TWI752401B - Network communication control method of multiple edge clouds and edge computing device and system - Google Patents

Network communication control method of multiple edge clouds and edge computing device and system Download PDF

Info

Publication number
TWI752401B
TWI752401B TW109100213A TW109100213A TWI752401B TW I752401 B TWI752401 B TW I752401B TW 109100213 A TW109100213 A TW 109100213A TW 109100213 A TW109100213 A TW 109100213A TW I752401 B TWI752401 B TW I752401B
Authority
TW
Taiwan
Prior art keywords
edge computing
information
service
target service
computing device
Prior art date
Application number
TW109100213A
Other languages
Chinese (zh)
Other versions
TW202123731A (en
Inventor
王韋程
Original Assignee
財團法人工業技術研究院
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
Priority claimed from US16/714,200 external-priority patent/US10771569B1/en
Application filed by 財團法人工業技術研究院 filed Critical 財團法人工業技術研究院
Publication of TW202123731A publication Critical patent/TW202123731A/en
Application granted granted Critical
Publication of TWI752401B publication Critical patent/TWI752401B/en

Links

Images

Landscapes

  • Mobile Radio Communication Systems (AREA)
  • Telephonic Communication Services (AREA)

Abstract

A network communication control method of multiple edge clouds comprises by a first edge computing platform, performing: acquiring operational information of a second edge computing platform, providing a terminal device with a target service, determining that there is a service abnormal situation, re-allocating the target service to the second edge computing platform and generating offload information and a target service location according to the service abnormal situation, the target service and the operational information of the second edge computing platform, transmitting the offload information to the second edge computing platform, and transmitting the target service location to the terminal device, in order to make the terminal device have a communication connection with the second edge computing platform according to the target service location. The offload information comprises the identity of the terminal device and the identity of the target service at least.

Description

多邊緣雲之網路通訊控制方法及邊緣運算裝置與系統Multi-edge cloud network communication control method and edge computing device and system

本揭露係關於一種網路通訊控制方法，特別係關於一種多邊緣雲之網路通訊控制方法。The present disclosure relates to a network communication control method, in particular, to a network communication control method of a multi-edge cloud.

隨著行動裝置的普及與網路服務的廣泛應用，諸如影音串流、虛擬實境、自動駕駛等，另加上各行業將業務服務擴展為網路服務的行為日益漸增，行動網路流量的需求亦持續地增加，因此一般回程（Backhaul）網路的負載流量已不足使用，遠端的雲端運算中心也無法負荷龐大的運算量。以往的網路架構在通訊技術日漸進步的情況下面臨了相當大的挑戰。With the popularization of mobile devices and the wide application of network services, such as video streaming, virtual reality, automatic driving, etc., and the increasing behavior of various industries to expand business services into network services, mobile network traffic The demand for cloud computing has also continued to increase. Therefore, the load traffic of the general backhaul network is insufficient, and the remote cloud computing center cannot handle the huge computing volume. The previous network architecture has faced considerable challenges with the advancement of communication technology.

為因應上述情況，行動邊緣運算（Mobile Edge Computing，MEC）的架構被提出，以期能夠解決流量需求及運算量日漸增加的問題。如字面文義，行動邊緣運算係在行動網路的邊緣提供雲端運算的能力與資訊科技（Information Technology，IT）服務的環境，主要概念在於減少核心網路設備的運算壓力，且提供行動營運商為顧客創造特定行動體驗的平台。然而，目前行動邊緣運算架構在遭遇異常時，往往需要重新佈建整個通訊環境，而耗費大量的時間，進而致使用戶的體驗感受不佳。In response to the above situation, the architecture of Mobile Edge Computing (MEC) is proposed in order to solve the problem of increasing traffic demand and computing volume. Literally, mobile edge computing is an environment that provides cloud computing capabilities and information technology (Information Technology, IT) services at the edge of the mobile network. The main concept is to reduce the computing pressure on core network equipment, and provide mobile operators for A platform for customers to create specific mobile experiences. However, when the current mobile edge computing architecture encounters an abnormality, it often needs to reconfigure the entire communication environment, which takes a lot of time, resulting in poor user experience.

鑒於上述，本揭露提供一種多邊緣雲之網路通訊控制方法及邊緣運算系統。In view of the above, the present disclosure provides a multi-edge cloud network communication control method and an edge computing system.

依據本揭露一實施例的多邊緣雲之網路通訊控制方法，適用於第一邊緣運算平台及第二邊緣運算平台，其中第一邊緣運算平台與第二邊緣運算平台進行資訊傳遞。所述網路通訊控制方法包含以第一邊緣運算平台執行：取得第二邊緣運算平台的運算資訊，提供終端裝置目標服務，判斷發生服務異常狀況，依據服務異常狀況、目標服務及第二邊緣運算平台的運算資訊，將目標服務重新配置至第二邊緣運算平台並產生卸載資訊及目標服務位置，將卸載資訊傳送至第二邊緣運算平台，以及將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至第二邊緣運算平台。其中，卸載資訊至少包含終端裝置的身分資訊及目標服務的身分資訊。The network communication control method for multi-edge cloud according to an embodiment of the present disclosure is applicable to a first edge computing platform and a second edge computing platform, wherein the first edge computing platform and the second edge computing platform perform information transmission. The network communication control method includes executing on the first edge computing platform: obtaining computing information of the second edge computing platform, providing target services of the terminal device, judging the occurrence of abnormal service conditions, and computing according to the abnormal service conditions, the target service and the second edge computing platform. The computing information of the platform, the target service is reconfigured to the second edge computing platform and the offload information and the target service location are generated, the offload information is sent to the second edge computing platform, and the target service location is sent to the terminal device, so that the terminal device according to The target service location is communicatively connected to the second edge computing platform. Wherein, the uninstall information includes at least the identity information of the terminal device and the identity information of the target service.

依據本揭露另一實施例的多邊緣雲之網路通訊控制方法，適用於第一邊緣運算平台、第二邊緣運算平台及中控平台，其中中控平台與第一邊緣運算平台及第二邊緣運算平台進行資訊傳遞。所述網路通訊控制方法包含：以第一邊緣運算平台提供一終端裝置一目標服務；以中控平台或第一邊緣運算平台執行多個步驟，所述多個步驟包含：判斷第一邊緣運算平台發生服務異常狀況；依據服務異常狀況、目標服務及第二邊緣運算平台的運算資訊，將目標服務重新配置至第二邊緣運算平台，並產生卸載資訊及一目標服務位置，以及將卸載資訊傳送至第二邊緣運算平台；以及以第一邊緣運算平台或第二邊緣運算平台將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至第二邊緣運算平台。其中卸載資訊至少包含終端裝置的身分資訊及目標服務的身分資訊。The network communication control method for multi-edge cloud according to another embodiment of the present disclosure is applicable to a first edge computing platform, a second edge computing platform, and a central control platform, wherein the central control platform, the first edge computing platform and the second edge computing platform The computing platform conducts information transmission. The network communication control method includes: using a first edge computing platform to provide a terminal device with a target service; using the central control platform or the first edge computing platform to execute multiple steps, the multiple steps include: judging the first edge computing A service abnormality occurs on the platform; according to the service abnormality, the target service and the computing information of the second edge computing platform, the target service is reconfigured to the second edge computing platform, and the offload information and a target service location are generated, and the offload information is transmitted. to the second edge computing platform; and using the first edge computing platform or the second edge computing platform to transmit the target service location to the terminal device, so that the terminal device communicates and connects to the second edge computing platform according to the target service location. The uninstall information includes at least the identity information of the terminal device and the identity information of the target service.

依據本揭露一實施例的邊緣運算裝置，包含記憶體及處理器，其中記憶體中儲存有多個指令，處理器電性連接於記憶體且執行記憶體中的指令以執行以下步驟：取得另一邊緣運算裝置的運算資訊，提供終端裝置目標服務，判斷發生服務異常狀況，依據服務異常狀況、目標服務及所述另一邊緣運算裝置的運算資訊，將目標服務重新配置至所述另一邊緣運算裝置並產生卸載資訊及目標服務位置，將卸載資訊傳送至所述另一邊緣運算裝置，以及將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至所述另一邊緣運算裝置。其中，卸載資訊至少包含終端裝置的身分資訊及目標服務的身分資訊。An edge computing device according to an embodiment of the present disclosure includes a memory and a processor, wherein a plurality of instructions are stored in the memory, and the processor is electrically connected to the memory and executes the instructions in the memory to perform the following steps: obtaining another The computing information of an edge computing device provides the target service of the terminal device, determines the occurrence of service abnormality, and reconfigures the target service to the other edge according to the service abnormality, the target service and the computing information of the other edge computing device The computing device generates offload information and a target service location, transmits the offload information to the other edge computing device, and transmits the target service location to the terminal device, so that the terminal device communicates with the other edge computing device according to the target service location device. Wherein, the uninstall information includes at least the identity information of the terminal device and the identity information of the target service.

依據本揭露一實施例的邊緣運算系統，包含上列實施例所述的邊緣運算裝置以及中控裝置。邊緣運算裝置用於執行第一網路通訊控制方法，所述第一網路通訊控制方法包含上列實施例所述之邊緣運算裝置的處理器執行該些指令所執行的該些步驟。中控裝置通訊連接於邊緣運算裝置，且用於執行第二網路通訊控制方法，所述第二網路通訊控制方法包含：判斷邊緣運算裝置發生服務異常狀況，依據服務異常狀況、目標服務及另一邊緣運算裝置的運算資訊，將目標服務重新配置至另一邊緣運算裝置並產生卸載資訊及目標服務位置，將卸載資訊傳送至另一邊緣運算裝置，以及使邊緣運算裝置或另一邊緣運算裝置將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至另一邊緣運算裝置。An edge computing system according to an embodiment of the present disclosure includes the edge computing device and the central control device described in the foregoing embodiments. The edge computing device is used for executing the first network communication control method, and the first network communication control method includes the steps performed by the processor of the edge computing device described in the above embodiments to execute the instructions. The central control device is communicatively connected to the edge computing device, and is used for executing a second network communication control method. The second network communication control method includes: judging that the edge computing device has a service abnormality, according to the service abnormality, the target service and the Computing information of another edge computing device, reconfigure the target service to another edge computing device and generate offload information and target service location, transmit the offload information to another edge computing device, and enable the edge computing device or another edge computing device The device transmits the target service location to the terminal device, so that the terminal device communicates with another edge computing device according to the target service location.

藉由上述結構，本案所揭示的多邊緣雲之網路通訊控制方法及邊緣運算系統可以因應多種服務異常狀況來執行提供服務之邊緣運算平台的重新配置，在遭遇服務異常狀況時無需重新佈建整個通訊環境，能夠具有服務不中斷之動態佈署能力，更可以據有負載平衡能力及異地備援能力。With the above structure, the multi-edge cloud network communication control method and edge computing system disclosed in this case can perform reconfiguration of the edge computing platform providing services in response to various service abnormal conditions, and do not need to be reconfigured when encountering service abnormal conditions. The entire communication environment can have dynamic deployment capabilities without service interruption, and can also have load balancing capabilities and remote backup capabilities.

以上之關於本揭露內容之說明及以下之實施方式之說明係用以示範與解釋本揭露之精神與原理，並且提供本揭露之專利申請範圍更進一步之解釋。The above description of the present disclosure and the following description of the embodiments are used to demonstrate and explain the spirit and principle of the present disclosure, and provide further explanations of the scope of the present disclosure.

以下在實施方式中詳細敘述本揭露之詳細特徵以及優點，其內容足以使任何熟習相關技藝者了解本揭露之技術內容並據以實施，且根據本說明書所揭露之內容、申請專利範圍及圖式，任何熟習相關技藝者可輕易地理解本揭露相關之目的及優點。以下之實施例係進一步詳細說明本揭露之觀點，但非以任何觀點限制本揭露之範疇。The detailed features and advantages of the present disclosure are described in detail below in the embodiments, and the content is sufficient to enable any person skilled in the relevant art to understand the technical content of the present disclosure and implement accordingly, and based on the content disclosed in this specification, the scope of the patent application and the drawings , any person skilled in the related art can easily understand the related objects and advantages of the present disclosure. The following examples further illustrate the point of the present disclosure in detail, but do not limit the scope of the present disclosure in any point of view.

本揭露提供一種邊緣運算系統，請參考圖1及圖2，其中圖1為依據本揭露一實施例所繪示的邊緣運算系統之通訊環境的示意圖，圖2則為依據本揭露一實施例所繪示的邊緣運算系統的功能方塊圖。如圖1所示，邊緣運算系統1包含第一邊緣運算裝置11及中控裝置13，兩者彼此通訊連接，邊緣運算系統1可以與終端裝置21通訊連接，亦可與第二邊緣運算裝置31通訊連接。進一步來說，第一邊緣運算裝置11可以通訊連接於終端裝置21，中控裝置13可以通訊連接於第二邊緣運算裝置31。於此要特別說明的是，圖1所示之與邊緣運算系統1連接的終端裝置21及第二邊緣運算裝置31的數量僅為示例，邊緣運算系統1亦可同時與多個終端裝置21連接，且與多個第二邊緣運算裝置31連接。此外，第二邊緣運算裝置31’亦可與一或多個終端裝置21連接。The present disclosure provides an edge computing system, please refer to FIG. 1 and FIG. 2 , wherein FIG. 1 is a schematic diagram of a communication environment of an edge computing system according to an embodiment of the present disclosure, and FIG. 2 is a schematic diagram of an edge computing system according to an embodiment of the present disclosure. The functional block diagram of the edge computing system is shown. As shown in FIG. 1 , the edge computing system 1 includes a first edge computing device 11 and a central control device 13 , which are connected in communication with each other. The edge computing system 1 can be communicatively connected with the terminal device 21 or with the second edge computing device 31 . communication connection. Further, the first edge computing device 11 can be communicatively connected to the terminal device 21 , and the central control device 13 can be communicatively connected to the second edge computing device 31 . It should be noted here that the number of terminal devices 21 and second edge computing devices 31 connected to the edge computing system 1 shown in FIG. 1 is only an example, and the edge computing system 1 can also be connected to multiple terminal devices 21 at the same time. , and is connected to a plurality of second edge computing devices 31 . In addition, the second edge computing device 31' can also be connected to one or more terminal devices 21.

第一邊緣運算裝置11可以係行動邊緣運算（Mobile Edge Computing，MEC）雲，可簡稱為邊緣雲。第一邊緣運算裝置11用於提供終端裝置21一或多種服務。第一邊緣運算裝置11可以透過基地台連接於終端裝置21。以網路架構而言，第一邊緣運算裝置11位於網路邊緣而相對靠近用戶端之終端裝置21。詳細來說，網路架構可以雲端伺服器（Cloud server）或網際網路（Internet）為網路核心，終端裝置21則位於網路架構的外圍而經由網路連接至雲端伺服器或網際網路，所述的網路邊緣即可以指相對靠近終端裝置21所在之用户端的區域網路（Local Area Network, LAN）與網際網路的交界，但不以此為限。進一步來說，第一邊緣運算裝置11可以係具有運算能力的單個伺服器，亦可以由多個具有不同功能且彼此連接的伺服器所組成，上述之終端裝置21則例如為智慧型手機、平板、筆記型電腦等行動電子裝置，本揭露不予限制。The first edge computing device 11 may be a mobile edge computing (Mobile Edge Computing, MEC) cloud, which may be referred to as an edge cloud for short. The first edge computing device 11 is used to provide one or more services to the terminal device 21 . The first edge computing device 11 can be connected to the terminal device 21 through the base station. In terms of network architecture, the first edge computing device 11 is located at the edge of the network and is relatively close to the terminal device 21 of the client. Specifically, the network architecture may be a cloud server or the Internet as the core of the network, and the terminal device 21 is located at the periphery of the network architecture and is connected to the cloud server or the Internet through the network , the network edge may refer to the boundary between the local area network (LAN) and the Internet which is relatively close to the user end where the terminal device 21 is located, but it is not limited thereto. Further, the first edge computing device 11 may be a single server with computing capabilities, or may be composed of a plurality of servers with different functions and connected to each other. The above-mentioned terminal device 21 is, for example, a smart phone, a tablet , notebook computers and other mobile electronic devices, this disclosure is not limited.

更進一步來說，如圖2所示，第一邊緣運算裝置11可以包含用戶介面功能（User Plan Function，UPF）模組111、網域名稱服務（Domain Name Service，DNS）模組112、資源管理模組113、服務管理模組114、第一服務模組115及第二服務模組116。第一邊緣運算裝置11的用戶介面功能模組111可以分別與第一服務模組115及第二服務模組116透過N6介面連接，亦可以與基地台透過N3介面連接，再透過基地台連接於終端裝置21。網域名稱服務模組112可以回應來自終端裝置21接收到的服務位置詢問，並將對應之服務喚醒。舉例來說，當使用者透過終端裝置21發出“www.youtube.com”之服務位置的詢問，網域名稱服務模組112便會回應“10.14.188.xxx”至終端裝置21，並將YouTube服務喚醒。資源管理模組113可以對第一邊緣運算裝置11的資源狀況進行監測、整理及管理；服務管理模組114則可以監控第一服務模組115及第二服務模組116的運作狀態，監控服務運行狀況，並在發生錯誤時回報，亦可以啟動新的服務模組（對於此實施例而言，即第一及第二服務模組115及116以外的模組）；第一服務模組115及第二服務模組116則分別可以提供一或多個網路服務。Furthermore, as shown in FIG. 2 , the first edge computing device 11 may include a User Plan Function (UPF) module 111 , a Domain Name Service (DNS) module 112 , and a resource management module 112 . The module 113 , the service management module 114 , the first service module 115 and the second service module 116 . The user interface function module 111 of the first edge computing device 11 can be connected to the first service module 115 and the second service module 116 through the N6 interface, respectively, and can also be connected to the base station through the N3 interface, and then connected to the base station through the base station. Terminal device 21 . The domain name service module 112 can respond to the service location query received from the terminal device 21 and wake up the corresponding service. For example, when the user sends an inquiry about the service location of "www.youtube.com" through the terminal device 21, the domain name service module 112 will respond "10.14.188.xxx" to the terminal device 21, and send YouTube Service wakes up. The resource management module 113 can monitor, organize and manage the resource status of the first edge computing device 11; the service management module 114 can monitor the operation status of the first service module 115 and the second service module 116, and monitor the service operating status, and report when an error occurs, and can also start a new service module (for this embodiment, that is, modules other than the first and second service modules 115 and 116); the first service module 115 and the second service module 116 can respectively provide one or more network services.

上述之模組可以各為獨立設置且彼此連接的伺服器，或者為運行於一或多個伺服器中的多個應用程序，本揭露並不限制運行上述模組之實體伺服器的數量。另外，於此要特別說明的是，圖2示例性地繪示第一邊緣運算裝置11包含兩個服務模組（第一服務模組115及第二服務模組116），然而第一邊緣運算裝置11亦可僅包含一個或兩個以上的服務模組，且服務模組的致能狀態可以動態地調整，其中所述致能狀態例如指示是否能提供服務。藉由第一邊緣運算裝置11的服務模組提供服務給終端裝置21，可以使第一邊緣運算裝置11進行較為複雜的運算，降低終端裝置21的運算負擔。另外，第一邊緣運算裝置11亦可提供儲存或處理終端裝置21有可能存取的資訊的服務。上述服務僅為舉例說明，然而第一邊緣運算裝置11所能提供的服務並不限於此。The above-mentioned modules may each be independently set up and connected servers, or may be multiple application programs running on one or more servers. The present disclosure does not limit the number of physical servers running the above-mentioned modules. In addition, it should be noted here that FIG. 2 exemplarily shows that the first edge computing device 11 includes two service modules (the first service module 115 and the second service module 116 ), but the first edge computing The device 11 may also include only one or more than two service modules, and the enabling status of the service modules may be dynamically adjusted, wherein the enabling status, for example, indicates whether the service can be provided. By providing services to the terminal device 21 by the service module of the first edge computing device 11 , the first edge computing device 11 can perform relatively complex operations, thereby reducing the computing burden of the terminal device 21 . In addition, the first edge computing device 11 may also provide a service for storing or processing information that the terminal device 21 may access. The above-mentioned services are only examples, but the services provided by the first edge computing device 11 are not limited thereto.

中控裝置13例如為協調流程器（Orchestrator），特別係5G核心的資料中心，用於與第一邊緣運算裝置11及第二邊緣運算裝置31進行資訊傳遞。其中，第二邊緣運算裝置31可以具有與上述第一邊緣運算裝置11一樣的架構及功能。第二邊緣運算裝置31可以係行動邊緣運算雲，用於提供終端裝置一或多個服務。進一步來說，第二邊緣運算裝置31可以係具有運算能力的單個伺服器，亦可以由多個具有不同功能且彼此連接的伺服器所組成。第二邊緣運算裝置31亦可以包含如前所述之用戶介面功能模組、網域名稱服務模組、資源管理模組、服務管理模組及一或多個服務模組。第二邊緣運算裝置31與第一邊緣運算裝置11的差別可在於資源狀況（例如多寡）、所能提供的服務的數量與種類，以及提供服務至特定終端裝置所產生的延遲（Latency）。The central control device 13 is, for example, an orchestrator, especially a 5G core data center, and is used to communicate information with the first edge computing device 11 and the second edge computing device 31 . The second edge computing device 31 may have the same structure and function as the first edge computing device 11 described above. The second edge computing device 31 may be a mobile edge computing cloud for providing one or more services to the terminal device. Further, the second edge computing device 31 may be a single server with computing capability, or may be composed of a plurality of servers having different functions and connected to each other. The second edge computing device 31 may also include a user interface function module, a domain name service module, a resource management module, a service management module and one or more service modules as described above. The difference between the second edge computing device 31 and the first edge computing device 11 may lie in resource status (eg, amount), the number and type of services that can be provided, and the latency (Latency) caused by providing services to a specific terminal device.

另外，中控裝置13除了與第一及第二邊緣運算裝置11及31通訊連接，亦可與更多的邊緣運算裝置通訊連接，以收集各邊緣運算裝置的資訊，例如裝置身分標識（ID）、裝置位置（例如邊緣運算裝置之IP）、所供服務ID、所供服務位置（例如對應服務模組之IP）、所連終端裝置ID、資源狀況、延遲等資訊，特別係週期性地收集。中控裝置13可以依據所收集的資訊來配置多個終端裝置與多個邊緣運算裝置之間的通訊連接關係。進一步來說，本揭露所提之中控裝置13可以依據終端裝置之需求將終端裝置分配給特定的邊緣運算裝置，也可以依據邊緣運算裝置之狀況進行重新配置，其中更進一步的實施方法將於後描述。In addition, the central control device 13 not only communicates with the first and second edge computing devices 11 and 31 , but also communicates with more edge computing devices to collect information on each edge computing device, such as device identification (ID) , device location (such as the IP of the edge computing device), service ID provided, service location provided (such as the IP of the corresponding service module), connected terminal device ID, resource status, delay and other information, especially collected periodically . The central control device 13 can configure the communication connection relationship between the plurality of terminal devices and the plurality of edge computing devices according to the collected information. Further, the central control device 13 proposed in the present disclosure can allocate the terminal device to a specific edge computing device according to the needs of the terminal device, and can also be reconfigured according to the status of the edge computing device, and further implementation methods will be described in the described later.

更進一步來說明中控裝置13的架構，如圖2所示，中控裝置13可以包含用戶介面功能模組131、網域名稱服務模組132、服務位置管理模組133、服務重新導向模組134及資源管理模組135。中控裝置13的用戶介面功能模組131可以與第一邊緣運算裝置11的用戶介面功能模組111可以透過N9介面連接，同理，用戶介面功能模組131亦可以與其他邊緣運算裝置透過N9介面連接。網域名稱服務模組132則具有同於前述之網域名稱服務模組112的功能，於此不再贅述。服務位置管理模組133可以具有資料庫，記錄中控裝置13所連之邊緣運算裝置各所能夠提供的服務，其中記錄方式例如為服務ID，並且服務位置管理模組133可以與服務重新導向模組134或資源管理模組135協同執行服務的重新配置，其中，重新配置的詳細實施內容將於後描述。上述模組可以各為獨立設置且彼此連接的伺服器，或者為運行於一或多個伺服器中的多個應用程序，本揭露並不限制運行上述模組之實體伺服器的數量。換句話說，中控裝置13可以為單個伺服器來實行，亦可以由多個具有不同功能的伺服器所組成。於另一實施例中，中控裝置13本身亦可以係具有中控功能的行動邊緣運算雲，可以透過基地台與終端裝置21連接並提供終端裝置21服務。於此實施例中，除了前述多個模組，中控裝置13更可以包含服務管理模組及一或多個服務模組。To further illustrate the structure of the central control device 13, as shown in FIG. 2, the central control device 13 may include a user interface function module 131, a domain name service module 132, a service location management module 133, and a service redirection module 134 and resource management module 135. The user interface function module 131 of the central control device 13 can be connected to the user interface function module 111 of the first edge computing device 11 through the N9 interface. Similarly, the user interface function module 131 can also be connected to other edge computing devices through the N9 interface. interface connection. The domain name service module 132 has the same functions as the above-mentioned domain name service module 112, and details are not described herein again. The service location management module 133 may have a database to record the services provided by the edge computing devices connected to the central control device 13, wherein the recording method is, for example, the service ID, and the service location management module 133 can be associated with the service redirection module. 134 or the resource management module 135 cooperate to perform service reconfiguration, wherein the detailed implementation content of the reconfiguration will be described later. The above-mentioned modules may each be independently set up and connected servers, or may be multiple application programs running on one or more servers. The present disclosure does not limit the number of physical servers running the above-mentioned modules. In other words, the central control device 13 may be implemented by a single server, or may be composed of a plurality of servers with different functions. In another embodiment, the central control device 13 itself may also be a mobile edge computing cloud with a central control function, which can be connected to the terminal device 21 through the base station and provide services for the terminal device 21 . In this embodiment, in addition to the aforementioned multiple modules, the central control device 13 may further include a service management module and one or more service modules.

本揭露亦提供一種多邊緣雲之網路通訊控制方法，請一併參考圖1～圖3，其中圖3係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的流程圖。如圖3所示，多邊緣雲之網路通訊控制方法可以包含步驟S11：以第一邊緣運算平台提供終端裝置一目標服務；步驟S12：以中控平台判斷第一邊緣運算平台發生服務異常狀況；步驟S13：以中控平台依據服務異常狀況、目標服務及第二邊緣運算平台的運算資訊，將目標服務重新配置至第二邊緣運算平台，並產生卸載（Offload）資訊及目標服務位置；步驟S14：以中控平台將卸載資訊傳送至第二邊緣運算平台；以及步驟S15：以第一邊緣運算平台或第二邊緣運算平台將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至第二邊緣運算平台。The present disclosure also provides a method for controlling network communication in a multi-edge cloud. Please refer to FIGS. 1 to 3 together. FIG. 3 is a flow chart of a method for controlling network communication in a multi-edge cloud according to an embodiment of the present disclosure. picture. As shown in FIG. 3 , the network communication control method of the multi-edge cloud may include step S11 : providing a target service of the terminal device with the first edge computing platform; step S12 : judging the abnormal service condition of the first edge computing platform with the central control platform ; Step S13: The central control platform reconfigures the target service to the second edge computing platform according to the abnormal condition of the service, the target service and the computing information of the second edge computing platform, and generates offload (Offload) information and the target service location; step S14: Use the central control platform to transmit the offloading information to the second edge computing platform; and Step S15: Use the first edge computing platform or the second edge computing platform to transmit the target service location to the terminal device, so that the terminal device communicates according to the target service location Connect to the second edge computing platform.

於一實施例中，上述第一邊緣運算平台及第二邊緣運算平台及中控平台分別以前述圖1及圖2之第一邊緣運算裝置11、第二邊緣運算裝置31及中控裝置13來實現；而於另一實施例中，第一邊緣運算平台、第二邊緣運算平台及中控平台可以分別係於伺服器中的虛擬機器（Virtual Machine）所運行之應用程序。以下將以第一邊緣運算裝置11、第二邊緣運算裝置31及中控裝置13所實現之第一邊緣運算平台、第二邊緣運算平台及中控平台來示例性地說明圖3所示之多邊緣雲之網路通訊控制方法。In one embodiment, the first edge computing platform, the second edge computing platform, and the central control platform are respectively implemented by the first edge computing device 11 , the second edge computing device 31 and the central control device 13 shown in FIG. 1 and FIG. 2 . In another embodiment, the first edge computing platform, the second edge computing platform, and the central control platform may be respectively an application program run by a virtual machine (Virtual Machine) in the server. In the following, the first edge computing platform, the second edge computing platform and the central control platform realized by the first edge computing device 11 , the second edge computing device 31 and the central control device 13 will be used as an example to illustrate the number shown in FIG. 3 . A network communication control method for edge cloud.

於步驟S11中，第一邊緣運算裝置11提供終端裝置21一目標服務，其中所述目標服務例如係人工智慧、虛擬實境、擴增實境、自動駕駛等應用服務，本揭露不予限制。進一步來說，第一邊緣運算裝置11與終端裝置21之間的通訊連結是中控裝置13先前依據第一邊緣運算裝置11的運算資訊所決定的，其中運算資訊包含資源狀況及第一邊緣運算裝置11提供所述目標服務之延遲。於步驟S12中，中控裝置13會判斷第一邊緣運算裝置11發生了服務異常狀況，其中所述服務異常狀況可能指示服務中止（Service Down）、服務移出（Service Swap Out）或平台失能（MEC Down）。為了因應服務異常狀況，於步驟S13～S15中，中控裝置13會進行重新配置以使另一邊緣運算裝置（第二邊緣運算裝置31）提供目標服務（即原第一邊緣運算裝置11所供之服務）給終端裝置21。In step S11 , the first edge computing device 11 provides the terminal device 21 with a target service, wherein the target service is, for example, application services such as artificial intelligence, virtual reality, augmented reality, and automatic driving, which are not limited in this disclosure. Further, the communication link between the first edge computing device 11 and the terminal device 21 is previously determined by the central control device 13 according to the computing information of the first edge computing device 11 , wherein the computing information includes the resource status and the first edge computing Delay for device 11 to provide the target service. In step S12, the central control device 13 determines that a service abnormality occurs in the first edge computing device 11, wherein the service abnormality may indicate a service down (Service Down), a service swap (Service Swap Out), or a platform failure ( MEC Down). In order to cope with abnormal service conditions, in steps S13 to S15 , the central control device 13 will be reconfigured so that another edge computing device (the second edge computing device 31 ) provides the target service (that is, the original first edge computing device 11 provides the target service). service) to the terminal device 21.

詳細說明對應於不同的服務異常狀況所執行之步驟S12～S15，請參考圖1～圖4，其中圖4係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。於服務異常狀況為服務中止或服務移出的運行狀態中，如圖4步驟A11所示，中控裝置13可以從第一邊緣運算裝置11接收到關聯於服務異常狀況的服務失效訊號，藉此便可判斷第一邊緣運算裝置11發生服務異常狀況。For a detailed description of steps S12 to S15 executed in response to different service abnormal conditions, please refer to FIG. 1 to FIG. 4 , wherein FIG. 4 is a signal of a multi-edge cloud network communication control method according to an embodiment of the present disclosure. Transfer schematic. In the running state where the service abnormality is service suspension or service removal, as shown in step A11 in FIG. 4 , the central control device 13 can receive a service failure signal related to the service abnormality from the first edge computing device 11 , thereby It can be determined that a service abnormality occurs in the first edge computing device 11 .

特別來說，服務中止可以係指原提供終端裝置21目標服務的服務模組失能而無法再繼續提供所述目標服務，此時服務管理模組114可以產生對應之服務失效訊號並提供至中控裝置13的服務位置管理模組133。另，服務移出則可以係指：在第一邊緣運算裝置11的資源管理模組113判斷資源不足時，服務管理模組114將優先序為後位者的服務移出，例如使對應的服務模組失能或改為提供其他服務，所述優先序可以依據實際所需設定，本揭露不予限制。當服務管理模組114決定移出之服務為原提供給終端裝置21的目標服務時，服務管理模組114便可以產生對應之服務失效訊號並提供給中控裝置13的服務位置管理模組133。所述對應於服務中止或移出之服務失效訊號可以包含第一邊緣運算裝置11的ID、目標服務的ID及原與發生異常之邊緣運算平台所連接的一或多個終端裝置21的ID。In particular, service suspension may refer to the failure of the service module that originally provided the target service of the terminal device 21 and cannot continue to provide the target service. At this time, the service management module 114 can generate a corresponding service failure signal and provide it to the the service location management module 133 of the control device 13 . In addition, service removal may refer to: when the resource management module 113 of the first edge computing device 11 determines that the resources are insufficient, the service management module 114 removes the service with the lower priority, for example, the corresponding service module Disable or change to provide other services, the priority can be set according to actual needs, which is not limited in this disclosure. When the service management module 114 determines that the removed service is the target service originally provided to the terminal device 21 , the service management module 114 can generate a corresponding service failure signal and provide it to the service location management module 133 of the central control device 13 . The service failure signal corresponding to service suspension or removal may include the ID of the first edge computing device 11 , the ID of the target service, and the IDs of one or more terminal devices 21 originally connected to the edge computing platform where the exception occurred.

當中控裝置13的服務位置管理模組133接收到服務失效訊號而得知第一邊緣運算裝置11發生服務異常狀況時，服務位置管理模組133及服務重新導向模組134可以據以進行重新配置。於此實施例中，所述重新配置包含重新定位（Re-locate）程序及重新導向（Re-direct）程序。其中，重新定位程序（步驟A12）可以包含：選擇能夠提供目標服務的邊緣運算裝置作為候選邊緣運算平台，以及依據候選邊緣運算平台的運算資訊決定目標邊緣運算平台。如前所述，中控裝置13可以週期性地收集所連接之一或多個邊緣運算裝置的資源狀況及所提供服務之延遲等資訊。進一步來說，中控裝置13的服務位置管理模組133可以從服務失效訊號取得目標服務ID，選擇包含具有目標服務ID之服務模組的邊緣運算裝置作為候選邊緣運算平台，接著從中選擇具有最佳運算資訊的邊緣運算裝置來作為目標邊緣運算平台，即預計接續提供目標服務給終端裝置21之邊緣運算平台。When the service location management module 133 of the central control device 13 receives the service failure signal and learns that the first edge computing device 11 has a service abnormality, the service location management module 133 and the service redirection module 134 can perform reconfiguration accordingly . In this embodiment, the reconfiguration includes a re-locate process and a redirection (Re-direct) process. The relocation procedure (step A12 ) may include: selecting an edge computing device capable of providing the target service as a candidate edge computing platform, and determining the target edge computing platform according to computing information of the candidate edge computing platform. As mentioned above, the central control device 13 may periodically collect information such as the resource status of one or more connected edge computing devices and the delay of the provided service. Further, the service location management module 133 of the central control device 13 can obtain the target service ID from the service failure signal, select the edge computing device including the service module with the target service ID as the candidate edge computing platform, and then select the edge computing platform with the most The edge computing device of Best Computing Information is used as the target edge computing platform, that is, the edge computing platform that is expected to continuously provide the target service to the terminal device 21 .

所述最佳運算資訊的判斷可以依目標服務的類型而有不同的標準。舉例來說，對於虛擬實境之服務而言，頻寬資源狀況可以作為主要考量因素，然而本揭露不以此為限。另外，有目標服務需求之終端裝置21的數量亦為重要考量因素之一，服務位置管理模組133可以從服務失效訊號所含之終端裝置21的ID得知有目標服務需求之終端裝置21的數量，判斷候選邊緣運算平台的資源狀況是否足夠提供目標服務給該數量的終端裝置21。特別來說，在行動邊緣運算環境中，提供目標服務之延遲可以作為首要考量因素，進一步來說，服務位置管理模組133可以從資源狀況符合一特定標準的候選邊緣運算平台中選擇具有最低延遲者作為目標邊緣運算平台。更進一步來說，所謂特定標準可以係指能夠提供目標服務給終端裝置21的最低門檻。The determination of the optimal operation information may have different criteria according to the type of the target service. For example, for the virtual reality service, the bandwidth resource status can be the main consideration factor, but the present disclosure is not limited to this. In addition, the number of terminal devices 21 with target service requirements is also one of the important considerations. The service location management module 133 can know the terminal device 21 with target service requirements from the ID of the terminal device 21 contained in the service failure signal. number, and determine whether the resource status of the candidate edge computing platform is sufficient to provide the target service to the number of terminal devices 21 . In particular, in a mobile edge computing environment, the delay in providing the target service can be a primary consideration. Further, the service location management module 133 can select the candidate edge computing platform with the lowest delay from the candidate edge computing platforms whose resource conditions meet a specific standard as the target edge computing platform. Furthermore, the so-called specific standard may refer to the minimum threshold for providing the target service to the terminal device 21 .

當服務位置管理模組133在執行上述重新定位程序而決定目標邊緣運算平台時，會產生重新導向要求並傳送給服務重新導向模組134（步驟A13）。所述重新導向要求包含目標邊緣運算平台的ID、目標服務ID以及終端裝置21的ID。以前述圖3之實施例而言，第二邊緣運算裝置31即被服務位置管理模組133選為目標邊緣運算平台，因此於此實施例中，重新導向要求會包含第二邊緣運算裝置31的ID。服務重新導向模組134依據重新導向要求將重新導向資訊傳送至第一邊緣運算裝置11（步驟A14），即前述之重新導向程序。其中重新導向資訊可以包含第二邊緣運算裝置31的ID、終端裝置21的ID及目標服務ID。進一步來說，第一邊緣運算裝置11的用戶介面功能模組111可以依據重新導向資訊來向第二邊緣運算裝置31的用戶介面功能模組提出針對終端裝置21的連結建立要求，所述連結之建立特別係透過N9介面建立終端裝置21的連結（Session Connection）。更進一步來說，第一邊緣運算裝置11在接收到重新導向資訊時會回傳一確認訊息（ACK）給服務重新導向模組134，以指示重新導向資訊傳遞成功。When the service location management module 133 determines the target edge computing platform by executing the relocation procedure, a redirection request is generated and sent to the service redirection module 134 (step A13 ). The redirection request includes the ID of the target edge computing platform, the target service ID, and the ID of the terminal device 21 . For the aforementioned embodiment of FIG. 3 , the second edge computing device 31 is selected as the target edge computing platform by the service location management module 133 . Therefore, in this embodiment, the redirection request will include the second edge computing device 31 . ID. The service redirection module 134 transmits the redirection information to the first edge computing device 11 according to the redirection request (step A14 ), which is the aforementioned redirection process. The redirection information may include the ID of the second edge computing device 31 , the ID of the terminal device 21 and the target service ID. Further, the user interface function module 111 of the first edge computing device 11 may request the user interface function module of the second edge computing device 31 to establish a connection for the terminal device 21 according to the redirection information, and the connection is established. In particular, the connection (Session Connection) of the terminal device 21 is established through the N9 interface. More specifically, when receiving the redirection information, the first edge computing device 11 will return an acknowledgment message (ACK) to the service redirection module 134 to indicate that the redirection information is successfully transmitted.

中控裝置13經上述程序將目標服務重新配置至第二邊緣運算裝置31，並且產生卸載資訊及目標服務位置，再分別將卸載資訊傳送至第二邊緣運算裝置31，將目標服務位置傳送至終端裝置21。進一步來說，中控裝置13的服務重新導向模組134將卸載資訊傳送至第二邊緣運算裝置31（步驟A15）。更進一步來說，服務重新導向模組134會等第一邊緣運算裝置11回傳對應於步驟A14的確認訊息後再執行步驟A15。於此實施例中，卸載資訊包含第一邊緣運算裝置11的ID、終端裝置21的ID以及目標服務ID。第二邊緣運算裝置31可依據卸載資訊同意前述第一邊緣運算裝置11之連結建立的要求。簡而言之，第一邊緣運算裝置11及第二邊緣運算裝置31可以分別從中控裝置13取得重新導向資訊及卸載資訊，據以與對方建立關於終端裝置21及目標服務的連結。另外，第二邊緣運算裝置31在接收到卸載資訊時，可以依據卸載資訊中的目標服務ID取得對應的服務模組的位置資訊（例如IP）作為目標服務位置，並回傳一確認訊息給服務重新導向模組134以指示卸載資訊傳遞成功。特別來說，此對應於步驟A15的確認訊息包含目標服務位置。The central control device 13 reconfigures the target service to the second edge computing device 31 through the above procedure, and generates the offload information and the target service location, and then respectively transmits the offload information to the second edge computing device 31, and transmits the target service location to the terminal. device 21. Further, the service redirection module 134 of the central control device 13 transmits the offload information to the second edge computing device 31 (step A15 ). More specifically, the service redirection module 134 will wait for the first edge computing device 11 to return a confirmation message corresponding to the step A14 before executing the step A15. In this embodiment, the offloading information includes the ID of the first edge computing device 11 , the ID of the terminal device 21 and the target service ID. The second edge computing device 31 can agree to the request for establishing the connection of the first edge computing device 11 according to the offload information. In short, the first edge computing device 11 and the second edge computing device 31 can obtain redirection information and uninstallation information from the central control device 13 respectively, so as to establish a connection with each other about the terminal device 21 and the target service. In addition, when receiving the uninstallation information, the second edge computing device 31 can obtain the location information (eg IP) of the corresponding service module according to the target service ID in the uninstallation information as the target service location, and return a confirmation message to the service Redirect module 134 to indicate that the uninstallation information was delivered successfully. Specifically, the confirmation message corresponding to step A15 includes the target service location.

中控裝置13的服務重新導向模組134另透過第一邊緣運算裝置11將目標服務位置傳送至終端裝置21。進一步來說，服務重新導向模組134可以等第二邊緣運算裝置31回傳對應於步驟A15的確認訊息後，從確認訊息取得目標服務位置，並將終端裝置21的ID及目標服務位置傳送至第一邊緣運算裝置11（步驟A16）。第一邊緣運算裝置11從中控裝置13取得目標服務位置，且可以依據終端裝置21的ID將目標服務位置傳送至終端裝置21（步驟A17）。特別來說，上述步驟A16及A17之資訊傳遞可以透過網域名稱服務通知（DNS Notify）、HTTP 301或HTTP 302之方式進行。終端裝置21在取得目標服務位置後，便可據以透過第一邊緣運算裝置11與第二邊緣運算裝置31之間的連結連接至第二邊緣運算裝置31中提供目標服務之服務模組（步驟A18及A19）。The service redirection module 134 of the central control device 13 further transmits the target service location to the terminal device 21 through the first edge computing device 11 . Further, the service redirection module 134 can wait for the second edge computing device 31 to return a confirmation message corresponding to step A15, obtain the target service location from the confirmation message, and send the ID of the terminal device 21 and the target service location to The first edge computing device 11 (step A16 ). The first edge computing device 11 obtains the target service location from the central control device 13, and can transmit the target service location to the terminal device 21 according to the ID of the terminal device 21 (step A17). In particular, the information transfer of the above steps A16 and A17 can be performed by means of DNS Notify, HTTP 301 or HTTP 302. After the terminal device 21 obtains the target service location, it can connect to the service module that provides the target service in the second edge computing device 31 through the connection between the first edge computing device 11 and the second edge computing device 31 (step A18 and A19).

除了上述之因應服務中止或服務移出的服務異常狀況，圖3所示之網路通訊控制方法亦可因應平台失能之服務異常狀況。請參考圖1～圖3及圖5，其中圖5係依據本揭露另一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。於圖5所示的實施例中，中控裝置13的資源管理模組135可以週期性地從第一邊緣運算裝置11取得其資訊。當資源管理模組135判斷經過一特定時間長度而未收到來自第一邊緣運算裝置11的資訊（步驟A21）時，便可以判斷第一邊緣運算裝置11失能。接著，中控裝置13可以進行重新配置。於此實施例中，資源管理模組135依據服務異常狀況產生重新定位要求並傳送至服務位置管理模組133（步驟A22）。進一步來說，所述重新定位要求可以包含第一邊緣運算裝置11的ID以及原與第一邊緣運算裝置11相連之終端裝置21的ID。In addition to the above-mentioned abnormal service conditions in response to service suspension or service removal, the network communication control method shown in FIG. 3 can also respond to service abnormal conditions in which the platform is disabled. Please refer to FIG. 1 to FIG. 3 and FIG. 5 , wherein FIG. 5 is a schematic diagram of signal transmission of a network communication control method of a multi-edge cloud according to another embodiment of the present disclosure. In the embodiment shown in FIG. 5 , the resource management module 135 of the central control device 13 may periodically obtain its information from the first edge computing device 11 . When the resource management module 135 determines that the information from the first edge computing device 11 has not been received after a certain period of time (step A21 ), it can determine that the first edge computing device 11 is disabled. Next, the central control device 13 can be reconfigured. In this embodiment, the resource management module 135 generates a relocation request according to the abnormal condition of the service and transmits it to the service location management module 133 (step A22 ). Further, the relocation request may include the ID of the first edge computing device 11 and the ID of the terminal device 21 originally connected to the first edge computing device 11 .

服務位置管理模組133依據重新定位要求而執行重新定位程序，以產生重新定位回覆並回傳至資源管理模組135（步驟A23）。如前所述，服務位置管理模組133可以記錄中控裝置13所連之邊緣運算裝置各所具有之服務的服務ID。因此，所述重新定位程序可以包含依據第一邊緣運算裝置11的ID取得第一邊緣運算裝置11原具有之一或多個服務，並將所述一或多個服務作為目標服務，而後續選擇候選邊緣運算平台及目標邊緣運算平台之實施方式皆同於前述圖4實施例，於此不再贅述。服務位置管理模組133產生之重新定位回覆可以包含目標邊緣運算平台的ID、目標服務ID以及終端裝置21的ID。以前述圖3之實施例而言，第二邊緣運算裝置31即被服務位置管理模組133選為目標邊緣運算平台，因此於此實施例中，重新定位回覆會包含第二邊緣運算裝置31的ID。The service location management module 133 executes the relocation procedure according to the relocation request, so as to generate a relocation response and send it back to the resource management module 135 (step A23 ). As mentioned above, the service location management module 133 can record the service IDs of the services possessed by the edge computing devices connected to the central control device 13 . Therefore, the relocation procedure may include obtaining one or more services originally possessed by the first edge computing device 11 according to the ID of the first edge computing device 11 , using the one or more services as the target service, and then selecting The implementations of the candidate edge computing platform and the target edge computing platform are the same as those of the aforementioned embodiment in FIG. 4 , and will not be repeated here. The relocation reply generated by the service location management module 133 may include the ID of the target edge computing platform, the target service ID, and the ID of the terminal device 21 . For the aforementioned embodiment of FIG. 3 , the second edge computing device 31 is selected as the target edge computing platform by the service location management module 133 . Therefore, in this embodiment, the relocation reply will include the information of the second edge computing device 31 . ID.

中控裝置13經上述程序將目標服務重新配置至第二邊緣運算裝置31，並產生卸載資訊及目標服務位置，再分別將卸載資訊傳送至第二邊緣運算裝置31，將目標服務位置傳送至終端裝置21。進一步來說，中控裝置13的資源管理模組135依據重新定位回覆產生包含終端裝置21的ID以及目標服務ID的卸載資訊並將卸載資訊傳送至第二邊緣運算裝置31（步驟A24）。如此一來，第二邊緣運算裝置31依據卸載資訊即可得知後續將要提供目標服務至終端裝置21，且第二邊緣運算裝置31在接收到卸載資訊時，可以依據卸載資訊中的目標服務ID取得對應的服務模組的位置資訊（例如IP）作為目標服務位置，並回傳一確認訊息給資源管理模組135以指示卸載資訊傳遞成功。特別來說，此確認訊息包含目標服務位置。The central control device 13 reconfigures the target service to the second edge computing device 31 through the above procedure, generates the offload information and the target service location, and then transmits the offload information to the second edge computing device 31 respectively, and transmits the target service location to the terminal. device 21. Further, the resource management module 135 of the central control device 13 generates the offload information including the ID of the terminal device 21 and the target service ID according to the relocation response, and transmits the offload information to the second edge computing device 31 (step A24 ). In this way, the second edge computing device 31 can know that the target service will be provided to the terminal device 21 in the future according to the uninstallation information, and when the second edge computing device 31 receives the uninstallation information, it can be based on the target service ID in the uninstallation information. The location information (eg IP) of the corresponding service module is obtained as the target service location, and a confirmation message is returned to the resource management module 135 to indicate that the uninstallation information is successfully delivered. Specifically, this confirmation message contains the target service location.

中控裝置13的資源管理模組135另透過第二邊緣運算裝置31將目標服務位置傳送至終端裝置21。進一步來說，資源管理模組135可以等第二邊緣運算裝置31回傳確認訊息後，從確認訊息取得目標服務位置，並將包含終端裝置21的ID及目標服務位置傳送至第二邊緣運算裝置31（步驟A25）。第二邊緣運算裝置31從中控裝置13取得目標服務位置，且可以依據終端裝置21的ID將目標服務位置傳送至終端裝置21（步驟A26）。特別來說，上述步驟A25及A26之資訊傳遞可以透過DNS Notify、HTTP 301或HTTP 302之方式進行。終端裝置21在取得目標服務位置後，便可據以連接至第二邊緣運算裝置31中提供目標服務之服務模組（步驟A27）。特別來說，於此實施例中，在執行邊緣運算平台轉換之前，終端裝置21即同時與第一邊緣運算裝置11及第二邊緣運算裝置31連線。詳細來說，終端裝置21所連接之基地台原先即分別與第一邊緣運算裝置11及第二邊緣運算裝置31有所連線。其中，第一邊緣運算裝置11作為主要邊緣運算平台，第二邊緣運算裝置31則作為備援邊緣運算平台，當主要邊緣運算平台故障時，備援邊緣運算平台便接管基地台。The resource management module 135 of the central control device 13 further transmits the target service location to the terminal device 21 through the second edge computing device 31 . Further, the resource management module 135 can obtain the target service location from the confirmation message after the second edge computing device 31 returns a confirmation message, and transmit the ID of the terminal device 21 and the target service location to the second edge computing device 31 (step A25). The second edge computing device 31 obtains the target service location from the central control device 13, and can transmit the target service location to the terminal device 21 according to the ID of the terminal device 21 (step A26). In particular, the information transfer of the above steps A25 and A26 can be performed by means of DNS Notify, HTTP 301 or HTTP 302. After obtaining the target service location, the terminal device 21 can connect to the service module in the second edge computing device 31 that provides the target service accordingly (step A27 ). In particular, in this embodiment, before performing the edge computing platform conversion, the terminal device 21 is simultaneously connected to the first edge computing device 11 and the second edge computing device 31 . Specifically, the base stations to which the terminal device 21 is connected are originally connected to the first edge computing device 11 and the second edge computing device 31 respectively. The first edge computing device 11 serves as the main edge computing platform, and the second edge computing device 31 serves as the backup edge computing platform. When the primary edge computing platform fails, the backup edge computing platform takes over the base station.

於一實施例中，在第一邊緣運算裝置11提供終端裝置目標服務（即圖3之步驟S11）之前，多邊緣雲之網路通訊控制方法更可以包含邊緣運算平台裝置註冊與對終端裝置21之服務要求的配置程序。請一併參考圖1、圖3及圖6，其中圖6係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的部分流程圖。在圖3的步驟S11之前，多邊緣雲之網路通訊控制方法更可以包含圖6所示之步驟S21：以第一邊緣運算平台向中控平台提出註冊要求；S22：以第一邊緣運算平台提供第一邊緣運算平台的運算資訊至中控平台；S23：以中控平台接收終端裝置的服務要求，並判斷服務要求對應於一目標服務；S24：以中控平台依據目標服務及第一邊緣運算平台的運算資訊，將目標服務配置至第一邊緣運算平台。In one embodiment, before the first edge computing device 11 provides the target service for the terminal device (ie, step S11 in FIG. 3 ), the network communication control method for the multi-edge cloud may further include registering the edge computing platform device and registering the terminal device 21 . Configurator required by the service. Please refer to FIG. 1 , FIG. 3 and FIG. 6 together, wherein FIG. 6 is a partial flowchart of a method for controlling network communication in a multi-edge cloud according to an embodiment of the present disclosure. Before step S11 in FIG. 3 , the network communication control method for the multi-edge cloud may further include step S21 shown in FIG. 6 : using the first edge computing platform to submit a registration request to the central control platform; S22 : using the first edge computing platform Provide the computing information of the first edge computing platform to the central control platform; S23: receive the service request of the terminal device by the central control platform, and determine that the service request corresponds to a target service; S24: use the central control platform according to the target service and the first edge The computing information of the computing platform configures the target service to the first edge computing platform.

進一步來說，圖1所示之通訊環境的建立包含了步驟S21～S24。以圖1來說明，於步驟S21及S22中，第一邊緣運算裝置11向中控裝置13提出註冊要求，中控裝置13可據以建立彼此間的連線，且在第一邊緣運算裝置11註冊完成後，第一邊緣運算裝置11可以提供自己的運算資訊（包含資源狀況、延遲等）至中控裝置13，特別係週期性地提供。另外，第二邊緣運算裝置31在重新配置程序執行前亦可以向中控裝置13提出註冊要求，並在註冊完成後提供自己的運算資訊至中控裝置13。此外，可以有更多的邊緣運算裝置向中控裝置13註冊並傳送自己的運算資訊，本揭露不予限制。Further, the establishment of the communication environment shown in FIG. 1 includes steps S21-S24. 1, in steps S21 and S22, the first edge computing device 11 submits a registration request to the central control device 13, and the central control device 13 can establish a connection with each other accordingly, and the first edge computing device 11 After the registration is completed, the first edge computing device 11 can provide its own computing information (including resource status, delay, etc.) to the central control device 13, especially periodically. In addition, the second edge computing device 31 can also submit a registration request to the central control device 13 before the reconfiguration program is executed, and provide its own computing information to the central control device 13 after the registration is completed. In addition, more edge computing devices may register with the central control device 13 and transmit their computing information, which is not limited in the present disclosure.

當邊緣運算系統1接收到終端裝置21的服務要求時，於步驟S23及S24中，中控裝置13判斷服務要求對應於目標服務，並依據目標服務及已註冊的邊緣運算裝置的運算資訊來配置目標服務。進一步來說，第一邊緣運算裝置11、第二邊緣運算裝置31或其他已向中控裝置13註冊的邊緣運算裝置的網域名稱服務模組可以接收到位於通訊範圍內的終端裝置21的服務要求（例如服務位置詢問），且執行上述接收步驟的邊緣運算裝置可以再將自己的身分資訊、服務的身分資訊、終端裝置21的身分資訊傳送至中控裝置13的服務位置管理模組133。於此實施例中，中控裝置13依據第一邊緣運算裝置11的運算資訊為對應於目標服務之最佳運算資訊，因而將目標服務配置至第一邊緣運算裝置11。所述最佳運算資訊的判斷的舉例及說明皆同於前列實施例的重新配置程序中的最佳運算資訊判斷，因此於此不再贅述。When the edge computing system 1 receives the service request from the terminal device 21, in steps S23 and S24, the central control device 13 determines that the service request corresponds to the target service, and configures it according to the target service and the computing information of the registered edge computing device target service. Further, the domain name service module of the first edge computing device 11 , the second edge computing device 31 or other edge computing devices registered with the central control device 13 can receive services from the terminal device 21 located within the communication range The edge computing device that performs the above-mentioned receiving step can then transmit its own identity information, service identity information, and identity information of the terminal device 21 to the service location management module 133 of the central control device 13 . In this embodiment, the central control device 13 allocates the target service to the first edge computing device 11 according to the optimal computing information corresponding to the target service according to the computing information of the first edge computing device 11 . The examples and descriptions of the determination of the optimal operation information are the same as the determination of the optimal operation information in the reconfiguration procedure of the previous embodiment, and thus are not repeated here.

另外如前所述，於一實施例中，中控裝置13可以係具有中控功能的行動邊緣運算雲，可以透過基地台與終端裝置21連接並提供終端裝置21服務。於此實施例中，中控裝置13在執行前列實施例所述之配置程序或重新配置程序時，可以將自己作為目標邊緣運算平台的選擇之一。特別來說，中控裝置13會將自己的優先序設定為最後，也就是說，中控裝置13會在判斷所連之所有邊緣運算裝置皆不適合提供服務後，才判斷自己是否適合提供服務。此外，當中控裝置13判斷無邊緣運算裝置能夠提供服務時，將產生配置失敗訊號至終端裝置21。In addition, as mentioned above, in one embodiment, the central control device 13 may be a mobile edge computing cloud with a central control function, which may be connected to the terminal device 21 through the base station and provide the terminal device 21 service. In this embodiment, when the central control device 13 executes the configuration procedure or the reconfiguration procedure described in the previous embodiments, it can select itself as one of the target edge computing platforms. In particular, the central control device 13 will set its own priority to the last, that is to say, the central control device 13 will determine whether it is suitable for providing services after determining that all connected edge computing devices are not suitable for providing services. In addition, when the central control device 13 determines that the edgeless computing device can provide services, it will generate a configuration failure signal to the terminal device 21 .

上述之多個實施例所提之邊緣運算系統及多邊緣雲之網路通訊控制方法涉及中控平台的設置及控制，適用於集中式的通訊環境。本揭露亦提出一種適用於分散式通訊環境的邊緣運算裝置。請參考圖7及圖8，其中圖7係依據本揭露一實施例所繪示的邊緣運算裝置之通訊環境的示意圖，圖8則係依據本揭露一實施例所繪示的邊緣運算裝置的功能方塊圖。圖7示例性地繪示分散式的通訊環境，其包含第一邊緣運算裝置11’、終端裝置21及第二邊緣運算裝置31’，其中第一邊緣運算裝置11’可以分別與終端裝置21及第二邊緣運算裝置31’通訊連接。於此要特別說明的是，圖7所示之與第一邊緣運算裝置11’連接的終端裝置21及第二邊緣運算裝置31’的數量僅為示例，第一邊緣運算裝置11’亦可同時與多個終端裝置21連接，且與多個第二邊緣運算裝置31’連接。此外，第二邊緣運算裝置31’亦可與一或多個終端裝置21連接。The edge computing system and the network communication control method of the multi-edge cloud mentioned in the above-mentioned various embodiments involve the setting and control of the central control platform, and are suitable for a centralized communication environment. The present disclosure also provides an edge computing device suitable for a distributed communication environment. Please refer to FIGS. 7 and 8 , wherein FIG. 7 is a schematic diagram of a communication environment of an edge computing device according to an embodiment of the present disclosure, and FIG. 8 is a function of an edge computing device according to an embodiment of the present disclosure. block diagram. FIG. 7 exemplarily shows a distributed communication environment, which includes a first edge computing device 11 ′, a terminal device 21 and a second edge computing device 31 ′, wherein the first edge computing device 11 ′ can communicate with the terminal devices 21 and 31 ′, respectively. The second edge computing device 31' is communicatively connected. It should be noted here that the number of the terminal device 21 and the second edge computing device 31 ′ connected to the first edge computing device 11 ′ shown in FIG. 7 is only an example, and the first edge computing device 11 ′ can also be simultaneously It is connected to a plurality of terminal devices 21, and is connected to a plurality of second edge computing devices 31'. In addition, the second edge computing device 31' can also be connected to one or more terminal devices 21.

第一邊緣運算裝置11’可以係行動邊緣運算雲（可簡稱邊緣雲），用於提供終端裝置21一或多種服務，且可以探詢鄰近的其他邊緣運算裝置（例如圖7所示之第二邊緣運算裝置31’）、取得鄰近邊緣運算裝置的運算資訊（包含資源狀況、延遲）並配置服務。換句話說，第一邊緣運算裝置11’具有與前列圖1及圖2所示之第一邊緣運算裝置11相同之功能，且更可以具有掌握鄰近邊緣裝置之資訊並進行服務配置的能力。同樣地，第一邊緣運算裝置11’可以係具有運算能力的單個伺服器，亦可以由多個具有不同功能且彼此連接的伺服器所組成。另，上述之終端裝置21例如為智慧型手機、平板、筆記型電腦等行動電子裝置，本揭露不予限制。The first edge computing device 11 ′ can be a mobile edge computing cloud (referred to as an edge cloud), which is used to provide one or more services to the terminal device 21 and can inquire about other neighboring edge computing devices (for example, the second edge computing device shown in FIG. 7 ). computing device 31 ′), obtain computing information (including resource status and delay) of neighboring edge computing devices, and configure services. In other words, the first edge computing device 11' has the same function as the first edge computing device 11 shown in FIG. 1 and FIG. 2 above, and can further have the capability of grasping the information of the neighboring edge devices and performing service configuration. Likewise, the first edge computing device 11' may be a single server with computing capability, or may be composed of multiple servers having different functions and connected to each other. In addition, the above-mentioned terminal device 21 is, for example, a mobile electronic device such as a smart phone, a tablet, and a notebook computer, which is not limited in the present disclosure.

進一步來說，如圖8所示，第一邊緣運算裝置11’除了包含如圖2所示之用戶介面功能模組111、網域名稱服務模組112、資源管理模組113、服務管理模組114、第一服務模組115及第二服務模組116之外，更可以包含鄰居管理模組117、參考訊號管理模組118、服務位置管理模組119及服務重新導向模組120。其中，與圖2之第一邊緣運算裝置11相同之模組的運作於此不再贅述。Further, as shown in FIG. 8 , the first edge computing device 11 ′ includes the user interface function module 111 , the domain name service module 112 , the resource management module 113 , and the service management module as shown in FIG. 2 . 114. In addition to the first service module 115 and the second service module 116, a neighbor management module 117, a reference signal management module 118, a service location management module 119 and a service redirection module 120 may be further included. The operation of the same module as that of the first edge computing device 11 in FIG. 2 will not be repeated here.

第一邊緣運算裝置11’的鄰居管理模組117可以探詢鄰近的其他邊緣運算裝置。更進一步來說，所謂鄰近邊緣運算裝置係指其所連接之基地台與第一邊緣運算裝置11’所連接之基地台兩者的通訊範圍有所重疊。所謂通訊範圍重疊意指當終端裝置21位於兩通訊範圍的交集處（或可理解為重疊處）時，鄰近邊緣運算裝置所連接之基地台與第一邊緣運算裝置11’所連接之基地台都具有與終端裝置21通訊的能力。鄰居管理模組117可以基於自動化鄰居關係（Automatic Neighbor Relation，ANR），特別係自動化管理鄰居關係表(Neighbor Relation Table，NRT)來建立鄰居關係清單，所述鄰居關係清單包含第二邊緣運算裝置31’的資訊。建立鄰居關係之邊緣運算裝置的用戶介面功能模組之間可以具有N9介面的連結。The neighbor management module 117 of the first edge computing device 11' can query other neighboring edge computing devices. Furthermore, the so-called adjacent edge computing device means that the communication range of the base station to which it is connected and the base station to which the first edge computing device 11' is connected overlap. The so-called overlapping communication ranges means that when the terminal device 21 is located at the intersection of the two communication ranges (or can be understood as overlapping), both the base station connected to the adjacent edge computing device and the base station connected to the first edge computing device 11' It has the ability to communicate with the terminal device 21 . The neighbor management module 117 may establish a neighbor relationship list based on an automatic neighbor relationship (Automatic Neighbor Relation, ANR), especially an automatic neighbor relationship table (Neighbor Relation Table, NRT), and the neighbor relationship list includes the second edge computing device 31 ' information. The user interface function modules of the edge computing device that establish the neighbor relationship may have N9 interface connections.

第一邊緣運算裝置11’的參考訊號管理模組118可以檢查是否有接收到其他邊緣運算裝置連接之基地台所傳出的訊號，若有則認定該邊緣運算裝置為具有鄰居關係之邊緣運算裝置。第一邊緣運算裝置11’的服務位置管理模組119及服務重新導向模組120則同於圖3之中控裝置13的服務位置管理模組133及服務重新導向模組134，用於執行服務的重新配置，其中，所述重新配置的詳細實施內容將於後描述。上述之模組可以各為獨立設置且彼此連接的伺服器，或者為運行於一或多個伺服器中的多個應用程序，本揭露並不限制運行上述模組之實體伺服器的數量。第二邊緣運算裝置31’可以具有與上述第一邊緣運算裝置11’一樣的架構及功能，於此不再贅述。The reference signal management module 118 of the first edge computing device 11' can check whether it has received a signal from a base station connected to other edge computing devices, and if so, determine that the edge computing device is an edge computing device with a neighbor relationship. The service location management module 119 and the service redirection module 120 of the first edge computing device 11' are the same as the service location management module 133 and the service redirection module 134 of the central control device 13 in FIG. reconfiguration, wherein the detailed implementation content of the reconfiguration will be described later. The above-mentioned modules may each be independently set up and connected servers, or may be multiple application programs running on one or more servers. The present disclosure does not limit the number of physical servers running the above-mentioned modules. The second edge computing device 31' may have the same structure and functions as the above-mentioned first edge computing device 11', and details are not described herein again.

本揭露亦提供一種適用於分散式通訊環境的多邊緣雲之網路通訊控制方法，請一併參考圖7～9，其中圖9係依據本揭露又一實施例所繪示的多邊緣雲之網路通訊控制方法的流程圖。如圖9所示，適用於分散式通訊環境的多邊緣雲之網路通訊控制方法可以包含以第一邊緣運算平台執行步驟S31：取得第二邊緣運算平台的運算資訊；步驟S32：提供終端裝置一目標服務；步驟S33：判斷發生服務異常狀況；步驟S34：依據服務異常狀況、目標服務及第二邊緣運算平台的運算資訊，將目標服務重新配置至第二邊緣運算平台，並產生卸載資訊及目標服務位置；步驟S35：將卸載資訊傳送至第二邊緣運算平台；步驟S36：將目標服務位置傳送至終端裝置，使終端裝置依據目標服務位置通訊連接至第二邊緣運算平台。The present disclosure also provides a network communication control method for a multi-edge cloud suitable for a distributed communication environment. Please refer to FIGS. 7-9 together, wherein FIG. 9 is a multi-edge cloud according to another embodiment of the present disclosure. The flow chart of the network communication control method. As shown in FIG. 9 , the network communication control method for multi-edge cloud suitable for distributed communication environment may include executing step S31 with the first edge computing platform: obtaining computing information of the second edge computing platform; step S32: providing a terminal device a target service; Step S33 : judging the occurrence of service abnormality; Step S34 : reconfiguring the target service to the second edge computing platform according to the service abnormality, the target service and the computing information of the second edge computing platform, and generating uninstall information and target service location; step S35: sending the offloading information to the second edge computing platform; step S36: sending the target service location to the terminal device, so that the terminal device communicates with the second edge computing platform according to the target service location.

於一實施例中，上述第一邊緣運算平台及第二邊緣運算平台分別以前述圖7及圖8之第一邊緣運算裝置11’及第二邊緣運算裝置31’來實現；而於另一實施例中，第一邊緣運算平台及第二邊緣運算平台可以分別係於伺服器中的虛擬機器所運行之應用程序。以下將以第一邊緣運算裝置11’及第二邊緣運算裝置31’所實現之第一邊緣運算平台及第二邊緣運算平台來示例性地說明圖9所示之多邊緣雲之網路通訊控制方法。In one embodiment, the first edge computing platform and the second edge computing platform are implemented by the first edge computing device 11 ′ and the second edge computing device 31 ′ in FIG. 7 and FIG. 8 , respectively; and in another implementation For example, the first edge computing platform and the second edge computing platform may be respectively associated with applications run by virtual machines in the server. In the following, the network communication control of the multi-edge cloud shown in FIG. 9 will be exemplarily explained by using the first edge computing platform and the second edge computing platform implemented by the first edge computing device 11 ′ and the second edge computing device 31 ′. method.

於步驟S31中，第一邊緣運算裝置11’可以取得所連之其他邊緣運算裝置（包含第二邊緣運算裝置31’）的運算資訊。進一步來說，第一邊緣運算裝置11’的鄰居管理模組117可以透過與其他邊緣運算裝置的鄰居管理模組之間的管理層網路（Management Network）來取得運算資訊。所述運算資訊包含資源狀況、所提供服務之延遲等。特別來說，第一邊緣運算裝置11’可以週期性地取得所連之其他邊緣運算裝置的運算資訊。另外，連接於第一邊緣運算裝置11’的其他邊緣運算裝置（包含第二邊緣運算裝置31’）亦可透過自己的鄰居管理模組取得第一邊緣運算裝置11’的運算資訊。意即，如前所述之具有鄰居關係的多個邊緣運算裝置之間可以互相傳遞運算資訊並儲存彼此的運算資訊。本揭露並不限制各邊緣運算裝置傳遞運算資訊的先後順序。In step S31, the first edge computing device 11' can obtain the computing information of other connected edge computing devices (including the second edge computing device 31'). Further, the neighbor management module 117 of the first edge computing device 11' can obtain computing information through a management network (Management Network) with neighbor management modules of other edge computing devices. The computing information includes resource status, delays of provided services, and the like. In particular, the first edge computing device 11' can periodically obtain the computing information of other connected edge computing devices. In addition, other edge computing devices (including the second edge computing device 31') connected to the first edge computing device 11' can also obtain the computing information of the first edge computing device 11' through their own neighbor management module. That is, as described above, a plurality of edge computing devices having a neighbor relationship can transfer computing information to each other and store each other's computing information. The present disclosure does not limit the order in which each edge computing device transmits computing information.

於步驟S32中，第一邊緣運算裝置11’提供終端裝置21目標服務，其中所述目標服務例如係人工智慧、虛擬實境、擴增實境、自動駕駛等應用服務，本揭露不予限制。進一步來說，第一邊緣運算裝置11’與終端裝置21之間的通訊連結是先前第一邊緣運算裝置11’、第二邊緣運算裝置31’或通訊環境中其他具有服務配置能力的邊緣運算裝置依據第一邊緣運算裝置11’的運算資訊所決定的，其中運算資訊包含資源狀況及第一邊緣運算裝置11’提供所述目標服務之延遲。於此要特別說明的是，上述步驟S32亦可以執行於步驟S31之前或與步驟S31同時執行，本揭露不予限制。於步驟S33中，第一邊緣運算裝置11’會判斷其本身發生了服務異常狀況，其中所述服務異常狀況可能為服務中止或服務移出。為了因應服務異常狀況，於步驟S34～S36中，第一邊緣運算裝置11’會進行重新配置以使另一邊緣運算裝置（第二邊緣運算裝置31’）提供原第一邊緣運算裝置11’所供之服務給終端裝置21。In step S32, the first edge computing device 11' provides the target service of the terminal device 21, wherein the target service is, for example, application services such as artificial intelligence, virtual reality, augmented reality, and automatic driving, which are not limited in this disclosure. Further, the communication link between the first edge computing device 11' and the terminal device 21 is the previous first edge computing device 11', the second edge computing device 31' or other edge computing devices with service configuration capability in the communication environment It is determined according to the computing information of the first edge computing device 11', wherein the computing information includes the resource status and the delay of the first edge computing device 11' providing the target service. It should be noted here that the above-mentioned step S32 may also be performed before or simultaneously with the step S31 , which is not limited in the present disclosure. In step S33, the first edge computing device 11' determines that a service abnormality has occurred in itself, wherein the service abnormality may be service suspension or service removal. In order to cope with abnormal service conditions, in steps S34 to S36 , the first edge computing device 11 ′ is reconfigured so that another edge computing device (the second edge computing device 31 ′) provides the The service is provided to the terminal device 21 .

詳細來說，請一併參考圖7～圖10，其中圖10係依據本揭露又一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。如圖10的步驟A31所示，當第一邊緣運算裝置11’用於提供目標服務之服務模組發生服務中止或服務移出之異常狀況時，第一邊緣運算裝置11’的服務管理模組114會產生對應之服務失效訊號並傳送至服務位置管理模組119。其中，服務中止及服務移出所指之詳細狀況皆如同前列圖4之實施例所述，於此不再贅述。當服務位置管理模組119接收到服務失效訊號時，服務位置管理模組119及服務重新導向模組120便可以據以進行重新配置。In detail, please refer to FIGS. 7 to 10 together, wherein FIG. 10 is a schematic diagram of signal transmission of a network communication control method of a multi-edge cloud according to another embodiment of the present disclosure. As shown in step A31 of FIG. 10 , when the service module used by the first edge computing device 11 ′ for providing the target service has an abnormal situation of service suspension or service removal, the service management module 114 of the first edge computing device 11 ′ A corresponding service failure signal will be generated and sent to the service location management module 119 . The detailed conditions of service suspension and service removal are the same as those described in the previous embodiment of FIG. 4 , and will not be repeated here. When the service location management module 119 receives the service failure signal, the service location management module 119 and the service redirection module 120 can perform reconfiguration accordingly.

所述重新配置可以包含重新定位程序及重新導向程序。其中，重新定位程序可以包含：依據服務失效訊號中的目標服務ID選擇候選邊緣運算平台，以及依據候選邊緣運算平台的運算資訊決定目標邊緣運算平台。如前所述，第一邊緣運算裝置11’可以週期性地收集所連接之其他邊緣運算裝置的資源狀況及所提供服務之延遲等資訊。進一步來說，服務位置管理模組119可以選擇包含具有目標服務ID之服務模組的邊緣運算裝置作為候選邊緣運算平台，並從中選擇具有最佳運算資訊的邊緣運算裝置來作為目標邊緣運算平台，即預計接續提供目標服務給終端裝置21之邊緣運算平台。所述最佳運算資訊的判斷的舉例及說明皆同於前列實施例中的最佳運算資訊判斷，因此於此不再贅述。The reconfiguration may include relocation procedures and redirection procedures. The relocation procedure may include: selecting a candidate edge computing platform according to the target service ID in the service failure signal, and determining a target edge computing platform according to computing information of the candidate edge computing platform. As mentioned above, the first edge computing device 11' can periodically collect information such as the resource status of other connected edge computing devices and the delay of the provided services. Further, the service location management module 119 can select the edge computing device including the service module with the target service ID as the candidate edge computing platform, and select the edge computing device with the best computing information from the edge computing device as the target edge computing platform, That is, the edge computing platform is expected to continuously provide the target service to the terminal device 21 . The examples and descriptions of the determination of the optimal operation information are the same as those of the determination of the optimal operation information in the foregoing embodiments, and thus will not be repeated here.

當服務位置管理模組119在執行上述重新定位程序而決定目標邊緣運算平台時，可以產生重新導向要求並傳送給服務重新導向模組120。所述重新導向要求包含目標邊緣運算平台的ID、目標服務ID以及終端裝置21的ID。以前述圖9之實施例而言，第二邊緣運算裝置31’即被服務位置管理模組119選為目標邊緣運算平台，因此於此實施例中，重新導向要求會包含第二邊緣運算裝置31’的ID。服務重新導向模組120可以依據重新導向要求將重新導向資訊傳送至用戶介面功能模組111，其中，重新導向資訊可以包含第二邊緣運算裝置31’的ID、終端裝置21的ID及目標服務ID。用戶介面功能模組111可以依據重新導向資訊來向第二邊緣運算裝置31’的用戶介面功能模組提出針對終端裝置21的連結建立要求，所述連結之建立特別係透過N9介面建立終端裝置21的連結（Session Connection）。When the service location management module 119 determines the target edge computing platform by executing the relocation procedure, a redirection request may be generated and sent to the service redirection module 120 . The redirection request includes the ID of the target edge computing platform, the target service ID, and the ID of the terminal device 21 . For the aforementioned embodiment of FIG. 9 , the second edge computing device 31 ′ is selected as the target edge computing platform by the service location management module 119 , so in this embodiment, the redirection request will include the second edge computing device 31 ' ID. The service redirection module 120 may transmit the redirection information to the user interface function module 111 according to the redirection request, wherein the redirection information may include the ID of the second edge computing device 31 ′, the ID of the terminal device 21 and the target service ID . The user interface function module 111 may request the user interface function module of the second edge computing device 31 ′ to establish a connection for the terminal device 21 according to the redirection information, and the connection establishment is specially established through the N9 interface to establish the connection of the terminal device 21 . Link (Session Connection).

第一邊緣運算裝置11’ 經上述程序將目標服務重新配置至第二邊緣運算裝置31’，並且產生卸載資訊及目標服務位置，再分別將卸載資訊傳送至第二邊緣運算裝置31’（步驟A32），將目標服務位置傳送至終端裝置21（步驟A33）。進一步來說，第一邊緣運算裝置11’可以藉由服務重新導向模組120將卸載資訊傳送至第二邊緣運算裝置31’。於此實施例中，卸載資訊包含第一邊緣運算裝置11’的ID、終端裝置21的ID以及目標服務ID。第二邊緣運算裝置31’可依據卸載資訊同意前述第一邊緣運算裝置11’之連結建立要求。簡而言之，第一邊緣運算裝置11’及第二邊緣運算裝置31’可以分別從服務重新導向模組120取得重新導向資訊及卸載資訊，據以與對方建立關於終端裝置21及目標服務的連結。第一邊緣運算裝置11’亦可以藉由服務重新導向模組120將目標服務位置傳送至終端裝置21。進一步來說，第一邊緣運算裝置11’的服務重新導向模組120在先前將目標服務重新配置至第二邊緣運算裝置31’時，可取得第二邊緣運算裝置31’中提供目標服務之服務模組的位置（例如IP）並將其作為目標服務位置。特別來說，服務重新導向模組120可以透過DNS Notify、HTTP 301或HTTP 302之方式將目標服務位置傳送至終端裝置21。終端裝置21在取得目標服務位置後，便可據以透過第一邊緣運算裝置11’與第二邊緣運算裝置31’之間的連結連接至第二邊緣運算裝置31’中提供目標服務之服務模組（步驟A34及A35）。The first edge computing device 11' reconfigures the target service to the second edge computing device 31' through the above procedure, generates offload information and target service location, and then transmits the offload information to the second edge computing device 31' respectively (step A32). ), the target service location is transmitted to the terminal device 21 (step A33). Further, the first edge computing device 11' can transmit the offload information to the second edge computing device 31' through the service redirection module 120. In this embodiment, the offload information includes the ID of the first edge computing device 11', the ID of the terminal device 21, and the target service ID. The second edge computing device 31' can agree to the connection establishment request of the first edge computing device 11' according to the offload information. In short, the first edge computing device 11' and the second edge computing device 31' can obtain redirection information and uninstallation information from the service redirection module 120, respectively, so as to establish a relationship with each other about the terminal device 21 and the target service. link. The first edge computing device 11' can also transmit the target service location to the terminal device 21 through the service redirection module 120. Further, when the service redirection module 120 of the first edge computing device 11' previously reconfigures the target service to the second edge computing device 31', it can obtain the service that provides the target service in the second edge computing device 31' The mod's location (e.g. IP) and use it as the target service location. Specifically, the service redirection module 120 can transmit the target service location to the terminal device 21 by means of DNS Notify, HTTP 301 or HTTP 302 . After the terminal device 21 obtains the target service location, it can connect to the service model in the second edge computing device 31 ′ that provides the target service through the connection between the first edge computing device 11 ′ and the second edge computing device 31 ′. group (steps A34 and A35).

本揭露亦提出一種整合式的網路通訊控制方法，可以切換運行上述各實施例所提之適用於集中式通訊環境的網路通訊控制方法或是用於分散式通訊環境的網路通訊控制方法，請一併參考圖2、圖8及圖11，其中圖11係依據本揭露另一實施例所繪示的邊緣運算系統之通訊環境的示意圖。如圖11所示，整合式的邊緣運算系統1’可以包含第一邊緣運算裝置11’及中控裝置13，兩者彼此通訊連接，邊緣運算系統1’可以與終端裝置21通訊連接，亦可與第二邊緣運算裝置31’通訊連接。進一步來說，第一邊緣運算裝置11’可以通訊連接於終端裝置21，亦可以通訊連接於第二邊緣運算裝置31’，中控裝置13則可以通訊連接於第二邊緣運算裝置31’。於此要特別說明的是，圖11所示之與邊緣運算系統1’連接的終端裝置21及第二邊緣運算裝置31’的數量僅為示例，邊緣運算系統1’亦可同時與多個終端裝置21連接，且與多個第二邊緣運算裝置31’連接。此外，第二邊緣運算裝置31’亦可與一或多個終端裝置連接。The present disclosure also provides an integrated network communication control method, which can switch between the network communication control method for a centralized communication environment or the network communication control method for a decentralized communication environment mentioned in the above embodiments. 2, FIG. 8 and FIG. 11 together, wherein FIG. 11 is a schematic diagram of a communication environment of an edge computing system according to another embodiment of the present disclosure. As shown in FIG. 11 , the integrated edge computing system 1 ′ can include a first edge computing device 11 ′ and a central control device 13 , which are connected in communication with each other. The edge computing system 1 ′ can be communicatively connected with the terminal device 21 , or It is connected in communication with the second edge computing device 31 ′. Further, the first edge computing device 11' can be communicatively connected to the terminal device 21, and can also be communicatively connected to the second edge computing device 31', and the central control device 13 can be communicatively connected to the second edge computing device 31'. It should be noted here that the number of the terminal device 21 and the second edge computing device 31 ′ connected to the edge computing system 1 ′ shown in FIG. 11 is only an example, and the edge computing system 1 ′ can also be connected to multiple terminals The device 21 is connected, and is connected to a plurality of second edge computing devices 31 ′. In addition, the second edge computing device 31' can also be connected to one or more terminal devices.

中控裝置13可以具有同於前列圖2之實施例所述的中控裝置13的模組及其功能運作，第一邊緣運算裝置11’則可以具有同於前列圖8實施例所述的邊緣運算裝置的模組及其功能運作。因此，邊緣運算系統1’可以藉由中控裝置13協同第一邊緣運算裝置11’來運行前列各實施例（包含圖3～6之實施例）所述之適用於集中式通訊環境的網路通訊控制方法，且可以藉由第一邊緣運算裝置11’來運行前列各實施例（包含圖9～10之實施例）所述之適用於分散式通訊環境的網路通訊控制方法。進一步來說，第二邊緣運算裝置31’亦可具有同於前列圖8實施例所述的邊緣運算裝置的模組及其功能運作。The central control device 13 may have the same modules and functions of the central control device 13 described in the previous embodiment of FIG. 2 , and the first edge computing device 11 ′ may have the same edge as described in the previous embodiment of FIG. The modules of the computing device and their functional operation. Therefore, the edge computing system 1 ′ can use the central control device 13 to cooperate with the first edge computing device 11 ′ to run the network applicable to the centralized communication environment described in the previous embodiments (including the embodiments in FIGS. 3 to 6 ). The communication control method, and the network communication control method applicable to the distributed communication environment described in the previous embodiments (including the embodiments in FIGS. 9 to 10 ) can be executed by the first edge computing device 11 ′. Further, the second edge computing device 31' can also have the same modules and functions as the edge computing device described in the previous embodiment of FIG. 8 .

於此實施例中，邊緣運算系統1’在一般運行時，係採用適用於集中式通訊環境的網路通訊控制方法，當第一邊緣運算裝置11’判斷與中控裝置13之間的連結斷線時，邊緣運算系統1’將改採適用於分散式通訊環境的網路通訊控制方法。舉例來說，於運行集中式通訊環境的網路通訊控制方法時，邊緣運算系統1’的中控裝置13在從第一邊緣運算裝置11’接收到運算資訊時，會回傳一確認訊息給第一邊緣運算裝置11’，因此，當第一邊緣運算裝置11’未接收到中控裝置13回傳之確認訊息時，第一邊緣運算裝置11’可以判斷與中控裝置13之間的連結已斷線，此時邊緣運算系統1’切換為分散式通訊環境的網路通訊控制方法（第一網路通訊控制方法）。特別來說，此實施例示例性地描述第一邊緣運算裝置11’為執行分散式通訊環境的網路通訊控制方法的邊緣運算裝置，而於另一實施例中，邊緣運算系統1’包含更多的邊緣運算裝置，且係藉由接收到終端裝置之服務請求的邊緣運算裝置來執行分散式通訊環境的網路通訊控制方法。當中控裝置13恢復上線時，中控裝置13會向第一邊緣運算裝置11’發送收集資訊請求。當第一邊緣運算裝置11’接收到收集資訊請求時，邊緣運算系統1’便會切換回集中式通訊環境的網路通訊控制方法（第二網路通訊控制方法）。In this embodiment, during normal operation of the edge computing system 1 ′, a network communication control method suitable for a centralized communication environment is adopted. When the first edge computing device 11 ′ determines that the connection with the central control device 13 is disconnected When online, the edge computing system 1' will adopt the network communication control method suitable for the distributed communication environment. For example, when running the network communication control method in the centralized communication environment, the central control device 13 of the edge computing system 1 ′ will return a confirmation message to the first edge computing device 11 ′ when receiving the computing information. The first edge computing device 11 ′, therefore, when the first edge computing device 11 ′ does not receive the confirmation message returned by the central control device 13 , the first edge computing device 11 ′ can determine the connection with the central control device 13 The line has been disconnected, and the edge computing system 1 ′ is switched to the network communication control method of the distributed communication environment (the first network communication control method). In particular, this embodiment exemplarily describes that the first edge computing device 11 ′ is an edge computing device that executes a network communication control method in a distributed communication environment, and in another embodiment, the edge computing system 1 ′ includes more There are many edge computing devices, and the network communication control method of the distributed communication environment is executed by the edge computing device that receives the service request from the terminal device. When the central control device 13 is back online, the central control device 13 will send a request to collect information to the first edge computing device 11'. When the first edge computing device 11' receives the information collection request, the edge computing system 1' will switch back to the network communication control method (the second network communication control method) of the centralized communication environment.

本揭露更提出一種邊緣運算裝置架構，特別來說，前列實施例所述之第一邊緣運算裝置11、中控裝置13及第一邊緣運算裝置11’皆可以具有此裝置架構。請參考圖12，圖12係依據本發明另一實施例所繪示的邊緣運算裝置的功能方塊圖。如圖12所示，邊緣運算裝置10包含記憶體101及處理器103，處理器103電性連接於記憶體101。記憶體101例如為快閃記憶體、唯讀記憶體、磁性記憶體或其他非揮發性儲存媒介。記憶體101中儲存有多個指令。進一步來說，所述指令至少包含前列實施例所述之第一邊緣運算裝置11/中控裝置13/第一邊緣運算裝置11’執行之步驟所對應的指令。處理器103例如為中央處理器、微控制器、可編程邏輯控制器等，可以執行記憶體101中所儲存的指令，以執行前列實施例所述之第一邊緣運算裝置11/中控裝置13/第一邊緣運算裝置11’執行之步驟。The present disclosure further proposes an edge computing device architecture. In particular, the first edge computing device 11, the central control device 13, and the first edge computing device 11' described in the foregoing embodiments can all have this device architecture. Please refer to FIG. 12 , which is a functional block diagram of an edge computing device according to another embodiment of the present invention. As shown in FIG. 12 , the edge computing device 10 includes a memory 101 and a processor 103 , and the processor 103 is electrically connected to the memory 101 . The memory 101 is, for example, a flash memory, a read-only memory, a magnetic memory, or other non-volatile storage media. A plurality of instructions are stored in the memory 101 . Further, the instructions at least include instructions corresponding to the steps performed by the first edge computing device 11/central control device 13/first edge computing device 11' described in the foregoing embodiments. The processor 103 is, for example, a central processing unit, a microcontroller, a programmable logic controller, etc., and can execute the instructions stored in the memory 101 to execute the first edge computing device 11/central control device 13 described in the previous embodiments /Steps performed by the first edge computing device 11'.

藉由上述架構，本揭露所提之邊緣運算系統、邊緣運算裝置、適用於集中式通訊環境的多邊緣雲之網路通訊控制方法以及適用於分散式通訊環境的多邊緣雲之網路通訊控制方法，可以因應多種服務異常狀況來執行提供服務之邊緣運算平台的重新配置，在遭遇服務異常狀況時無需重新佈建整個通訊環境，能夠具有服務不中斷之動態佈署能力，更可以據有負載平衡能力及異地備援能力。With the above structure, the present disclosure provides an edge computing system, an edge computing device, a network communication control method for a multi-edge cloud suitable for a centralized communication environment, and a network communication control for a multi-edge cloud suitable for a distributed communication environment. The method can perform the reconfiguration of the edge computing platform that provides services in response to various service abnormal conditions. When encountering service abnormal conditions, it is not necessary to reconfigure the entire communication environment. It can have the ability to dynamically deploy services without interruption, and can also be based on load. Balance ability and off-site backup ability.

雖然本揭露以前述之實施例揭露如上，然其並非用以限定本揭露。在不脫離本揭露之精神和範圍內，所為之更動與潤飾，均屬本揭露之專利保護範圍。關於本揭露所界定之保護範圍請參考所附之申請專利範圍。Although the present disclosure is disclosed in the foregoing embodiments, it is not intended to limit the present disclosure. Changes and modifications made without departing from the spirit and scope of this disclosure fall within the scope of patent protection of this disclosure. For the protection scope defined by this disclosure, please refer to the attached patent application scope.

1、1’:邊緣運算系統 11、11’:第一邊緣運算裝置 111、131:用戶介面功能模組 112、132:網域名稱服務模組 113、135:資源管理模組 114:服務管理模組 115:第一服務模組 116:第二服務模組 117:鄰居管理模組 118:參考訊號管理模組 119、133:服務位置管理模組 120、134:服務重新導向模組 13:中控裝置 21:終端裝置 31、31’:第二邊緣運算裝置 10:邊緣運算裝置 101:記憶體 103:處理器 S11~S15:步驟 A11~A19:步驟 A21~A27:步驟 S21~S24:步驟 S31~S36:步驟 A31~A35:步驟1, 1': edge computing system 11, 11': The first edge computing device 111, 131: User interface function module 112, 132: Domain Name Service Module 113, 135: Resource management module 114: Service Management Module 115: The first service module 116: Second Service Module 117: Neighbor Management Module 118: Reference signal management module 119, 133: Service location management module 120, 134: Service redirection mod 13: Central control device 21: Terminal device 31, 31': The second edge computing device 10: Edge computing device 101: Memory 103: Processor S11~S15: Steps A11~A19: Steps A21~A27: Steps S21~S24: Steps S31~S36: Steps A31~A35: Steps

圖1係依據本揭露一實施例所繪示的邊緣運算系統之通訊環境的示意圖。 圖2係依據本揭露一實施例所繪示的邊緣運算系統的功能方塊圖。 圖3係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的流程圖。 圖4係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。 圖5係依據本揭露另一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。 圖6係依據本揭露一實施例所繪示的多邊緣雲之網路通訊控制方法的部分流程圖。 圖7係依據本揭露一實施例所繪示的邊緣運算裝置之通訊環境的示意圖。 圖8係係依據本揭露一實施例所繪示的邊緣運算裝置的功能方塊圖。 圖9係依據本揭露又一實施例所繪示的多邊緣雲之網路通訊控制方法的流程圖。 圖10係依據本揭露又一實施例所繪示的多邊緣雲之網路通訊控制方法的訊號傳遞示意圖。 圖11係依據本揭露另一實施例所繪示的邊緣運算系統之通訊環境的示意圖。 圖12係依據本揭露另一實施例所繪示的邊緣運算裝置的功能方塊圖。FIG. 1 is a schematic diagram of a communication environment of an edge computing system according to an embodiment of the present disclosure. FIG. 2 is a functional block diagram of an edge computing system according to an embodiment of the present disclosure. FIG. 3 is a flowchart of a method for controlling network communication in a multi-edge cloud according to an embodiment of the present disclosure. FIG. 4 is a schematic diagram of signal transmission of a network communication control method for a multi-edge cloud according to an embodiment of the present disclosure. FIG. 5 is a schematic diagram of signal transmission of a network communication control method for a multi-edge cloud according to another embodiment of the present disclosure. FIG. 6 is a partial flowchart of a method for controlling network communication in a multi-edge cloud according to an embodiment of the present disclosure. FIG. 7 is a schematic diagram of a communication environment of an edge computing device according to an embodiment of the present disclosure. FIG. 8 is a functional block diagram of an edge computing device according to an embodiment of the present disclosure. FIG. 9 is a flowchart of a method for controlling network communication in a multi-edge cloud according to yet another embodiment of the present disclosure. FIG. 10 is a schematic diagram of signal transmission of a network communication control method for a multi-edge cloud according to yet another embodiment of the present disclosure. FIG. 11 is a schematic diagram of a communication environment of an edge computing system according to another embodiment of the present disclosure. FIG. 12 is a functional block diagram of an edge computing device according to another embodiment of the present disclosure.

S31~S36:步驟S31~S36: Steps

Claims (20)

一種多邊緣雲之網路通訊控制方法，適用於一第一邊緣運算平台及一第二邊緣運算平台，該第一邊緣運算平台與該第二邊緣運算平台進行資訊傳遞，該網路通訊控制方法包含：以該第一邊緣運算平台取得該第二邊緣運算平台的一運算資訊；以該第一邊緣運算平台提供一終端裝置一目標服務；以該第一邊緣運算平台判斷發生一服務異常狀況；以該第一邊緣運算平台依據該服務異常狀況、該目標服務及該第二邊緣運算平台的該運算資訊，將該目標服務重新配置至該第二邊緣運算平台，並產生一卸載資訊及一目標服務位置；以該第一邊緣運算平台將該卸載資訊傳送至該第二邊緣運算平台；以及以該第一邊緣運算平台將該目標服務位置傳送至該終端裝置，使該終端裝置依據該目標服務位置通訊連接至該第二邊緣運算平台；其中該卸載資訊至少包含該終端裝置的一身分資訊及該目標服務的一身分資訊。A network communication control method for multi-edge cloud, suitable for a first edge computing platform and a second edge computing platform, the first edge computing platform and the second edge computing platform perform information transmission, the network communication control method Including: obtaining a computing information of the second edge computing platform with the first edge computing platform; providing a target service of a terminal device with the first edge computing platform; judging the occurrence of a service abnormal condition with the first edge computing platform; Using the first edge computing platform to reconfigure the target service to the second edge computing platform according to the abnormal condition of the service, the target service and the computing information of the second edge computing platform, and generate an offload information and a target a service location; using the first edge computing platform to transmit the offloading information to the second edge computing platform; and using the first edge computing platform to transmit the target service location to the terminal device, so that the terminal device is based on the target service The location communication is connected to the second edge computing platform; wherein the offloading information at least includes an identity information of the terminal device and an identity information of the target service. 如請求項1所述的網路通訊控制方法，其中依據該服務異常狀況、該目標服務及該第二邊緣運算平台的該運算資訊，將該目標服務重新配置至該第二邊緣運算平台包含：依據一或多個候選邊緣運算平台中的每一候選邊緣運算平台的運算資訊以及該目標服務，從該一或多個候選邊緣運算平台中選擇該第二邊緣運算平台；以及依據該服務異常狀況及該選擇的結果產生一重新導向資訊。The network communication control method according to claim 1, wherein according to the abnormal condition of the service, the target service and the computing information of the second edge computing platform, reconfiguring the target service to the second edge computing platform comprises: selecting the second edge computing platform from the one or more candidate edge computing platforms according to the computing information of each candidate edge computing platform among the one or more candidate edge computing platforms and the target service; and according to the abnormal condition of the service and the result of the selection generates a redirection message. 如請求項2所述的網路通訊控制方法，其中該卸載資訊更包含該第一邊緣運算平台的一身分資訊，且該網路通訊控制方法更包含以該第一邊緣運算平台依據該重新導向資訊與該第二邊緣運算平台彼此建立連結。The network communication control method of claim 2, wherein the offloading information further includes an identity information of the first edge computing platform, and the network communication control method further includes using the first edge computing platform to redirect according to the redirection The information and the second edge computing platform are linked to each other. 如請求項1所述的網路通訊控制方法，其中該服務異常狀況指示該第一邊緣運算平台的該目標服務移出，且該網路通訊控制方法更包含以該第一邊緣運算平台依據一優先序決定將該目標服務移出。The network communication control method of claim 1, wherein the service abnormality indicates that the target service of the first edge computing platform is moved out, and the network communication control method further comprises using the first edge computing platform according to a priority The program decides to move the target service out. 如請求項1所述的網路通訊控制方法，其中該服務異常狀況指示該第一邊緣運算平台的該目標服務中止。The network communication control method of claim 1, wherein the service abnormality indicates that the target service of the first edge computing platform is terminated. 如請求項1所述的網路通訊控制方法，其中依據該服務異常狀況、該目標服務及該第二邊緣運算平台的該運算資訊，將該目標服務重新配置至該第二邊緣運算平台包含：選擇能夠提供該目標服務的一或多個候選邊緣運算平台，其中該一或多個候選邊緣運算平台各具有一運算資訊，該一或多個候選邊緣運算平台包含該第二邊緣運算平台，且該運算資訊包含資源狀況及對應於該目標服務的延遲；以及判斷該第二邊緣運算平台的該運算資訊為一最佳運算資訊，而從該一或多個候選邊緣運算平台中選擇該第二邊緣運算平台。The network communication control method according to claim 1, wherein according to the abnormal condition of the service, the target service and the computing information of the second edge computing platform, reconfiguring the target service to the second edge computing platform comprises: selecting one or more candidate edge computing platforms capable of providing the target service, wherein each of the one or more candidate edge computing platforms has computing information, the one or more candidate edge computing platforms includes the second edge computing platform, and The computing information includes resource status and delay corresponding to the target service; and determining that the computing information of the second edge computing platform is optimal computing information, and selecting the second edge computing platform from the one or more candidate edge computing platforms Edge computing platform. 如請求項1所述的網路通訊控制方法，在以該第一邊緣運算平台取得該第二邊緣運算平台的該運算資訊之前更包含以該第一邊緣運算平台建立一鄰居關係清單，其中該鄰居關係清單包含該第二邊緣運算平台。The network communication control method according to claim 1, before obtaining the computing information of the second edge computing platform with the first edge computing platform, further comprising establishing a neighbor relationship list with the first edge computing platform, wherein the The neighbor relationship list includes the second edge computing platform. 一種多邊緣雲之網路通訊控制方法，適用於一第一邊緣運算平台、一第二邊緣運算平台及一中控平台，該中控平台與該第一邊緣運算平台及該第二邊緣運算平台進行資訊傳遞，該網路通訊控制方法包含：以該第一邊緣運算平台提供一終端裝置一目標服務；以該中控平台或該第一邊緣運算平台執行多個步驟，其中該些步驟包含：判斷該第一邊緣運算平台發生一服務異常狀況；依據該服務異常狀況、該目標服務及該第二邊緣運算平台的一運算資訊，將該目標服務重新配置至該第二邊緣運算平台，並產生一卸載資訊及一目標服務位置；以及將該卸載資訊傳送至該第二邊緣運算平台；以及以該第一邊緣運算平台或該第二邊緣運算平台將該目標服務位置傳送至該終端裝置，使該終端裝置依據該目標服務位置通訊連接至該第二邊緣運算平台；其中該卸載資訊至少包含該終端裝置的一身分資訊及該目標服務的一身分資訊。A network communication control method for multi-edge cloud, suitable for a first edge computing platform, a second edge computing platform and a central control platform, the central control platform, the first edge computing platform and the second edge computing platform For information transmission, the network communication control method includes: providing a terminal device with a target service on the first edge computing platform; performing a plurality of steps on the central control platform or the first edge computing platform, wherein the steps include: Determining that a service abnormality occurs on the first edge computing platform; according to the service abnormality, the target service and a computing information of the second edge computing platform, reconfigure the target service to the second edge computing platform, and generate an offloading information and a target service location; and sending the offloading information to the second edge computing platform; and using the first edge computing platform or the second edge computing platform to send the target serving location to the terminal device, so that The terminal device is connected to the second edge computing platform according to the target service location communication; wherein the offloading information at least includes an identity information of the terminal device and an identity information of the target service. 如請求項8所述的多邊緣雲之網路通訊控制方法，其中以該中控平台或該第一邊緣運算平台執行該些步驟包含：當該第一邊緣運算平台判斷未接收到該中控平台的一確認訊息時，以該第一邊緣運算平台執行該些步驟。The multi-edge cloud network communication control method according to claim 8, wherein performing the steps with the central control platform or the first edge computing platform includes: when the first edge computing platform determines that the central control platform has not received When a confirmation message of the platform is received, the steps are performed by the first edge computing platform. 如請求項8所述的多邊緣雲之網路通訊控制方法，其中以該中控平台或該第一邊緣運算平台執行該些步驟包含：當該第一邊緣運算平台從該中控平台接收一收集資訊請求時，以該中控平台執行該些步驟。The multi-edge cloud network communication control method according to claim 8, wherein performing the steps with the central control platform or the first edge computing platform comprises: when the first edge computing platform receives a message from the central control platform When collecting information requests, use the central control platform to perform these steps. 一種邊緣運算裝置，包含一記憶體及一處理器，該處理器電性連接於該記憶體，該記憶體中儲存有多個指令，且該處理器執行該些指令以執行以下步驟：取得另一邊緣運算裝置的一運算資訊；提供一終端裝置一目標服務；判斷發生一服務異常狀況；依據該服務異常狀況、該目標服務及該另一邊緣運算裝置的該運算資訊，將該目標服務重新配置至該另一邊緣運算裝置，並產生一卸載資訊及一目標服務位置；將該卸載資訊傳送至該另一邊緣運算裝置；以及將該目標服務位置傳送至該終端裝置，使該終端裝置依據該目標服務位置通訊連接至該另一邊緣運算裝置；其中該卸載資訊至少包含該終端裝置的一身分資訊及該目標服務的一身分資訊。An edge computing device includes a memory and a processor, the processor is electrically connected to the memory, a plurality of instructions are stored in the memory, and the processor executes the instructions to perform the following steps: obtaining another A computing information of an edge computing device; providing a target service to a terminal device; judging that a service abnormality has occurred; resetting the target service according to the service abnormality, the target service and the computing information of the other edge computing device configure to the other edge computing device, and generate an offload information and a target service location; send the offload information to the other edge computing device; and send the target service location to the terminal device, so that the terminal device according to the The target service location is communicatively connected to the other edge computing device; wherein the offloading information at least includes an identity information of the terminal device and an identity information of the target service. 如請求項11所述的邊緣運算裝置，其中該處理器所執行之依據該服務異常狀況、該目標服務及該另一邊緣運算裝置的該運算資訊，將該目標服務重新配置至該另一邊緣運算裝置包含：依據一或多個候選邊緣運算平台中的每一候選邊緣運算平台的運算資訊以及該目標服務，從該一或多個候選邊緣運算平台中選擇該另一邊緣運算裝置；以及依據該服務異常狀況及該選擇的結果產生一重新導向資訊。The edge computing device of claim 11, wherein the processor reconfigures the target service to the other edge according to the service abnormality, the target service and the computing information of the other edge computing device The computing device includes: selecting the other edge computing device from the one or more candidate edge computing platforms according to computing information of each candidate edge computing platform in the one or more candidate edge computing platforms and the target service; and according to The service exception and the result of the selection generate a redirect message. 如請求項12所述的邊緣運算裝置，其中該卸載資訊更包含該邊緣運算裝置的一身分資訊，且該處理器更執行依據該重新導向資訊與該另一邊緣運算裝置建立連結。The edge computing device of claim 12, wherein the offload information further includes an identity information of the edge computing device, and the processor further establishes a connection with the other edge computing device according to the redirection information. 如請求項11所述的邊緣運算裝置，其中該服務異常狀況指示該邊緣運算裝置的該目標服務移出，且該處理器更執行依據一優先序決定將該目標服務移出。The edge computing device of claim 11, wherein the service abnormality indicates that the target service of the edge computing device is removed, and the processor further determines to remove the target service according to a priority order. 如請求項11所述的邊緣運算裝置，其中該服務異常狀況指示該邊緣運算裝置的該目標服務中止。The edge computing device of claim 11, wherein the service abnormality indicates that the target service of the edge computing device is suspended. 如請求項11所述的邊緣運算裝置，其中該處理器所執行之依據該服務異常狀況、該目標服務及該另一邊緣運算裝置的該運算資訊，將該目標服務重新配置至該另一邊緣運算裝置包含：選擇能夠提供該目標服務的一或多個候選邊緣運算平台，其中該一或多個候選邊緣運算平台各具有一運算資訊，該一或多個候選邊緣運算平台包含該另一邊緣運算裝置，且該運算資訊包含資源狀況及對應於該目標服務的延遲；以及判斷該另一邊緣運算裝置的該運算資訊為一最佳運算資訊，而從該一或多個候選邊緣運算平台中選擇該另一邊緣運算裝置。The edge computing device of claim 11, wherein the processor reconfigures the target service to the other edge according to the service abnormality, the target service and the computing information of the other edge computing device The computing device includes: selecting one or more candidate edge computing platforms capable of providing the target service, wherein each of the one or more candidate edge computing platforms has computing information, and the one or more candidate edge computing platforms include the other edge computing platform computing device, and the computing information includes resource status and delay corresponding to the target service; and judging that the computing information of the other edge computing device is an optimal computing information, and selects the computing information from the one or more candidate edge computing platforms Select the other edge computing device. 如請求項11所述的邊緣運算裝置，其中該處理器在執行取得該另一邊緣運算裝置的該運算資訊之前，更執行建立一鄰居關係清單，且該鄰居關係清單包含該另一邊緣運算裝置。The edge computing device of claim 11, wherein the processor further executes to create a neighbor relationship list before executing the acquisition of the computing information of the other edge computing device, and the neighbor relationship list includes the other edge computing device . 一種邊緣運算系統，包含：如請求項11所述的該邊緣運算裝置，用於執行一第一網路通訊控制方法，其中該第一網路通訊控制方法包含該處理器執行該些指令所執行的該些步驟；以及一中控裝置，通訊連接於該邊緣運算裝置，該中控裝置用於執行一第二網路通訊控制方法，其中該第二網路通訊控制方法包含：判斷該邊緣運算裝置發生該服務異常狀況；依據該服務異常狀況、該目標服務及該另一邊緣運算裝置的該運算資訊，將該目標服務重新配置至該另一邊緣運算裝置並產生該卸載資訊及該目標服務位置；將該卸載資訊傳送至該另一邊緣運算裝置；以及使該邊緣運算裝置或該另一邊緣運算裝置將該目標服務位置傳送至該終端裝置，使該終端裝置依據該目標服務位置通訊連接至該另一邊緣運算裝置。An edge computing system, comprising: the edge computing device as described in claim 11, for executing a first network communication control method, wherein the first network communication control method comprises the processor executing the instructions. and a central control device communicatively connected to the edge computing device, the central control device is used to execute a second network communication control method, wherein the second network communication control method comprises: judging the edge computing The service abnormality occurs in the device; according to the service abnormality, the target service and the computing information of the other edge computing device, reconfigure the target service to the other edge computing device and generate the offload information and the target service location; send the offload information to the other edge computing device; and cause the edge computing device or the other edge computing device to send the target service location to the terminal device, so that the terminal device communicates and connects according to the target service location to the other edge computing device. 如請求項18所述的邊緣運算系統，其中當該邊緣運算裝置判斷未接收到該中控裝置的一確認訊息時，該邊緣運算系統運行該第一網路通訊控制方法。The edge computing system of claim 18, wherein when the edge computing device determines that a confirmation message from the central control device has not been received, the edge computing system executes the first network communication control method. 如請求項18所述的邊緣運算系統，其中當該邊緣運算裝置從該中控裝置接收一收集資訊請求時，該邊緣運算系統運行該第二網路通訊控制方法。The edge computing system of claim 18, wherein when the edge computing device receives a request for collecting information from the central control device, the edge computing system executes the second network communication control method.
TW109100213A 2019-12-13 2020-01-03 Network communication control method of multiple edge clouds and edge computing device and system TWI752401B (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US16/714,200 2019-12-13
US16/714,200 US10771569B1 (en) 2019-12-13 2019-12-13 Network communication control method of multiple edge clouds and edge computing system
US201916722961A 2019-12-20 2019-12-20
US16/722,961 2019-12-20

Publications (2)

Publication Number Publication Date
TW202123731A TW202123731A (en) 2021-06-16
TWI752401B true TWI752401B (en) 2022-01-11

Family

ID=77516978

Family Applications (1)

Application Number Title Priority Date Filing Date
TW109100213A TWI752401B (en) 2019-12-13 2020-01-03 Network communication control method of multiple edge clouds and edge computing device and system

Country Status (1)

Country Link
TW (1) TWI752401B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI801133B (en) * 2022-02-11 2023-05-01 中華電信股份有限公司 System of transferring data in mobile edge computing, method and computer readable medium thereof
TWI821038B (en) * 2022-11-22 2023-11-01 財團法人工業技術研究院 Computing task dispatching method, terminal electronic device and computing system using the same

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130166712A1 (en) * 2011-12-27 2013-06-27 Microsoft Corporation Cloud-edge topologies
TWI627871B (en) * 2017-09-05 2018-06-21 財團法人工業技術研究院 Control method for network communication system and multi-access edge computing ecosystem device
TWI627872B (en) * 2017-09-05 2018-06-21 財團法人工業技術研究院 Control method for network communication system including base station network management server and multi-access edge computing ecosystem device
US20180173570A1 (en) * 2016-12-16 2018-06-21 Accenture Global Solutions Limited Edge computing platform

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130166712A1 (en) * 2011-12-27 2013-06-27 Microsoft Corporation Cloud-edge topologies
US20180173570A1 (en) * 2016-12-16 2018-06-21 Accenture Global Solutions Limited Edge computing platform
TWI627871B (en) * 2017-09-05 2018-06-21 財團法人工業技術研究院 Control method for network communication system and multi-access edge computing ecosystem device
TWI627872B (en) * 2017-09-05 2018-06-21 財團法人工業技術研究院 Control method for network communication system including base station network management server and multi-access edge computing ecosystem device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CONGFENG JIANG等人，「Toward Computation Offloading in Edge Computing: A Survey」, date of publication August 30, 2019. *

Also Published As

Publication number Publication date
TW202123731A (en) 2021-06-16

Similar Documents

Publication Publication Date Title
US10771569B1 (en) Network communication control method of multiple edge clouds and edge computing system
CN112788089B (en) Multi-edge cloud network communication control method, edge computing device and system
TWI724106B (en) Business flow control method, device and system between data centers
US7225356B2 (en) System for managing operational failure occurrences in processing devices
US9762669B2 (en) Service management roles of processor nodes in distributed node service management
TWI727496B (en) Network communication control method of multiple edge clouds and edge computing system
TWI476581B (en) Method, apparatus and computer program product for providing high availability in an active/active appliance cluster
US20150146716A1 (en) Communication System Architecture
CN107439031B (en) Radio-property-based access point load balancing in controller clusters
TWI752401B (en) Network communication control method of multiple edge clouds and edge computing device and system
CN105612539B (en) Producer system partitioning among leasing agent systems
US20220053602A1 (en) Apparatus and method for supporting continuity of edge computing service in mobile network
US9641558B2 (en) Communication system architecture
EP3203686B1 (en) Virtualized network function interworking
CN116048538B (en) Service grid deployment method and device for DPU
TW202110223A (en) Conditional configuration in a wireless communication network
CN113366814B (en) Method for managing resource allocation in edge computing system
US11659478B2 (en) Endpoint computing device multi-network slice utilization system
CN114143569B (en) Webpage recording and live broadcasting method and system
US11757987B2 (en) Load balancing systems and methods
CN116471343B (en) Equipment board card management method, load balancing method, device, server and medium
CN115361271B (en) SSH server switching and connecting method, cloud server and storage medium
US12015521B2 (en) Using an application programming interface (API) gateway to manage communications in a distributed system
CN117176788A (en) Access control method and device of Internet of things equipment, storage medium and electronic device
JP6246677B2 (en) COMMUNICATION SYSTEM, CONTROL DEVICE, AND PROCESSING DEVICE SWITCHING METHOD