个。
可选的，上述预设边缘服务器可以是上述无线接入点覆盖范围内的边缘服务器，也可以是同一个网络中上述终端接入的无线接入点以外的其它无线接入点覆盖范围内的边缘服务器。
步骤305、上述预设边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求。
步骤306、上述预设边缘服务器将上述资源寻找请求广播给其它边缘服务器和/或中心服务器。
可选的，上述其它边缘服务器和/或中心服务器包括上述无线接入点所在网络中除上述预设边缘服务器之外的边缘服务器和/或中心服务器。
在具体实施例中，上述边缘服务器在接收到任务卸载设备发送的资源寻找请求之后，解析该请求，获取该请求中的计算资源的信息，然后将该计算资源的信息与本服务器自己的计算资源逐一比较，以判断自身是否满足上述任务所需计算资源的需求。在确定自身不满足的情况下，上述预设边缘服务器可以将上述资源寻找请求广播给上述其它边缘服务器和/或中心服务器。
可选的，上述预设边缘服务器和上述其它边缘服务器和/或中心服务器可以是以全连接(full mesh)的形式组网。
上述其它边缘服务器和/或中心服务器接收到广播的资源寻找请求之后，解析该资源寻找请求获取到请求中的完成上述任务所需的计算资源的信息，然后将该计算资源与各自服务器中的计算资源逐一做比对，以判断是否满足上述任务所需的计算资源的需求。
可选的，如果该其它边缘服务器和/或中心服务器中没有服务器满足需求，且该其它边缘服务器和/或中心服务器中包括上述中心服务器，那么所有不满足需求的服务器向上述预设边缘服务器发送不满足需求的响应信息，然后上述预设边缘服务器将这些响应信息发送给上述任务卸载设备，任务卸载设备根据这些响应信息判断出本次资源寻找失败，并将该资源寻找失败的响应发送给上述终端。
可选的，如果该其它边缘服务器和/或中心服务器中没有服务器满足需求，且该其它边缘服务器和/或中心服务器中不包括上述中心服务器，那么所有不满足需求的服务器向上述预设边缘服务器发送不满足需求的响应信息，然后上述预设边缘服务器根据这些响应信息将上述资源寻找请求发送给上述中心服务器，由中心服务器根据该请求寻找满足需求的服务器。
可选的，如果该其它边缘服务器和/或中心服务器中有满足需求的服务器，那么满足需求的服务器将自身的地址信息发送给上述预设边缘服务器，然后，预设边缘服务器将接收到的地址信息发送给上述任务卸载设备。
如果上述其它边缘服务器和/或中心服务器中只有一个满足需求的服务器。在这种情况下，上述任务卸载设备将该一个满足需求的服务器确定为执行上述任务的服务器。确定出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
如果上述其它边缘服务器和/或中心服务器中有多个满足需求的服务器。在这种情况下，任务卸载设备可以根据接收到的该多个满足需求的服务器的地址信息的先后顺序来选择用于执行上述任务的服务器。例如，可以选择该最先接收到的地址信息对应的服务器作为执行上述任务的服务器。当然，任务卸载设备也可以选择接收到的第二个或第三个地址信息对应的服务器作为执行上述任务的服务器。具体选择的服务器根据具体的情况决定，本方案对此不做限定。选择出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
在其中一种可能的实施方式中，上述“步骤304、上述任务卸载设备将上述资源寻找请求发送给预设边缘服务器”之后，还包括：上述边缘服务器根据上述资源寻找请求确定上述预设边缘服务器满足上述计算资源的需求。
在这种情况下，上述预设边缘服务器将自身的地址信息发送给任务卸载设备，任务卸载设备接收到该预设边缘服务器的地址信息之后，确定该预设边缘服务器为执行上述任务的服务器。确定出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
在一些可能的实施方式中，上述“步骤306、上述预设边缘服务器将上述资源寻找请求广播给其它边缘服务器和/或中心服务器”之后，还可以包括如下的实施方式：
上述其它边缘服务器和/或中心服务器接收到上述资源寻找请求之后，同样的，先解析该请求获取到请求中的完成上述任务所需的计算资源的信息，然后将该计算资源与各自服务器中的计算资源逐一做比对，以判断是否满足上述任务所需的计算资源的需求。
可选的，如果该其它边缘服务器和/或中心服务器中没有服务器满足需求，那么具体的实现方式可以对应参照上述图3步骤306中对应的描述，此处不再赘述。
可选的，如果该其它边缘服务器和/或中心服务器中有服务器满足需求，该其它边缘服务器和/或中心服务器中满足需求的服务器根据该请求向上述预设边缘服务器发送自身所在的位置信息和地址信息。然后，预设边缘服务器将接收到的地址信息和位置信息发送给上述任务卸载设备。
同样的，如果上述其它边缘服务器和/或中心服务器中只有一个满足需求的服务器。在这种情况下，上述任务卸载设备将该一个满足需求的服务器确定为执行上述任务的服务器。确定出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
如果上述其它边缘服务器和/或中心服务器中有多个满足需求的服务器。在这种情况下，任务卸载设备先根据接收到的该多个满足需求的服务器的位置信息和上述终端的位置信息计算出终端分别与该多个满足需求的服务器的距离，然后任务卸载设备选择该多个满足需求的服务器中与终端距离最近的服务器作为执行上述任务的服务器。选择出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
在一些可能的实施方式中，在“步骤303、上述任务卸载设备根据上述资源寻找请求确定上述预设范围内没有服务器满足上述计算资源的需求”之后，在“步骤305、上述边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求”之前还包括：上述任务卸载设备还将上述终端的位置信息发送给预设边缘服务器。
在这种情况下，在“步骤305、上述边缘服务器根据上述资源寻找请求确定上述预设边缘服务器不满足上述计算资源的需求”之后，除了将上述资源寻找请求广播给其它边缘服务器和/或中心服务器之外，还将上述终端的位置信息广播给该其它边缘服务器和/或中心服务器。
在具体实施例中，如果该其它边缘服务器和/或中心服务器中有满足需求的服务器，那么满足需求的服务器根据接收到的上述终端的位置信息计算本服务器与上述终端的距离，然后将该计算得到的距离和自身的地址信息发送给上述预设边缘服务器。预设边缘服务器将接收到的距离和地址信息发送给上述任务卸载设备。
同样的，如果上述其它边缘服务器和/或中心服务器中只有一个满足需求的服务器。在这种情况下，上述任务卸载设备将该一个满足需求的服务器确定为执行上述任务的服务器。确定出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
如果上述其它边缘服务器和/或中心服务器中有多个满足需求的服务器。在这种情 况下，任务卸载设备先根据接收到的该多个满足需求的服务器分别与终端的距离，选择出该多个满足需求的服务器中与终端距离最近的服务器作为执行上述任务的服务器。选择出执行上述任务的服务器之后的具体实现可以参见“上述任务卸载设备选择出执行上述任务的服务器之后”列出的两种处理方式，此处不再赘述。
在其中一种可能的实施方式中，上述任务卸载设备接收到终端发送的资源寻找请求之后，可以先不判断上述预设范围内是否有服务器满足计算资源的需求，而是直接将上述资源寻找请求发送给预设边缘服务器。预设边缘服务器可以根据该资源寻找请求判断自身是否满足需求，满足则向任务卸载设备发送自身的地址信息，如果不满则向上述其它边缘服务器和/或中心服务器广播该资源寻找请求。然后广播之后的具体实施方式可以参见上述方法实施例中对应的描述，此处不再赘述。
在其中一种可能的实施方式中，上述任务卸载设备接收到终端发送的资源寻找请求之后，可以先不判断上述预设范围内是否有服务器满足计算资源的需求，而是将上述资源寻找请求发送给预设边缘服务器。预设边缘服务器也可以先不根据该资源寻找请求判断自身是否满足需求，而是直接向上述其它边缘服务器和/或中心服务器广播该资源寻找请求。然后广播之后的具体实施方式可以参见上述方法实施例中对应的描述，此处不再赘述。
基于上述方法实施例，下面介绍本申请实施例提供的一种边缘计算服务器的寻找方法中的最优实施例。参见图4，图4示例性示出了图1所述系统中一个无线接入点覆盖范围内的终端和边缘服务器的架构示意图。
参见图4，一个无线接入点中部署有任务卸载设备402(可以部署一个或多个任务卸载设备402)，任务卸载设备402中部署了业务卸载代理。无线接入点覆盖范围内包括一个或多个终端401，终端401中部署了任务计算管理器。无线接入点覆盖范围内包括一个或多个边缘服务器403，边缘服务器403中部署了任务接收器。边缘服务器403可以与中心服务器404互相通信，中心服务器404中部署了任务协调器。
任务卸载设备402中的业务卸载代理可以通过无线接入点获取到预设范围内的边缘服务器403的信息，提供局部调度能力，以快速寻找到满足终端需求的服务器从而响应终端的请求；同时在预设范围内没有服务器满足终端需求的情况下，通过对请求做进一步的封装，发送给预设边缘服务器的任务接收器，以进行更大范围的寻找操作。该预设边缘服务器可以与前述方法实施例中的预设边缘服务器相同。该预设范围为该无线接入点所覆盖的空间范围。
边缘服务器403中的任务接收器，向下可以给无线接入点提供接收任务请求的应用程序接口(application programming interface，API)，向上可以提供发送请求广播给其它边缘服务器，或者发送到中心侧的任务协调器的API。
中心服务器404的任务协调器，它的API主要提供给边缘服务器的任务接收器，当边缘服务器广播的方式也不能满足任务的需求时，任务协调器可以做全局的调度。
需要说明的是，上述业务卸载代理、任务计算管理器、任务接收器和任务协调器可以是由软件来实现，具体的实现方式根据具体情况决定，本方案对此不做限制。
在具体的实施例中，终端401生成需要执行的任务的资源寻找请求(该请求可以与前述方法实施例中的资源请求相同)后，可以先将该请求发送给该终端401中的计算任务管理器中，由该计算任务管理器计算该终端401是否有资源完成该任务。
如果上述终端401中没有资源完成该任务，那么终端401会将上述资源寻找请求发送给无线接入点，然后任务卸载设备402基于网络切片技术，从无线接入点获取到该请求，并将该请求发送至业务卸载代理。具体的，业务卸载代理使用任务卸载设备的能力API，例如Traffic off load，注册服务类型或流量切片，再由任务卸载设备将指定的流向解调并发送到在任务卸载设备的业务卸载代理上。
上述业务卸载代理获取到上述请求之后，解析该请求，获取到该请求中的完成上述任务所需要的计算资源的信息，例如可以包括处理器的内存空闲情况、图形处理器(graphics processing unit，GPU)资源的空闲情况、硬盘空间的空闲情况以及输入/输出(Input/Output，I/O)接口或设备的空闲情况等信息。同时，该业务卸载代理通过上述无线接入点获取到预设范围内的边缘服务器403的计算资源空闲情况。
然后，上述业务卸载代理将从上述请求中获取到的计算资源与预设范围内的边缘服务器403的空闲的计算资源逐一对比，以判断是否有服务器满足请求中的计算资源的需求。
如果有服务器满足上述需求，则上述业务卸载代理将上述请求重新封装好之后发送给满足需求的边缘服务器403中的任务接收器，由任务接收器对该请求进行解析，然后根据解析结果向上述任务卸载设备402发送本服务器的地址信息。该任务卸载设备402接收到该地址信息之后，将该地址信息发送至本服务器的业务卸载代理，由业务卸载代理进行后续的处理。需要说明的是前述方法实施例中由任务卸载设备处理的操作可以由该任务卸载设备中的业务卸载代理来处理，该后续的处理可以参见前述方法实施例中的对应的描述，此处不再赘述。
如果上述判断后没有服务器满足上述需求，那么上述业务卸载代理将上述请求重新封装好之后发送给预设边缘服务器的任务接收器，由该任务接收器做后续的处理。需要说明的是前述方法实施例中由预设边缘服务器处理的操作可以由该预设边缘服务器中的任务接收器来处理，该后续的处理可以参见前述方法实施例中的对应的描述，此处不再赘述。
在其中一种可能的实施方式中，上述业务卸载代理将上述请求重新封装好之后发送给预设边缘服务器的任务接收器，包括将在该请求中添加了上述终端401的位置信息之后的请求重新封装好之后发送给预设边缘服务器的任务接收器，然后由该任务接收器做后续的处理。同样的，需要说明的是前述方法实施例中由预设边缘服务器处理的操作可以由该预设边缘服务器中的任务接收器来处理，该后续的处理可以参见前述方法实施例中的对应的描述，此处不再赘述。
上述主要从终端、任务卸载设备和服务器(包括边缘服务器和中心服务器)之间交互的角度对本申请实施例提供的方案进行了介绍。可以理解的是，各个网元，例如终端、任务卸载设备和服务器等为了实现上述对应的功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开 的实施例描述的各示例的单元及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对终端、任务卸载设备和服务器等进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
在采用对应各个功能划分各个功能模块的情况下，图5示出了上述任意一个方法实施例中所涉及的任务卸载设备的一种可能的逻辑结构示意图，任务卸载设备500包括：
接收单元501，用于接收上述终端发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，其中，任务卸载设备500部署在上述终端接入的无线接入点中；
确定单元502，用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内有满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
选择单元503，用于从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行上述任务的服务器。
在一些可能的实施方式中，选择单元503具体用于：
将上述资源寻找请求发送给满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器；
接收满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器根据上述资源寻找请求发送的地址信息；
根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行上述任务的服务器。
在一些可能的实施方式中，任务卸载设备500还包括发送单元，用于在选择单元503根据接收的上述地址信息的先后顺序从满足上述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行上述任务的服务器之后，向上述终端发送响应上述资源寻找请求的第一响应信息；
接收单元501，还用于接收上述终端根据上述第一响应信息发送的上述任务；
上述发送单元，还用于将上述任务发送给上述选择出的服务器。
在一些可能的实施方式中，任务卸载设备500还包括发送单元和计算单元；
确定单元502，还用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述发送单元，用于将上述资源寻找请求发送给预设边缘服务器；
接收单元501，还用于接收上述预设边缘服务器发送的第二响应信息，上述第二响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第二响应信息包括上 述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述计算单元，用于根据上述满足上述计算资源的需求的服务器的位置信息和上述终端的位置信息，计算上述满足上述计算资源的需求的服务器与上述终端之间的距离；
选择单元503，还用于根据上述计算得到的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在一些可能的实施方式中，任务卸载设备500还包括发送单元；
确定单元502，还用于根据上述资源寻找请求确定上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器；
上述发送单元，用于将上述资源寻找请求和上述终端的位置信息发送给预设边缘服务器；
接收单元501，还用于接收上述预设边缘服务器发送的第三响应信息，上述第三响应信息为接收到上述预设边缘服务器广播的上述资源寻找请求和上述终端的位置信息的服务器中，满足上述计算资源的需求的服务器向上述预设边缘服务器发送的，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
选择单元503，还用于根据上述满足上述计算资源的需求的服务器与上述终端之间的距离，从上述满足上述计算资源的需求的服务器中，选择距离上述终端最近的服务器用于执行上述任务。
在采用对应各个功能划分各个功能模块的情况下，图6示出了上述任意一项所述方法实施例中所涉及的预设边缘服务器的一种可能的逻辑结构示意图，边缘服务器600包括：
接收单元601，用于接收上述任务卸载设备发送的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息，边缘服务器600为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；
确定单元602，用于根据上述资源寻找请求确定边缘服务器600不满足上述计算资源的需求；
广播单元603，用于将上述资源寻找请求广播给其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，边缘服务器600还包括发送单元；
接收单元601，还用于在广播单元603将上述资源寻找请求广播给上述其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第二响应信息，上述第二响应信息包括上述满足上述计算资源的需求的服务器的地址信息和位置信息；
上述发送单元，用于将上述第二响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，接收单元601，还用于在确定单元602根据上述资源寻找请求确定边缘服务器600不满足上述计算资源的需求之前，接收上述任务卸载设备发送的上述终端的位置信息；
广播单元603，还用于在确定单元602根据上述资源寻找请求确定边缘服务器600不满足上述计算资源的需求之后，将上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器。
在其中一种可能的实施方式中，接收单元601，还用于在广播单元603将上述资源寻找请求和上述终端的位置信息广播给上述其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第三响应信息，上述第三响应信息包括上述满足上述计算资源的需求的服务器的地址信息，以及包括上述满足上述计算资源的需求的服务器与上述终端之间的距离；
上述发送单元，用于将上述第三响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，接收单元601，还用于在广播单元603将上述资源寻找请求广播给其它边缘服务器和/或中心服务器之后，接收上述其它边缘服务器和/或中心服务器中满足上述计算资源的需求的服务器发送的第四响应信息，上述第四响应信息包括上述满足上述计算资源的需求的服务器的地址信息；
上述发送单元，用于将上述第四响应信息发送给上述任务卸载设备。
在其中一种可能的实施方式中，边缘服务器600还包括发送单元；
确定单元602，还用于在接收单元601接收上述任务卸载设备发送的上述资源寻找请求之后，根据上述资源寻找请求确定边缘服务器600满足上述计算资源的需求；
上述发送单元，用于向上述任务卸载设备发送边缘服务器600的地址信息。
在采用对应各个功能划分各个功能模块的情况下，图7示出了上述任意一项所述方法实施例中所涉及的其它边缘服务器和/或中心服务器中任一个服务器的一种可能的逻辑结构示意图，服务器700包括：
接收单元701，用于接收预设边缘服务器广播的资源寻找请求，上述资源寻找请求携带有完成任务需要的计算资源的信息；上述资源寻找请求为上述预设边缘服务器接收上述任务卸载设备发送的资源寻找请求后，判断自身不满足上述计算资源的需求的情况下广播的；上述预设边缘服务器为预设的在上述终端接入的无线接入点覆盖的空间范围内没有满足上述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行上述任务的服务器；上述任务卸载设备部署在上述终端接入的无线接入点中；上述第一边缘服务器包括上述系统中除预设边缘服务器之外的服务器；
确定单元702，用于根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求；
发送单元703，用于向上述任务卸载设备发送响应上述资源寻找请求的第二响应信息，上述第二响应信息包括上述第一边缘服务器的地址信息和位置信息。
在其中一种可能的实施方式中，服务器700还包括计算单元；
接收单元701，还用于在确定单元702根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之前，接收预设边缘服务器广播的上述终端的位置信息；
上述计算单元，用于在确定单元702根据上述资源寻找请求确定上述第一边缘服 务器满足上述计算资源的需求之后，根据上述终端的位置信息计算上述第一边缘服务器与上述终端之间的距离；
发送单元703，还用于向上述任务卸载设备发送响应上述资源寻找请求的第三响应信息，上述第三响应信息包括上述第一边缘服务器的地址信息，以及包括上述第一边缘服务器与上述终端之间的距离。
在其中一种可能的实施方式中，发送单元703，还用于在确定单元702根据上述资源寻找请求确定上述第一边缘服务器满足上述计算资源的需求之后，向上述任务卸载设备发送响应上述资源寻找请求的第四响应信息，上述第四响应信息包括上述第一边缘服务器的地址信息。
图8所示，为本申请实施例提供的边缘计算服务器的寻找方法实施例中所涉及的任务卸载设备的一种可能的硬件结构示意图。任务卸载设备800包括：处理器801、存储器802和通信接口803。处理器801、通信接口803以及存储器802可以相互连接或者通过总线804相互连接。
示例性的，存储器802用于存储任务卸载设备800的计算机程序和数据，存储器802可以包括但不限于是随机存储记忆体(random access memory，RAM)、只读存储器(read-only memory，ROM)、可擦除可编程只读存储器(erasable programmable read only memory，EPROM)或便携式只读存储器(compact disc read-only memory，CD-ROM)等。通信接口803用于支持任务卸载设备800进行通信，例如接收或发送数据。
示例性的，处理器801可以是中央处理器单元、通用处理器、数字信号处理器、专用集成电路、现场可编程门阵列或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。处理器也可以是实现计算功能的组合，例如包含一个或多个微处理器组合，数字信号处理器和微处理器的组合等等。处理器801可以用于读取上述存储器802中存储的程序，执行上述图2、图3以及可能的实施方式所述方法中任务卸载设备所实现的操作。
图9所示，为本申请实施例提供的边缘计算服务器的寻找方法实施例中所涉及的预设边缘服务器的一种可能的硬件结构示意图。边缘服务器900包括：处理器901、存储器902和通信接口903。处理器901、通信接口903以及存储器902可以相互连接或者通过总线904相互连接。
示例性的，存储器902用于存储边缘服务器900的计算机程序和数据，存储器902可以包括但不限于RAM、ROM、EPROM或CD-ROM等。通信接口903用于支持边缘服务器900进行通信，例如接收或发送数据。
示例性的，处理器901可以是中央处理器单元、通用处理器、数字信号处理器、专用集成电路、现场可编程门阵列或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。处理器也可以是实现计算功能的组合，例如包含一个或多个微处理器组合，数字信号处理器和微处理器的组合等等。处理器901可以用于读取上述存储器902中存储的程序，执行上述图2、图3以及可能的实施方式所述方法中预设边缘服务器所实现的操作。
图10所示，为本申请实施例提供的边缘计算服务器的寻找方法实施例中所涉及的其它边缘服务器和/或中心服务器中任一个服务器的一种可能的硬件结构示意图。服务器1000包括：处理器1001、存储器1002和通信接口1003。处理器1001、通信接口1003以及存储器1002可以相互连接或者通过总线1004相互连接。
示例性的，存储器1002用于存储服务器1000的计算机程序和数据，存储器1002可以包括但不限于RAM、ROM、EPROM或CD-ROM等。通信接口1003用于支持服务器1000进行通信，例如接收或发送数据。
示例性的，处理器1001可以是中央处理器单元、通用处理器、数字信号处理器、专用集成电路、现场可编程门阵列或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。处理器也可以是实现计算功能的组合，例如包含一个或多个微处理器组合，数字信号处理器和微处理器的组合等等。处理器1001可以用于读取上述存储器1002中存储的程序，执行上述图2、图3以及可能的实施方式所述方法中其它边缘服务器和/或中心服务器中任一个服务器所实现的操作。
本申请实施例还提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述图2、图3以及可能的实施方式所述方法中任务卸载设备所实现的操作。
本申请实施例还提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述图2、图3以及可能的实施方式所述方法中预设边缘服务器所实现的操作。
本申请实施例还提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现上述图2、图3以及可能的实施方式所述方法中其它边缘服务器和/或中心服务器中任一个边缘服务器所实现的操作。
综上所述，相比于现有技术的所有的资源及调度数据都由中心云分发以及配置，边缘云只做任务执行的架构，本发明通过部署在无线接入点中的任务卸载设备完成资源和数据的调度，即就近选择满足终端需求的服务器来计算完成终端需要计算的任务，实现了最低时延地响应终端的需求且节约网络开销。
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims (24)

  1. 一种边缘计算服务器的寻找方法，其特征在于，所述方法应用于边缘计算系统，所述边缘计算系统包括终端、任务卸载设备、中心服务器以及多个边缘服务器；所述方法包括：
    所述任务卸载设备接收所述终端发送的资源寻找请求，所述资源寻找请求携带有完成任务需要的计算资源的信息，其中，所述任务卸载设备部署在所述终端接入的无线接入点中；
    所述任务卸载设备根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内有满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器；
    所述任务卸载设备从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器。
  2. 根据所述权利要求1所述方法，其特征在于，所述任务卸载设备从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器，包括：
    所述任务卸载设备将所述资源寻找请求发送给满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器；
    所述任务卸载设备接收满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器根据所述资源寻找请求发送的地址信息；
    所述任务卸载设备根据接收的所述地址信息的先后顺序从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行所述任务的服务器。
  3. 根据权利要求2所述方法，其特征在于，所述任务卸载设备根据接收的所述地址信息的先后顺序从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行所述任务的服务器之后，所述方法还包括：
    所述任务卸载设备向所述终端发送响应所述资源寻找请求的第一响应信息；
    所述任务卸载设备接收所述终端根据所述第一响应信息发送的所述任务；
    所述任务卸载设备将所述任务发送给所述选择出的服务器。
  4. 根据权利要求1至3任一项所述方法，其特征在于，所述方法还包括：
    所述任务卸载设备根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器；
    所述任务卸载设备将所述资源寻找请求发送给预设边缘服务器；
    所述任务卸载设备接收所述预设边缘服务器发送的第二响应信息，所述第二响应信息为接收到所述预设边缘服务器广播的所述资源寻找请求的服务器中，满足所述计算资源的需求的服务器向所述预设边缘服务器发送的，所述第二响应信息包括所述满足所述计算资源的需求的服务器的地址信息和位置信息；
    所述任务卸载设备根据所述满足所述计算资源的需求的服务器的位置信息和所述终端的位置信息，计算所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述任务卸载设备根据所述计算得到的距离，从所述满足所述计算资源的需求的服务器中，选择距离所述终端最近的服务器用于执行所述任务。
  5. 根据权利要求1至3任一项所述方法，其特征在于，所述方法还包括：
    所述任务卸载设备根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器；
    所述任务卸载设备将所述资源寻找请求和所述终端的位置信息发送给预设边缘服务器；
    所述任务卸载设备接收所述预设边缘服务器发送的第三响应信息，所述第三响应信息为接收到所述预设边缘服务器广播的所述资源寻找请求和所述终端的位置信息的服务器中，满足所述计算资源的需求的服务器向所述预设边缘服务器发送的，所述第三响应信息包括所述满足所述计算资源的需求的服务器的地址信息，以及包括所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述任务卸载设备根据所述满足所述计算资源的需求的服务器与所述终端之间的距离，从所述满足所述计算资源的需求的服务器中，选择距离所述终端最近的服务器用于执行所述任务。
  6. 一种边缘计算服务器的寻找方法，其特征在于，所述方法应用于边缘计算系统，所述边缘计算系统包括终端、任务卸载设备、中心服务器以及多个边缘服务器，其中，所述任务卸载设备部署在所述终端接入的无线接入点中；所述方法包括：
    预设边缘服务器接收所述任务卸载设备发送的资源寻找请求，所述资源寻找请求携带有完成任务需要的计算资源的信息，所述预设边缘服务器为预设的在所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行所述任务的服务器；
    所述预设边缘服务器根据所述资源寻找请求确定所述预设边缘服务器不满足所述计算资源的需求；
    所述预设边缘服务器将所述资源寻找请求广播给其它边缘服务器和/或中心服务器。
  7. 根据权利要求6所述方法，其特征在于，所述预设边缘服务器将所述资源寻找请求广播给其它边缘服务器和/或中心服务器之后，还包括：
    所述预设边缘服务器接收所述其它边缘服务器和/或中心服务器中满足所述计算资源的需求的服务器发送的第二响应信息，所述第二响应信息包括所述满足所述计算资源的需求的服务器的地址信息和位置信息；
    所述预设边缘服务器将所述第二响应信息发送给所述任务卸载设备。
  8. 根据权利要求6所述方法，其特征在于，所述预设边缘服务器根据所述资源寻找请求确定所述预设边缘服务器不满足所述计算资源的需求之前，还包括：
    所述预设边缘服务器接收所述任务卸载设备发送的所述终端的位置信息；
    所述预设边缘服务器根据所述资源寻找请求确定所述预设边缘服务器不满足所述计算资源的需求之后，还包括：
    所述预设边缘服务器将所述终端的位置信息广播给所述其它边缘服务器和/或中心服务器。
  9. 根据权利要求8所述方法，其特征在于，所述预设边缘服务器将所述资源寻找请求和所述终端的位置信息广播给所述其它边缘服务器和/或中心服务器之后，还包括：
    所述预设边缘服务器接收所述其它边缘服务器和/或中心服务器中满足所述计算资源的需求的服务器发送的第三响应信息，所述第三响应信息包括所述满足所述计算资源的需求的服务器的地址信息，以及包括所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述预设边缘服务器将所述第三响应信息发送给所述任务卸载设备。
  10. 根据权利要求6至9任一项所述方法，其特征在于，所述预设边缘服务器接收所述任务卸载设备发送的资源寻找请求之后，还包括：
    所述预设边缘服务器根据所述资源寻找请求确定所述预设边缘服务器满足所述计算资源的需求；
    所述预设边缘服务器向所述任务卸载设备发送所述预设边缘服务器的地址信息。
  11. 一种任务卸载设备，其特征在于，所述任务卸载设备为边缘计算系统中的设备，所述边缘计算系统还包括终端、中心服务器以及多个边缘服务器；所述任务卸载设备包括：
    接收单元，用于接收所述终端发送的资源寻找请求，所述资源寻找请求携带有完成任务需要的计算资源的信息，其中，所述任务卸载设备部署在所述终端接入的无线接入点中；
    确定单元，用于根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内有满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器；
    选择单元，用于从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择出用于执行所述任务的服务器。
  12. 根据权利要求11所述任务卸载设备，其特征在于，所述选择单元具体用于：
    将所述资源寻找请求发送给满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器；
    接收满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器根据所述资源寻找请求发送的地址信息；
    根据接收的所述地址信息的先后顺序从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行所述任务的服务器。
  13. 根据权利要求12所述方任务卸载设备，其特征在于，所述任务卸载设备还包 括发送单元，用于在所述选择单元根据接收的所述地址信息的先后顺序从满足所述计算资源的需求的中心服务器和/或至少一个边缘服务器中选择用于执行所述任务的服务器之后，向所述终端发送响应所述资源寻找请求的第一响应信息；
    所述接收单元，还用于接收所述终端根据所述第一响应信息发送的所述任务；
    所述发送单元，还用于将所述任务发送给所述选择出的服务器。
  14. 根据权利要求11至13任一项所述任务卸载设备，其特征在于，所述任务卸载设备还包括发送单元和计算单元；
    所述确定单元，还用于根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器；
    所述发送单元，用于将所述资源寻找请求发送给预设边缘服务器；
    所述接收单元，还用于接收所述预设边缘服务器发送的第二响应信息，所述第二响应信息为接收到所述预设边缘服务器广播的所述资源寻找请求的服务器中，满足所述计算资源的需求的服务器向所述预设边缘服务器发送的，所述第二响应信息包括所述满足所述计算资源的需求的服务器的地址信息和位置信息；
    所述计算单元，用于根据所述满足所述计算资源的需求的服务器的位置信息和所述终端的位置信息，计算所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述选择单元，还用于根据所述计算得到的距离，从所述满足所述计算资源的需求的服务器中，选择距离所述终端最近的服务器用于执行所述任务。
  15. 根据权利要求11至13任一项所述任务卸载设备，其特征在于，所述任务卸载设备还包括发送单元；
    所述确定单元，还用于根据所述资源寻找请求确定所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器；
    所述发送单元，用于将所述资源寻找请求和所述终端的位置信息发送给预设边缘服务器；
    所述接收单元，还用于接收所述预设边缘服务器发送的第三响应信息，所述第三响应信息为接收到所述预设边缘服务器广播的所述资源寻找请求和所述终端的位置信息的服务器中，满足所述计算资源的需求的服务器向所述预设边缘服务器发送的，所述第三响应信息包括所述满足所述计算资源的需求的服务器的地址信息，以及包括所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述选择单元，还用于根据所述满足所述计算资源的需求的服务器与所述终端之间的距离，从所述满足所述计算资源的需求的服务器中，选择距离所述终端最近的服务器用于执行所述任务。
  16. 一种边缘服务器，其特征在于，所述边缘服务器为边缘计算系统中的服务器，所述边缘计算系统还包括终端、任务卸载设备、中心服务器以及除了所述边缘服务器之外的多个边缘服务器，其中，所述任务卸载设备部署在所述终端接入的无线接入点 中；所述边缘服务器包括：
    接收单元，用于接收所述任务卸载设备发送的资源寻找请求，所述资源寻找请求携带有完成任务需要的计算资源的信息，所述边缘服务器为预设的在所述终端接入的无线接入点覆盖的空间范围内没有满足所述计算资源的需求的中心服务器或边缘服务器的情况下，用于寻找其它满足需求的服务器来执行所述任务的服务器；
    确定单元，用于根据所述资源寻找请求确定所述边缘服务器不满足所述计算资源的需求；
    广播单元，用于将所述资源寻找请求广播给其它边缘服务器和/或中心服务器。
  17. 根据权利要求16所述边缘服务器，其特征在于，所述边缘服务器还包括发送单元；
    所述接收单元，还用于在所述广播单元将所述资源寻找请求广播给所述其它边缘服务器和/或中心服务器之后，接收所述其它边缘服务器和/或中心服务器中满足所述计算资源的需求的服务器发送的第二响应信息，所述第二响应信息包括所述满足所述计算资源的需求的服务器的地址信息和位置信息；
    所述发送单元，用于将所述第二响应信息发送给所述任务卸载设备。
  18. 根据权利要求16所述边缘服务器，其特征在于，
    所述接收单元，还用于在所述确定单元根据所述资源寻找请求确定所述边缘服务器不满足所述计算资源的需求之前，接收所述任务卸载设备发送的所述终端的位置信息；
    所述广播单元，还用于在所述确定单元根据所述资源寻找请求确定所述边缘服务器不满足所述计算资源的需求之后，将所述终端的位置信息广播给所述其它边缘服务器和/或中心服务器。
  19. 根据权利要求18所述边缘服务器，其特征在于，所述边缘服务器还包括发送单元；
    所述接收单元，还用于在所述广播单元将所述资源寻找请求和所述终端的位置信息广播给所述其它边缘服务器和/或中心服务器之后，接收所述其它边缘服务器和/或中心服务器中满足所述计算资源的需求的服务器发送的第三响应信息，所述第三响应信息包括所述满足所述计算资源的需求的服务器的地址信息，以及包括所述满足所述计算资源的需求的服务器与所述终端之间的距离；
    所述发送单元，用于将所述第三响应信息发送给所述任务卸载设备。
  20. 根据权利要求16至19任一项所述边缘服务器，其特征在于，所述边缘服务器还包括发送单元；
    所述确定单元，还用于在所述接收单元接收所述任务卸载设备发送的所述资源寻找请求之后，根据所述资源寻找请求确定所述边缘服务器满足所述计算资源的需求；
    所述发送单元，用于向所述任务卸载设备发送所述边缘服务器的地址信息。
  21. 一种任务卸载设备，其特征在于，包括处理器、存储器和通信接口，所述存储器存储有计算机程序，所述处理器用于调用所述计算机程序来执行如权利要求1至5任一项所述的方法。
  22. 一种边缘服务器，其特征在于，包括处理器、存储器和通信接口，所述存储器存储有计算机程序，所述处理器用于调用所述计算机程序来执行如权利要求6至10任一项所述的方法。
  23. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现如权利要求1至5任一项所述的方法。
  24. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现如权利要求6至10任一项所述的方法。
PCT/CN2020/104835 2019-08-07 2020-07-27 边缘计算服务器的寻找方法及相关设备 WO2021023042A1 (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201910727350.0 2019-08-07
CN201910727350 2019-08-07
CN201911007983.0 2019-10-22
CN201911007983.0A CN112351055B (zh) 2019-08-07 2019-10-22 边缘计算服务器的寻找方法及相关设备

Publications (1)

Publication Number Publication Date
WO2021023042A1 true WO2021023042A1 (zh) 2021-02-11

Family

ID=74367792

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/104835 WO2021023042A1 (zh) 2019-08-07 2020-07-27 边缘计算服务器的寻找方法及相关设备

Country Status (2)

Country Link
CN (1) CN112351055B (zh)
WO (1) WO2021023042A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113326126A (zh) * 2021-05-28 2021-08-31 湘潭大学 任务处理方法、任务调度方法、装置及计算机设备
CN114047965A (zh) * 2021-10-12 2022-02-15 润联软件系统(深圳)有限公司 计算卸载方法、卫星服务器及计算机可读存储介质
CN115946132A (zh) * 2023-03-15 2023-04-11 江西丹巴赫机器人股份有限公司 一种智能密集库与物流机器人系统及其故障监测方法
WO2024056993A1 (en) * 2022-09-15 2024-03-21 Arm Limited Computational resource discovery for offloading processing upon request

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115134416B (zh) * 2021-03-22 2023-04-25 中国联合网络通信集团有限公司 虚拟现实业务处理系统和方法
CN113242319A (zh) * 2021-07-09 2021-08-10 军事科学院系统工程研究院网络信息研究所 一种基于视频云服务架构的云边协同工作方法和系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017121478A1 (en) * 2016-01-14 2017-07-20 Sony Mobile Communications Inc. User equipment selection for mobile edge computing
CN108920279A (zh) * 2018-07-13 2018-11-30 哈尔滨工业大学 一种多用户场景下的移动边缘计算任务卸载方法
CN110062026A (zh) * 2019-03-15 2019-07-26 重庆邮电大学 移动边缘计算网络中资源分配和计算卸载联合优化方案

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6627289B2 (ja) * 2015-07-06 2020-01-08 日本電気株式会社 無線アクセスネットワークノード、エッジサーバ、ポリシ管理ノード、及びこれらの方法
US10484451B2 (en) * 2016-10-18 2019-11-19 Huawei Technologies Co., Ltd. Virtual network state management in mobile edge computing
CN108494612B (zh) * 2018-01-19 2021-06-08 西安电子科技大学 一种提供移动边缘计算服务的网络系统及其服务方法
CN108901046B (zh) * 2018-06-14 2020-10-09 北京大学 面向移动边缘计算的协同任务卸载算法及系统设计方案

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017121478A1 (en) * 2016-01-14 2017-07-20 Sony Mobile Communications Inc. User equipment selection for mobile edge computing
CN108920279A (zh) * 2018-07-13 2018-11-30 哈尔滨工业大学 一种多用户场景下的移动边缘计算任务卸载方法
CN110062026A (zh) * 2019-03-15 2019-07-26 重庆邮电大学 移动边缘计算网络中资源分配和计算卸载联合优化方案

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
CAO YUTING; CHEN HAOPENG; JIANG JIANWEI; HU FEI: "TaSRD: Task Scheduling Relying on Resource and Dependency in Mobile Edge Computing", 2018 IEEE INTERNATIONAL CONFERENCE ON PROGRESS IN INFORMATICS AND COMPUTING (PIC), IEEE, 14 December 2018 (2018-12-14), pages 287 - 295, XP033545181, DOI: 10.1109/PIC.2018.8706333 *
INTEL, CEA, PANASONIC, HHI: "Use cases for mmWave and Mobile Edge computing, two building blocks of 5G networks: the vision of the European funded projects 5G-MiEdge", 3GPP DRAFT; S1-172127 DISCUSSION 5G-MIEDGE, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. SA WG1, no. Porto, Portugal; 20170508 - 20170512, 7 May 2017 (2017-05-07), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP051259619 *
KAI XIONG, SUPENG LENG, KE ZHANG, HAO LIU: "Research on heterogeneous radio access and resource allocation algorithm in vehicular fog computing", CHINESE JOURNAL ON INTERNET OF THINGS, vol. 3, no. 2, 30 June 2019 (2019-06-30), pages 20 - 27, XP055777645, DOI: 10.11959/j.issn.2096−3750.2019.00108 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113326126A (zh) * 2021-05-28 2021-08-31 湘潭大学 任务处理方法、任务调度方法、装置及计算机设备
CN113326126B (zh) * 2021-05-28 2024-04-05 湘潭大学 任务处理方法、任务调度方法、装置及计算机设备
CN114047965A (zh) * 2021-10-12 2022-02-15 润联软件系统(深圳)有限公司 计算卸载方法、卫星服务器及计算机可读存储介质
WO2024056993A1 (en) * 2022-09-15 2024-03-21 Arm Limited Computational resource discovery for offloading processing upon request
CN115946132A (zh) * 2023-03-15 2023-04-11 江西丹巴赫机器人股份有限公司 一种智能密集库与物流机器人系统及其故障监测方法

Also Published As

Publication number Publication date
CN112351055A (zh) 2021-02-09
CN112351055B (zh) 2022-06-10

Similar Documents

Publication Publication Date Title
WO2021023042A1 (zh) 边缘计算服务器的寻找方法及相关设备
US9906464B2 (en) Optimization of multimedia service over an IMS network
CN108270813B (zh) 一种异构多协议栈方法、装置及系统
US20060143266A1 (en) Communication system
US20130254261A1 (en) System and Method of Managing Servers for Streaming Desktop Applications
CN115039391A (zh) 用于提供边缘计算服务的方法和装置
EP2972854B1 (en) Distributed data center technology
CN107135268B (zh) 基于信息中心网络的分布式任务计算方法
KR20210119504A (ko) 통신 방법 및 장치, 엔티티 및 컴퓨터 판독가능 저장 매체
CN111770535A (zh) 一种基于意图的网络配置方法、装置及系统
WO2021223454A1 (zh) 投屏方法和系统
US20200136926A1 (en) Dynamic adaptive network
CN112399484A (zh) 数据传输方法、装置、存储介质及电子设备
CN105656978A (zh) 一种资源共享方法及装置
WO2017034933A1 (en) Preferred network information
CN112714146B (zh) 一种资源调度方法、装置、设备及计算机可读存储介质
CN109981778B (zh) 内容分发网络的服务实现方法、装置、设备及存储介质
CN111245878B (zh) 一种基于混合云计算和雾计算的计算卸载通信网络的方法
CN113783963B (zh) 数据传输方法、服务器节点、网关设备、网络系统
US20140341033A1 (en) Transmission management device, system, and method
CN116170509A (zh) 算力调度方法、装置及存储介质
CN113366814B (zh) 对边缘计算系统中的资源分配进行管理的方法
CN113791899A (zh) 一种面向移动Web增强现实的边缘服务器管理系统
US20240111599A1 (en) Code execution on a distributed unit
WO2024021517A1 (zh) 网络协同处理方法、系统、装置及相关设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20850265

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20850265

Country of ref document: EP

Kind code of ref document: A1