A cloud service based path construction method and related device

Abstract

The application provides a path construction method based on cloud service and related equipment. When a direct path between a first central cloud area management node and a first edge cloud fails, a target path between the first central cloud area management node and the first edge cloud is constructed, so that the two can keep communication. In the application, the first edge cloud is managed by the first central cloud area management node through a first path. The method of the application comprises: the first central cloud area management node determines that the first path fails, and acquires candidate paths between the first central cloud area management node and the first edge cloud except the first path; the first central cloud area management node selects a target path from the candidate paths, the target path takes the first central cloud area management node and the first edge cloud as end points, and takes a second central cloud area management node and / or a second edge cloud as a relay point; the first central cloud area management node constructs the target path, and manages the first edge cloud through the target path.

Description

Technical Field

[0001] This application relates to the field of cloud technology, and in particular to a path construction method based on cloud services and related equipment. Background Technology

[0002] Because edge cloud is closer to users or data sources, it can provide services to the terminal devices used by users nearby, thereby generating faster network service response and meeting users' various needs in real-time business, security and privacy protection.

[0003] Generally, different edge clouds distributed in various locations can be uniformly managed by a central cloud. The central cloud can be divided into multiple regions based on geographical factors. Each central cloud region's management surface (also called a central cloud region management node) can be used to manage a portion of the edge clouds, thereby improving management efficiency. For any given central cloud region management node, a direct path is established between that central cloud region management node and the edge clouds it manages, enabling communication between the central cloud region and the edge clouds it manages.

[0004] However, when the path between the central cloud regional management node and an edge cloud fails, cloud service provider personnel need to go to the site where the edge cloud is deployed to perform maintenance to ensure that the edge cloud continues to provide services to users, but it cannot guarantee that communication between the central cloud regional management node and the edge cloud will be restored. Summary of the Invention

[0005] This application provides a cloud service-based path construction method and related equipment. When the direct path between the first central cloud area management node and the first edge cloud fails, a target path between the first central cloud area management node and the first edge cloud is constructed, so that communication can be maintained between the first central cloud area management node and the first edge cloud.

[0006] A first aspect of this application provides a path construction method based on cloud services. In this method, a first edge cloud is managed by a first central cloud regional management node through a first path. The method includes:

[0007] Once the first central cloud area management node identifies a first path failure, it can perform path planning to obtain at least one candidate path between the first central cloud area and the first edge cloud. It is understood that these candidate paths all originate from (or terminate) the first central cloud area, terminate (or originate) the first edge cloud, and pass through relay points.

[0008] The first central cloud region management node selects one candidate path as the target path from at least one candidate path. The target path can be one of the following: (1) The target path includes: a second path between the first central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. (2) The target path includes: a fourth path between the first central cloud region management node and the second central cloud region management node, a fifth path between the second central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. (3) The target path includes: a fourth path between the first central cloud region management node and the second central cloud region management node, and a sixth path between the second central cloud region management node and the first edge cloud.

[0009] The first central cloud region management node constructs the target path, so based on this target path, the first central cloud region and the first edge cloud can maintain communication, that is, the first central cloud region management node can manage the first edge cloud through the target path.

[0010] As can be seen from the above method, if the first central cloud region management node determines that the first path between the first central cloud region management node and the first edge cloud is faulty, it can first obtain candidate paths between the first central cloud region management node and the first edge cloud. The first edge cloud is managed by the first central cloud region management node through the first path. Then, the first central cloud region management node determines the target path from the candidate paths. This target path passes through relay points such as the second edge cloud and / or the second central cloud region management node. Finally, the first central cloud region management node constructs the target path. Based on this target path, the first central cloud region management node and the first edge cloud can maintain communication. That is, the first central cloud region management node can manage the first edge cloud through the target path, thereby improving the reliability and maintainability of the entire cloud service system, reducing the manpower costs required for operation and maintenance, and accelerating the logical and large-scale commercialization of edge clouds to a certain extent.

[0011] In one possible implementation, a second path has been established between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through this second path. The construction of the target path by the first central cloud region management node includes: constructing a third path between the first edge cloud and the second edge cloud, thereby obtaining the target path based on the second path between the first central cloud region management node and the second edge cloud, and the third path between the second edge cloud and the first edge cloud. In this way, the first central cloud region and the first edge cloud can maintain communication through this target path, meaning that the first central cloud region management node can manage the first edge cloud through the target path.

[0012] In one possible implementation, a fourth path has been constructed between the first central cloud region management node and the second central cloud region management node, and a fifth path has been constructed between the second central cloud region management node and the second edge cloud. The second edge cloud is managed by the second central cloud region management node through the fifth path. The first central cloud region management node constructs the target path by constructing a third path between the first and second edge clouds, thereby obtaining the target path based on the fourth path between the first and second central cloud region management nodes, the fifth path between the second and second edge clouds, and the third path between the second and first edge clouds. In this way, the first central cloud region and the first edge cloud can maintain communication through this target path, that is, the first central cloud region management node can manage the first edge cloud through the target path.

[0013] In one possible implementation, a fourth path has been established between the first central cloud region management node and the second central cloud region management node. The first central cloud region management node constructs a target path by constructing a sixth path between the second central cloud region management node and the first edge cloud, thereby obtaining a target path based on the fourth path between the first and second central cloud region management nodes and the sixth path between the second central cloud region management node and the first edge cloud. In this way, the first central cloud region and the first edge cloud can maintain communication through this target path, meaning the first central cloud region management node can manage the first edge cloud through the target path.

[0014] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the path with the minimum number of hops among at least one candidate path, or the path with the maximum network bandwidth among at least one candidate path. In the aforementioned implementations, the factors considered by the first central cloud area management node in selecting the target path can be any one of network latency, hop count, and network bandwidth, or any combination of these factors. Therefore, the first central cloud area can select the most suitable relay path.

[0015] In one possible implementation, the first path, the second path, the third path, the fourth path, the fifth path, and the sixth path are all direct paths.

[0016] In one possible implementation, the first edge cloud and the second edge cloud are located in the same data center. If both the first edge cloud and the second edge cloud are connected to the local network of the data center, it means that the first edge cloud and the second edge cloud are reachable, that is, the direct path between them can be built based on the local network of the data center.

[0017] In one possible implementation, the first edge cloud and the second edge cloud are located in the same data center. If the first edge cloud and / or the second edge cloud are not connected to the local network of the data center, it means that the first edge cloud and the second edge cloud are unreachable.

[0018] In one possible implementation, the method further includes: the first central cloud region management node acquiring the status information of the first edge cloud, which may include data such as the operating status of the first edge cloud, the operating indicators of the first edge cloud, and the network quality between the first edge cloud and the first central cloud region. When these data are within the normal range, the first central cloud region management node can determine that the first path between the first central cloud region management node and the first edge cloud is normal; when these data are outside the normal range, the first central cloud region management node can determine that the first path between the first central cloud region management node and the first edge cloud is faulty.

[0019] A second aspect of this application provides a cloud service-based path construction method, wherein a first edge cloud is managed by a first central cloud regional management node through a first path, the method comprising:

[0020] In the event of a failure in the first central cloud area management node itself or a failure in the path between the first central cloud area management node and the client (i.e., the first central cloud area management node itself is not faulty, but there is a network failure between the first central cloud area management node and the staff's client), the second central cloud area management node receives the access request for the first edge cloud sent by the client.

[0021] Based on the access request, the second central cloud area management node can perform path planning to obtain at least one candidate path between the second central cloud area management node and the first edge cloud, and select the target path from the at least one candidate path.

[0022] After the second central cloud area management node determines the target path between the second central cloud area management node and the first edge cloud, it can construct the target path. Based on the target path, the second central cloud area management node and the first edge cloud can communicate, that is, the second central cloud area management node can manage the first edge cloud through the target path.

[0023] In related technologies, if the primary central cloud area management node itself fails or the path between the primary central cloud area management node and the client fails, all edge clouds managed by the primary central cloud area management node will be affected, with a large impact radius. As can be seen from the above method: after staff discover a failure in the primary central cloud area management node, they can input an access request to the secondary central cloud area management node through the client. Then, the secondary central cloud area management node can obtain candidate paths between itself and the primary edge cloud based on the access request. Next, the secondary central cloud area management node selects a target path from the candidate paths. Finally, the secondary central cloud area management node constructs the target path. Based on this target path, communication can be achieved between the secondary central cloud area management node and the primary edge cloud. That is, the secondary central cloud area management node can manage the primary edge cloud in place of the primary central cloud area management node through the target path, thereby improving the reliability and maintainability of the entire cloud service system and reducing the impact radius caused by the failure of the primary central cloud area management node.

[0024] In one possible implementation, the target path is a sixth path between the second central cloud region management node and the first edge cloud. The construction of this target path by the second central cloud region management node includes: the second central cloud region management node constructing a sixth path between itself and the first edge cloud. In this way, communication between the second central cloud region and the first edge cloud can be achieved through this path; that is, the second central cloud region management node can manage the first edge cloud through this path.

[0025] In one possible implementation, a second path has been established between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through this second path. The target path includes a fifth path between the second central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. The second central cloud region management node constructs the target path by building both the fifth path between itself and the second edge cloud, and the third path between the second edge cloud and the first edge cloud. This results in a target path constructed based on the fifth path between the second central cloud region management node and the second edge cloud, and the third path between the second edge cloud and the first edge cloud. In this way, the second central cloud region and the first edge cloud can communicate through this target path; that is, the second central cloud region management node can manage the first edge cloud through this target path.

[0026] In one possible implementation, a fourth path has been constructed between the second central cloud region management node and the first central cloud region. The target path includes the fourth path between the second central cloud region management node and the first central cloud region, and a first path between the first central cloud region management node and the first edge cloud. The second central cloud region management node constructs the target path by directly reusing the fourth path between itself and the first central cloud region, and the first path between itself and the first edge cloud, thereby obtaining the target path based on the fourth path and the first path. In this way, the second central cloud region and the first edge cloud can communicate through this target path, that is, the second central cloud region management node can manage the first edge cloud through this target path.

[0027] In one possible implementation, a second path has been established between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through this second path. A fourth path has also been established between the second central cloud region management node and the first central cloud region. The target path includes the fourth path between the second central cloud region management node and the first central cloud region, the second path between the first central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. The second central cloud region management node constructs the target path by building the third path between the second edge cloud and the first edge cloud, thus obtaining the target path based on the fourth path between the second central cloud region management node and the first central cloud region, the second path between the first central cloud region management node and the second edge cloud, and the third path between the second edge cloud and the first edge cloud. In this way, the second central cloud region and the first edge cloud can communicate through this target path, meaning the second central cloud region management node can manage the first edge cloud through this target path.

[0028] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the path with the minimum number of hops among at least one candidate path, or the path with the maximum network bandwidth among at least one candidate path. In the aforementioned implementations, the factors considered by the first central cloud area management node in selecting the target path can be any one of network latency, hop count, and network bandwidth, or any combination of these factors. Therefore, the first central cloud area can select the most suitable relay path.

[0029] In one possible implementation, the first path, the second path, the third path, the fourth path, the fifth path, and the sixth path are all direct paths.

[0030] In one possible implementation, the first edge cloud and the second edge cloud are located in the same data center. If both the first edge cloud and the second edge cloud are connected to the local network of the data center, it means that the first edge cloud and the second edge cloud are reachable, that is, the direct path between them can be built based on the local network of the data center.

[0031] A third aspect of this application provides a cloud service-based path construction apparatus, which is the aforementioned first central cloud region management node. A first edge cloud is managed by the apparatus via a first path. The apparatus includes: an acquisition module, configured to determine a first path failure and acquire at least one candidate path between the first central cloud region management node and the first edge cloud, excluding the first path; a selection module, configured to select a target path from the at least one candidate path, the target path including: a second path between the first central cloud region management node and a second edge cloud, and a third path between the second edge cloud and the first edge cloud; or, a fourth path between the first central cloud region management node and the second central cloud region management node, a fifth path between the second central cloud region management node and the second edge cloud, and the third path; or, a fourth path and a sixth path between the second central cloud region management node and the first edge cloud; and a construction module, configured to construct the target path and manage the first edge cloud through the target path.

[0032] As can be seen from the above device, if the first central cloud area management node determines that the first path between the first central cloud area management node and the first edge cloud is faulty, it can first obtain candidate paths between the first central cloud area management node and the first edge cloud. The first edge cloud is managed by the first central cloud area management node through the first path. Then, the first central cloud area management node determines the target path from the candidate paths. This target path passes through relay points such as the second edge cloud and / or the second central cloud area management node. Finally, the first central cloud area management node constructs the target path. Based on this target path, the first central cloud area management node and the first edge cloud can maintain communication. That is, the first central cloud area management node can manage the first edge cloud through the target path, thereby improving the reliability and maintainability of the entire cloud service system, reducing the manpower costs required for operation and maintenance, and accelerating the logic and large-scale commercialization of edge clouds to a certain extent.

[0033] In one possible implementation, a second path has been constructed, and the second edge cloud is managed by the first central cloud region management node through the second path. A construction module is built specifically to construct a third path, thereby obtaining a target path constructed based on the second and third paths.

[0034] In one possible implementation, the fourth and fifth paths have been constructed, and the second edge cloud is managed by the second central cloud region management node through the fifth path. A construction module is built specifically to construct the third path, thereby obtaining the target path based on the fourth, fifth, and third paths.

[0035] In one possible implementation, a fourth path has been constructed, and a building module is specifically used to construct a sixth path, thereby obtaining a target path constructed based on the fourth and sixth paths.

[0036] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the target path is the path with the fewest hops among at least one candidate path, or the target path is the path with the maximum network bandwidth among at least one candidate path.

[0037] In one possible implementation, the first path is a direct path.

[0038] A fourth aspect of this application provides a cloud service-based path construction apparatus, which is the aforementioned second central cloud region management node. A first edge cloud is managed by the first central cloud region management node through a first path. The apparatus includes: a receiving module, configured to receive an access request for the first edge cloud sent by a client in the event of a failure of the first central cloud region management node or a failure of the path between the first central cloud region management node and the client; a determining module, configured to obtain at least one candidate path between the second central cloud region management node and the first edge cloud based on the access request; a selecting module, configured to select a target path from the at least one candidate path; and a constructing module, configured to construct the target path and manage the first edge cloud through the target path.

[0039] In related technologies, if the first central cloud area management node itself fails or the path between the first central cloud area management node and the client fails, all edge clouds managed by the first central cloud area management node will be affected, with a large impact radius. As can be seen from the above device: after staff discover a failure in the first central cloud area management node, they can input an access request to the second central cloud area management node through the client. Then, the second central cloud area management node can obtain candidate paths between itself and the first edge cloud based on the access request. Next, the second central cloud area management node selects a target path from the candidate paths. Finally, the second central cloud area management node constructs the target path. Based on this target path, communication can be achieved between the second central cloud area management node and the first edge cloud. That is, the second central cloud area management node can manage the first edge cloud in place of the first central cloud area management node through the target path, thereby improving the reliability and maintainability of the entire cloud service system and reducing the impact radius caused by the failure of the first central cloud area management node.

[0040] In one possible implementation, the target path is the sixth path between the second central cloud area management node and the first edge cloud, and the building module is specifically used to build the sixth path.

[0041] In one possible implementation, a second path has been constructed between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through the second path. The target path includes a fifth path between the second central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. A construction module is specifically used to construct the fifth path and the third path, thereby obtaining the target path constructed based on the fifth path and the third path.

[0042] In one possible implementation, a fourth path has been constructed between the second central cloud region management node and the first central cloud region. The target path includes the fourth path and the first path. The construction module is specifically used to directly obtain the target path constructed based on the fourth path and the first path.

[0043] In one possible implementation, a second path has been constructed between the first central cloud region management node and the second edge cloud, and the second edge cloud is managed by the first central cloud region management node through the second path. A fourth path has been constructed between the second central cloud region management node and the first central cloud region. The target path includes the fourth path, the second path, and a third path between the second edge cloud and the first edge cloud. A construction module is specifically used to construct the third path, thereby obtaining the target path constructed based on the fourth path, the second path, and the third path.

[0044] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the target path is the path with the fewest hops among at least one candidate path, or the target path is the path with the maximum network bandwidth among at least one candidate path.

[0045] In one possible implementation, the first path is a direct path.

[0046] A fifth aspect of this application provides a network device including a memory and a processor;

[0047] The memory stores code, and the processor is configured to execute the code. When the code is executed, the network device performs the method described in the second aspect or any possible implementation of the second aspect.

[0048] A sixth aspect of this application provides a computer storage medium storing a computer program that, when executed by a computer, causes the computer to perform the method as described in the first aspect, any possible implementation of the first aspect, the second aspect, or any possible implementation of the second aspect.

[0049] A seventh aspect of this application provides a computer program product storing instructions that, when executed by a computer, cause the computer to perform the method as described in the first aspect, any possible implementation of the first aspect, the second aspect, or any possible implementation of the second aspect.

[0050] In this embodiment, if the first central cloud region management node determines that the first path between the first central cloud region management node and the first edge cloud is faulty, it can first obtain candidate paths between the first central cloud region management node and the first edge cloud. The first edge cloud is managed by the first central cloud region management node through the first path. Then, the first central cloud region management node determines the target path from the candidate paths. This target path passes through relay points such as the second edge cloud and / or the second central cloud region management node. Finally, the first central cloud region management node constructs the target path. Based on this target path, the first central cloud region management node and the first edge cloud can maintain communication. That is, the first central cloud region management node can manage the first edge cloud through the target path, thereby improving the reliability and maintainability of the entire cloud service system, reducing the manpower costs required for operation and maintenance, and accelerating the logical and large-scale commercialization of the edge cloud to a certain extent. Attached Figure Description

[0051] Figure 1 A schematic diagram of the structure of the cloud service system provided in the embodiments of this application;

[0052] Figure 2A flowchart illustrating a cloud service-based path construction method provided in this application embodiment;

[0053] Figure 3 This is another schematic diagram of the cloud service system provided in the embodiments of this application;

[0054] Figure 4 This is another schematic diagram of the cloud service system provided in the embodiments of this application;

[0055] Figure 5 This is another schematic diagram of the cloud service system provided in the embodiments of this application;

[0056] Figure 6 Another flowchart illustrating the cloud service-based path construction method provided in this application embodiment;

[0057] Figure 7 This is another schematic diagram of the cloud service system provided in the embodiments of this application;

[0058] Figure 8 This is another schematic diagram of the cloud service system provided in the embodiments of this application;

[0059] Figure 9 A schematic diagram of a cloud service-based path building apparatus provided in this application embodiment;

[0060] Figure 10 Another schematic diagram of the cloud service-based path building apparatus provided in this application embodiment;

[0061] Figure 11 This is a schematic diagram of the structure of a network device provided in an embodiment of this application. Detailed Implementation

[0062] This application provides a cloud service-based path construction method and related equipment. When the direct path between the first central cloud area management node and the first edge cloud fails, a target path between the first central cloud area management node and the first edge cloud is constructed, so that communication can be maintained between the first central cloud area management node and the first edge cloud.

[0063] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0064] The embodiments of this application are applied to cloud service systems. Figure 1 A schematic diagram of the cloud service system provided in the embodiments of this application is shown below. Figure 1 As shown, this system distributes the infrastructure resources used for edge computing and manages these resources uniformly. The deployment points with more resources and concentrated locations are called the central cloud, while the deployment points with fewer resources and wider distribution are called edge clouds (or edge cloud nodes). Because edge clouds are more widely distributed, they are closer to the user's terminal devices; that is, edge clouds can be understood as cloud computing platforms close to terminal devices, while the central cloud can be understood as a cloud platform that manages multiple edge clouds uniformly.

[0065] In this cloud service system, each edge cloud can be deployed through a series of edge infrastructure resources, including but not limited to: distributed data centers, wireless server rooms or clusters, carrier communication networks, core network equipment, base stations, edge gateways, home gateways, computing devices, storage devices, and other edge devices and their corresponding network environments. It is understood that the location, capabilities, and included infrastructure resources of different edge clouds can be the same or different; no restrictions are imposed here.

[0066] For any edge cloud, various resources can be provided externally, such as computing resources like central processing units (CPUs) and graphics processing units (GPUs), storage resources like memory and hard drives, and network resources like bandwidth. Furthermore, the edge cloud can create corresponding cloud instances based on images, and provide various cloud computing services through these cloud instances. An image is the basic file required to create a cloud instance in the edge cloud. For example, it can be an image file containing the operating system, applications, or operational configurations needed to provide cloud computing services to users. It can be a file that conforms to edge cloud computing deployment requirements and is created according to a specific set of files in a certain format. Image formats are diverse, including virtual machine (VM) image files, container (Docker) image files, and various types of application packages. The image format can be related to the virtualization technology used by the cloud computing service, which is not limited here. Corresponding to the image, the implementation form of a cloud instance can be a virtual machine, a container, or an application, etc.

[0067] In addition, the terminal devices that access this cloud service system usually refer to the demand side of cloud computing services, such as smartphones, tablets, smartwatches, personal computers, etc. used by users.

[0068] It is worth noting that, due to the widespread distribution of edge clouds, to improve the management efficiency of the central cloud over edge clouds, the central cloud can be divided into multiple regions based on geographical factors, i.e., multiple central cloud regions. Each central cloud region's management surface (hereinafter referred to as the central cloud region management node) can be used to manage multiple edge clouds. Specifically, for any central cloud region management node, a direct path is established between the central cloud region management node and the edge clouds it manages; that is, a network connection is established between the central cloud region management node and the edge clouds it manages through the cloud-edge network, enabling communication between the central cloud region management node and the edge clouds it manages. For any two central cloud region management nodes, a direct path is established between the central cloud region management nodes; that is, a network connection is established between the central cloud region management nodes through the inter-cloud network, enabling communication between the central cloud region management nodes.

[0069] In this context, a direct path refers to the path between two central cloud regional management nodes, or between a central cloud regional management node and an edge cloud, where there are no other central cloud management nodes or other edge clouds. Routers, gateways, or other commonly used network communication devices can be set up between direct paths, but this embodiment of the invention does not limit this.

[0070] In related technologies, when the direct path between the central cloud regional management node and an edge cloud fails, cloud service provider personnel need to go to the site where the edge cloud is deployed for maintenance to ensure that the edge cloud continues to provide services to users. However, this does not guarantee that communication between the central cloud regional management node and the edge cloud will be restored. To solve this problem, this application provides a path construction method based on cloud services, which can be applied to... Figure 1 Before implementing this method, the cloud service system shown first introduces the application and deployment of edge cloud.

[0071] When a user needs cloud computing services, they can first send a cloud service application to the central cloud. This application information includes the user's data center location information, public network information, and local network information. Specifically, the data center location information includes the data center's geographical location and the model of the equipment used in the data center; the public network information includes the user's Internet Protocol (IP) address on the public network; and the local network information includes the network segment of the data center's local network (also known as the user's local network) and the model of the network equipment deployed on the local network.

[0072] Next, the central cloud, based on the data center's site information, allocates one or more edge clouds located within that data center to the user and sends an authorization request to the user, inquiring whether the user agrees to have the edge clouds connected to their local network. After the user agrees, the cloud service provider's staff can use public network information to construct a direct path between the edge cloud serving the user and the central cloud's regional management node that manages these edge clouds, and then connect the edge cloud serving the user to the user's local network based on local network information. It's important to note that even if the user does not agree to the authorization request, the cloud service provider's staff can still construct a direct path between the edge cloud serving the user and the central cloud's regional management node based on public network information, but will not connect the edge cloud serving the user to the user's local network.

[0073] Once the edge cloud serving this user is started, it can complete self-registration with the corresponding central cloud regional management node. Then, the edge cloud serving this user can be managed by the corresponding central cloud regional management node. At this point, the edge cloud deployment is complete.

[0074] Subsequently, if the direct path between an edge cloud and the central cloud regional management node that manages the edge cloud fails, or if the central cloud regional management node that manages the edge cloud fails, the cloud service-based path construction method provided in this application embodiment can be executed. For ease of explanation, the following will use two edge clouds in the same data center as an example, referred to as the first edge cloud and the second edge cloud, and the central cloud regional management node that manages the first edge cloud as the first central cloud regional management node. Once the first edge cloud is deployed, since it is managed by the first central cloud regional management node, a direct path (i.e., the aforementioned first path) has been established between the first edge cloud and the first central cloud regional management node, meaning they are connected via the cloud-edge network and can communicate.

[0075] Understandably, this data center can also deploy third edge clouds, fourth edge clouds, and so on. The central cloud can also include second central cloud regional management nodes, third central cloud regional management nodes, and so on. Direct paths have been established between different central cloud regional management nodes (for example, a direct path has been established between the first central cloud regional management node and the second central cloud regional management node, i.e., the aforementioned fourth path has been established). In other words, different central cloud regional management nodes have achieved network connectivity through the inter-cloud network and can communicate with each other.

[0076] When the direct path between the first edge cloud and the first central cloud regional management node fails, a path construction method based on cloud services provided in this application embodiment can be executed. Figure 2 A flowchart illustrating a cloud service-based path construction method provided in this application embodiment, the method comprising:

[0077] 201. If the first central cloud area management node determines that the direct path between the first central cloud area management node and the first edge cloud is faulty, then at least one candidate path between the first central cloud area management node and the first edge cloud is obtained.

[0078] The first central cloud regional management node can detect in real time whether the direct path between the first central cloud regional management node and the first edge cloud is faulty. Specifically, during operation, the first edge cloud can send its status information to the first central cloud regional management node in real time. This status information may include the operating status of the first edge cloud, its operating indicators, and the network quality between the first edge cloud and the first central cloud regional management node. When these data are within the normal range, the first central cloud regional management node can determine that the direct path between the first central cloud regional management node and the first edge cloud is normal. When these data are outside the normal range, the first central cloud regional management node can determine that the direct path between the first central cloud regional management node and the first edge cloud is faulty.

[0079] It should be understood that, optionally, the first central cloud area management node may determine whether the direct path between the first central cloud area management node and the first edge cloud is faulty without relying on the status information of the first edge cloud. For example, if the first central cloud area management node fails to receive data uploaded by the first edge cloud for a certain period of time, it can be determined that the direct path between the first central cloud area management node and the first edge cloud is faulty, etc.

[0080] After the first central cloud area management node determines that the direct path between it and the first edge cloud is faulty, it can perform path planning to obtain at least one candidate path between them. It should be noted that these candidate paths all originate from (or terminate at) the first central cloud area management node, terminate at (or originate from) the first edge cloud, and pass through relay points. Candidate paths can be presented in various ways; the following section will illustrate this with three examples:

[0081] Example 1: such as Figure 3 As shown ( Figure 3 (This is another schematic diagram of the cloud service system provided in this application embodiment). Assume that the data center where the first edge cloud is located also deploys a second edge cloud and a third edge cloud. The first, second, and third edge clouds are all connected to the local network of the data center (i.e., the first and second edge clouds are reachable from each other, and the first and third edge clouds are also reachable from each other). Furthermore, the second and third edge clouds are both managed by the first central cloud regional management node. Then, the first central cloud regional management node can plan two candidate paths: the first candidate path is: first central cloud regional management node - second edge cloud - first edge cloud; the second candidate path is: first central cloud regional management node - third edge cloud - first edge cloud.

[0082] Example 2: such as Figure 4 As shown ( Figure 4(This is another schematic diagram of the cloud service system provided in this application embodiment). Assume that the data center where the first edge cloud is located also deploys a second edge cloud and a third edge cloud. The central cloud includes a first central area and a second central cloud area management node. The first edge cloud, the second edge cloud, and the third edge cloud are all connected to the local network of the data center (i.e., the first edge cloud and the second edge cloud are reachable from each other), and the second edge cloud and the third edge cloud are managed by the second central cloud area management node. Then, the first central cloud area management node can plan three candidate paths: the first candidate path is: first central cloud area management node - second central cloud area management node - second edge cloud - first edge cloud; the second candidate path is: first central cloud area management node - second central cloud area management node - third edge cloud - first edge cloud; and the third candidate path is: first central cloud area management node - second central cloud area management node - first edge cloud.

[0083] Example 3: such as Figure 5 As shown ( Figure 5 (This is another schematic diagram of the cloud service system provided in this application embodiment). Assume that the data center where the first edge cloud is located also deploys a second edge cloud. The central cloud includes a first central region, a second central cloud region management node, and a third central cloud region management node. Neither the first nor the second edge cloud is connected to the local network of the data center (i.e., the first and second edge clouds are unreachable from each other), and the second edge cloud is managed by the first central cloud region management node. Then, the first central cloud region management node can plan two candidate paths: the first candidate path is: first central cloud region management node - second central cloud region management node - first edge cloud; the second candidate path is: first central cloud region management node - third central cloud region management node - first edge cloud.

[0084] It should be understood that Figure 3 The example shown is illustrative, illustrating only that the data center has three edge clouds and the central cloud comprises one region. It does not impose limitations on the number of edge clouds or the number of central cloud region management nodes. Similarly, Figure 4 The examples shown and Figure 5 The examples shown are similar, and will not be repeated here.

[0085] 202. The first central cloud area management node selects a target path from at least one candidate path. The target path has the first central cloud area management node and the first edge cloud as endpoints and the second central cloud area management node and / or the second edge cloud as relay points.

[0086] After obtaining candidate paths between the first central cloud area management node and the first edge cloud, the first central cloud area management node can select one candidate path as the target path between the two. Specifically, the factors considered by the first central cloud area management node in selecting the target path can be any one of network latency, hop count, and network bandwidth, or any combination of these factors. It is worth noting that the priority order of these three factors is network latency, hop count, and network bandwidth, in that order. Therefore, the first central cloud area management node can select the target path in several ways: (i) The first central cloud area management node determines the path with the lowest network latency from the candidate paths as the target path between the first central cloud area management node and the first edge cloud. (ii) The first central cloud area management node determines the path with the fewest hops from the candidate paths as the target path between the first central cloud area management node and the first edge cloud. (iii) The first central cloud area management node determines the path with the highest network bandwidth from the candidate paths as the target path between the first central cloud area management node and the first edge cloud. (iv) After determining that all candidate paths have the same network latency, the first central cloud area management node determines the path with the fewest hops as the target path between the first central cloud area management node and the first edge cloud. (v) After determining that all candidate paths have the same network latency, the first central cloud area management node determines the path with the largest network bandwidth as the target path between the first central cloud area management node and the first edge cloud. (vi) After determining that all candidate paths have the same hop count, the first central cloud area management node determines the path with the largest network bandwidth as the target path between the first central cloud area management node and the first edge cloud. (vii) After determining that all candidate paths have the same network latency and the same hop count, the first central cloud area management node determines the path with the largest network bandwidth as the target path between the first central cloud area management node and the first edge cloud, and so on.

[0087] For ease of explanation, the following text will assume that the first central cloud region management node selects the target path only considering network latency, and will combine... Figures 3 to 5 The examples shown are for illustrative purposes. For example... Figure 3 In the example shown, after comparing the network latency of the first candidate path and the second candidate path, the first central cloud area management node determines the first candidate path (first central cloud area management node - second edge cloud - first edge cloud) as the target path. Figure 4 In the example shown, after comparing the network latency of the first candidate path, the second candidate path, and the third candidate path, the first central cloud area management node determines the first candidate path (first central cloud area management node - second central cloud area management node - second edge cloud - first edge cloud) as the target path. Figure 5 In the example shown, after comparing the network latency of the first candidate path and the second candidate path, the first central cloud area management node determines the first candidate path (first central cloud area management node - second central cloud area management node - first edge cloud) as the target path.

[0088] 203. The first central cloud area management node constructs the target path and manages the first edge cloud through the target path.

[0089] The first central cloud region management node determines the target path between the first central cloud region management node and the first edge cloud, and can construct this target path. To further understand the aforementioned process, the following section combines... Figures 3 to 5 The following example illustrates this:

[0090] Figure 3 In the example shown, the first candidate path is determined as the target path. Within this path, a direct path between the first central cloud region management node and the second edge cloud (i.e., the aforementioned second path, which has already been pre-built since the second edge cloud is managed by the first central cloud region management node) has already been constructed. Therefore, the first central cloud region management node needs to construct a direct path between the first edge cloud and the second edge cloud (i.e., the aforementioned third path), thus obtaining the target path based on the direct path between the first central cloud region management node and the second edge cloud, and the direct path between the first edge cloud and the second edge cloud. In this way, the first central cloud region management node and the first edge cloud can maintain communication through this target path; that is, the first central cloud region management node can re-manage the first edge cloud through the target path.

[0091] Figure 4 In the example shown, the first candidate path is determined as the target path. In this path, a direct path has been established between the first central cloud region management node and the second central cloud region management node, and a direct path has also been established between the second central cloud region management node and the second edge cloud (i.e., the aforementioned fifth path, which was pre-established because the second edge cloud is managed by the second central cloud region management node). Therefore, the first central cloud region management node needs to establish a direct path between the first edge cloud and the second edge cloud, thus obtaining the target path constructed based on the direct paths between the first and second central cloud region management nodes, the second central cloud region management node and the second edge cloud, and the first edge cloud and the second edge cloud. In this way, the first central cloud region management node and the first edge cloud can maintain communication through this target path; that is, the first central cloud region management node can re-manage the first edge cloud through the target path.

[0092] Figure 5In the example shown, the first candidate path is determined as the target path. In this path, a direct path (i.e., the aforementioned fourth path) between the first central cloud region management node and the second central cloud region management node has already been constructed. Therefore, the first central cloud region management node needs to construct a direct path between the second central cloud region management node and the first edge cloud, thus obtaining the target path constructed based on the direct path between the first and second central cloud region management nodes, and the direct path between the second central cloud region management node and the first edge cloud. In this way, the first central cloud region management node and the first edge cloud can maintain communication through this target path; that is, the first central cloud region management node can re-manage the first edge cloud through the target path.

[0093] It should be noted that, in Figure 3 and Figure 4 In the example shown, since both the first edge cloud and the second edge cloud are connected to the local network of the data center, when the first central cloud area management node builds a direct path between the two, it can send the IP address and port of the first edge cloud in the local network to the second edge cloud, so that the second edge cloud and the first edge cloud can complete the network connection with the local network.

[0094] It should be understood that Figures 3 to 5 In the diagram, thin lines represent existing direct paths, while thick lines represent newly constructed direct paths.

[0095] In this embodiment, if the first central cloud region management node determines that the first path between the first central cloud region management node and the first edge cloud is faulty, it can first obtain candidate paths between the first central cloud region management node and the first edge cloud. The first edge cloud is managed by the first central cloud region management node through the first path. Then, the first central cloud region management node determines the target path from the candidate paths. This target path passes through relay points such as the second edge cloud and / or the second central cloud region management node. Finally, the first central cloud region management node constructs the target path. Based on this target path, the first central cloud region management node and the first edge cloud can maintain communication. That is, the first central cloud region management node can manage the first edge cloud through the target path, thereby improving the reliability and maintainability of the entire cloud service system, reducing the manpower costs required for operation and maintenance, and accelerating the logical and large-scale commercialization of the edge cloud to a certain extent.

[0096] When the first central cloud region management node fails, another cloud service-based path construction method provided in this application embodiment can be executed. Figure 6 Another flowchart illustrating the cloud service-based path construction method provided in this application embodiment, the method includes:

[0097] 601. In the event of a failure in the first central cloud area management node or a failure in the path between the first central cloud area management node and the client, the second central cloud area management node receives the access request sent by the client for the first edge cloud.

[0098] When a cloud service provider's staff's client cannot access the first central cloud area management node (i.e., the first central cloud area management node itself is not faulty, but there is a network failure between the first central cloud area management node and the staff's client) or discovers a fault in the first central cloud area management node itself, the staff can select the nearest second central cloud area management node as the entry point in the central cloud and send an access request to the second central cloud area management node through the client. This access request is used to instruct the second central cloud area management node to access the first edge cloud managed by the first central cloud area management node. In other words, the staff wants to communicate with the first edge cloud managed by the first central cloud area management node through the second central cloud area management node.

[0099] 602. The second central cloud area management node obtains at least one candidate path between the second central cloud area management node and the first edge cloud based on the access request.

[0100] After receiving an access request, the second central cloud area management node can perform path planning based on the request, thereby obtaining candidate paths between the second central cloud area management node and the first edge cloud. It should be noted that these candidate paths all start (or end) at the second central cloud area management node and end (or start) at the first edge cloud, and may or may not pass through relay points. Candidate paths can be presented in various ways, which will be illustrated below with two examples:

[0101] Example 1: such as Figure 7 As shown ( Figure 7 (This is another schematic diagram of the cloud service system provided in this application embodiment). Assume that the data center where the first edge cloud is located also deploys a second edge cloud. Both the first and second edge clouds are connected to the local network of the data center (i.e., the first and second edge clouds are reachable from each other), and the second edge cloud is managed by the first central cloud regional management node. Then, the second central cloud regional management node can detect the operating status of the first central cloud regional management node. If it is determined that the first central cloud regional management node itself is not faulty, the second central cloud regional management node can plan two candidate paths: the first candidate path is: second central cloud regional management node - first central cloud regional management node - first edge cloud; the second candidate path is: second central cloud regional management node - first central cloud regional management node - second edge cloud - first edge cloud.

[0102] Example 2: such as Figure 8 As shown ( Figure 8(This is another schematic diagram of the cloud service system provided in this application embodiment). Assume that the data center where the first edge cloud is located also deploys a second edge cloud. Both the first and second edge clouds are connected to the local network of the data center (i.e., the first and second edge clouds are reachable from each other), and the second edge cloud is managed by the first central cloud regional management node. Then, the second central cloud regional management node can detect the operating status of the first central cloud regional management node. If it is determined that the first central cloud regional management node itself is faulty, the second central cloud regional management node can plan two candidate paths: the first candidate path is: second central cloud regional management node - first edge cloud; the second candidate path is: second central cloud regional management node - second edge cloud - first edge cloud.

[0103] It should be understood that, Figure 7 Based on this, if the first edge cloud and / or the second edge cloud are not connected to the data center's internal interconnect network (i.e., the first edge cloud and the second edge cloud are not reachable from each other), then there is no second candidate path. Similarly, Figure 8 The examples shown are similar, and will not be repeated here.

[0104] It should also be understood that Figure 7 The example shown is illustrative, illustrating only the deployment of two edge clouds within a data center and the central cloud comprising two regions. It does not impose limitations on the number of edge clouds or the number of regional management nodes within the central cloud. Similarly, Figure 8 The examples shown are similar, and will not be repeated here.

[0105] 603. The second central cloud area management node selects the target path from at least the candidate paths.

[0106] After obtaining candidate paths between the second central cloud area management node and the first edge cloud, the second central cloud area management node can select one candidate path as the target path between the two. The factors considered by the second central cloud area management node in selecting the target path can be any one of network latency, hop count, and network bandwidth, or any combination of these factors. It is worth noting that the priority order of these three factors is network latency, hop count, and network bandwidth, in that order. For details on the process of the second central cloud area management node selecting the target path, please refer to [link to relevant documentation]. Figure 2 The relevant explanations regarding the selection of the target path by the first central cloud area management node in the illustrated embodiment will not be repeated here.

[0107] 604. The second central cloud area management node constructs the target path and manages the first edge cloud through the target path.

[0108] After the first central cloud region management node determines the target path between the second central cloud region and the first edge cloud, it can construct that target path. To further understand the aforementioned process, the following section will combine... Figure 7 and Figure 8 The following example illustrates this:

[0109] Figure 7 In the example shown, if the first candidate path is determined as the target path, the direct path between the first central cloud region management node and the second central cloud region management node (i.e., the aforementioned fourth path) has been constructed, and the direct path between the first central cloud region management node and the first edge cloud (i.e., the aforementioned first path, which has been pre-constructed because the first edge cloud is managed by the first central cloud region management node) has also been constructed. Therefore, the second central cloud region management node directly reuses these two direct paths to obtain the target path constructed based on the direct path between the first and second central cloud region management nodes, and the direct path between the first central cloud region management node and the first edge cloud. In this way, the second central cloud region management node and the first edge cloud can communicate through this target path, that is, the second central cloud region management node can manage the first edge cloud through this path.

[0110] Figure 7 In the example shown, if the second candidate path is determined as the target path, the direct path between the first central cloud region management node and the second central cloud region management node (i.e., the aforementioned fourth path) has already been constructed, and the direct path between the first central cloud region management node and the second edge cloud (i.e., the aforementioned second path, which has already been constructed in advance because the second edge cloud is managed by the first central cloud region management node) has also been constructed. Therefore, the second central cloud region management node needs to construct a direct path between the first edge cloud and the second edge cloud (i.e., the aforementioned third path), thereby obtaining the target path constructed based on the direct path between the first central cloud region management node and the second central cloud region management node, the direct path between the first central cloud region management node and the second edge cloud, and the direct path between the first edge cloud and the second edge cloud. In this way, the second central cloud region management node and the first edge cloud can communicate through this target path, that is, the second central cloud region management node can manage the first edge cloud through this path.

[0111] Figure 8 In the example shown, if the first candidate path is determined as the target path, the second central cloud area management node directly constructs a direct path between the second central cloud area management node and the first edge cloud (i.e., the aforementioned sixth path). In this way, the second central cloud area management node and the first edge cloud can communicate through this direct path, that is, the second central cloud area management node can manage the first edge cloud through this path.

[0112] Figure 8 In the example shown, if the second candidate path is determined as the target path, the second central cloud area management node directly constructs a direct path between itself and the second edge cloud (i.e., the aforementioned fifth path; since the second edge cloud was originally managed by the first central cloud area management node, this path was not previously constructed). It also constructs a direct path between the second edge cloud and the first edge cloud (i.e., the aforementioned third path), thus obtaining the target path constructed based on the direct paths between the second central cloud area management node and the second edge cloud, and between the second and first edge clouds. In this way, the second central cloud area management node and the first edge cloud can communicate through this target path; that is, the second central cloud area management node can manage the first edge cloud through this path.

[0113] It should be understood that Figure 7 and Figure 8 In the diagram, thin lines represent existing direct paths, while thick lines represent newly constructed direct paths.

[0114] In related technologies, if the primary central cloud area management node itself fails or the path between the primary central cloud area management node and the client fails, all edge clouds managed by the primary central cloud area management node will be affected, with a large impact radius. As can be seen from the above method: after staff discover a failure in the primary central cloud area management node, they can input an access request to the secondary central cloud area management node through the client. Then, the secondary central cloud area management node can obtain candidate paths between itself and the primary edge cloud based on the access request. Next, the secondary central cloud area management node selects a target path from the candidate paths. Finally, the secondary central cloud area management node constructs the target path. Based on this target path, communication can be achieved between the secondary central cloud area management node and the primary edge cloud. That is, the secondary central cloud area management node can manage the primary edge cloud in place of the primary central cloud area management node through the target path, thereby improving the reliability and maintainability of the entire cloud service system and reducing the impact radius caused by the failure of the primary central cloud area management node.

[0115] The above is a detailed description of the path construction method based on cloud services provided in the embodiments of this application. The following will introduce the path construction apparatus based on cloud services provided in the embodiments of this application. Figure 9 This is a schematic diagram of a cloud service-based path construction device provided in an embodiment of this application. This device is the aforementioned first central cloud region management node. The first edge cloud is managed by the device through a first path, such as... Figure 9 As shown, the device includes:

[0116] The acquisition module 901 is used to determine the failure of the first path and acquire at least one candidate path other than the first path between the first central cloud area management node and the first edge cloud; for example, the acquisition module 901 can execute Figure 2 Step 201 in the illustrated embodiment.

[0117] Selection module 902 is configured to select a target path from at least one candidate path, the target path including: a second path between the first central cloud area management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud; or, a fourth path between the first central cloud area management node and the second central cloud area management node, a fifth path between the second central cloud area management node and the second edge cloud, and the third path; or, the fourth path, and a sixth path between the second central cloud area management node and the first edge cloud; for example, selection module 902 may execute Figure 2 Step 202 in the illustrated embodiment.

[0118] Module 903 is used to build the target path and manage the first edge cloud through the target path. For example, Module 903 can execute... Figure 2 Step 203 in the illustrated embodiment.

[0119] In one possible implementation, a second path has been constructed, and the second edge cloud is managed by the first central cloud region management node through the second path. Module 903 is specifically used to construct a third path, thereby obtaining a target path constructed based on the second and third paths.

[0120] In one possible implementation, the fourth and fifth paths have been constructed, the second edge cloud is managed by the second central cloud area management node through the fifth path, and the construction module 903 is specifically used to construct the third path, thereby obtaining the target path constructed based on the fourth, fifth and third paths.

[0121] In one possible implementation, the fourth path has been constructed, and a construction module 903 is specifically used to construct the sixth path, thereby obtaining the target path constructed based on the fourth path and the sixth path.

[0122] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the target path is the path with the fewest hops among at least one candidate path, or the target path is the path with the maximum network bandwidth among at least one candidate path.

[0123] In one possible implementation, the first path is a direct path.

[0124] Figure 10Another schematic diagram of a cloud service-based path construction device provided in this application embodiment is shown, which is the aforementioned second central cloud region management node. The first edge cloud is managed by the first central cloud region management node through a first path. The device includes:

[0125] The receiving module 1001 is configured to receive an access request for the first edge cloud sent by the client in the event of a failure in the first central cloud area management node or a failure in the path between the first central cloud area management node and the client; for example, the receiving module 1001 may perform... Figure 6 Step 601 in the illustrated embodiment.

[0126] The determination module 1002 is used to obtain at least one candidate path between the second central cloud area management node and the first edge cloud based on the access request; for example, the determination module 1002 can execute... Figure 6 Step 602 in the illustrated embodiment.

[0127] Selection module 1003 is used to select a target path from at least one candidate path; for example, selection module 1003 can execute... Figure 6 Step 603 in the illustrated embodiment.

[0128] Module 1004 is used to build the target path and manage the first edge cloud through the target path. For example, Module 1004 can execute... Figure 6 Step 604 in the illustrated embodiment.

[0129] In one possible implementation, the target path is the sixth path between the second central cloud area management node and the first edge cloud, and the construction module 1004 is specifically used to construct the sixth path.

[0130] In one possible implementation, a second path has been constructed between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through the second path. The target path includes a fifth path between the second central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. The construction module 1004 is specifically used to construct the fifth path and the third path, thereby obtaining the target path constructed based on the fifth path and the third path.

[0131] In one possible implementation, a fourth path has been constructed between the second central cloud region management node and the first central cloud region. The target path includes the fourth path and the first path. The construction module 1004 is specifically used to directly obtain the target path constructed based on the fourth path and the first path.

[0132] In one possible implementation, a second path has been constructed between the first central cloud region management node and the second edge cloud, and the second edge cloud is managed by the first central cloud region management node through the second path. A fourth path has been constructed between the second central cloud region management node and the first central cloud region. The target path includes the fourth path, the second path, and a third path between the second edge cloud and the first edge cloud. The construction module 1004 is specifically used to construct the third path, thereby obtaining the target path constructed based on the fourth path, the second path, and the third path.

[0133] In one possible implementation, the target path is the path with the minimum network latency among at least one candidate path, or the target path is the path with the fewest hops among at least one candidate path, or the target path is the path with the maximum network bandwidth among at least one candidate path.

[0134] In one possible implementation, the first path is a direct path.

[0135] It should be noted that the information interaction and execution process between the modules / units of the above-mentioned device are based on the same concept as the method embodiment of this application, and the resulting technical effects are the same as those of the method embodiment of this application. For details, please refer to the description in the method embodiment shown above in the embodiment of this application, and it will not be repeated here.

[0136] Figure 11 This is a schematic diagram of the structure of a network device provided in an embodiment of this application. Figure 11 As shown, one embodiment of the network device in this application may include one or more central processing units 1101, memory 1102, input / output interfaces 1103, wired or wireless network interfaces 1104, and power supplies 1105.

[0137] The memory 1102 can be temporary or persistent storage. Furthermore, the central processing unit 1101 can be configured to communicate with the memory 1102 and execute a series of instructions stored in the memory 1102 on a network device.

[0138] In this embodiment, the central processing unit 1101 can execute the aforementioned... Figure 2 The operations performed in the first central cloud region in the illustrated embodiment or Figure 6 The specific operations performed in the second central cloud region in the illustrated embodiment will not be described in detail here.

[0139] In this embodiment, the specific functional module division in the central processing unit 1101 can be the same as described above. Figure 9 or Figure 10 The division of modules such as the first acquisition module, the second acquisition module, the determination module, and the construction module described in the text is similar, and will not be repeated here.

[0140] This application also relates to a computer storage medium, including computer-readable instructions, which, when executed, implement as follows: Figure 2 or Figure 6 The method described.

[0141] This application also relates to a computer program product containing instructions that, when run on a computer, cause the computer to perform actions such as... Figure 2 or Figure 6 The method described.

[0142] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0143] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0144] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0145] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A path construction method based on cloud services, characterized in that, The first edge cloud is managed by the first central cloud regional management node through the first path, and the method includes: The first central cloud area management node determines that the first path is faulty and obtains at least one candidate path between the first central cloud area management node and the first edge cloud, excluding the first path. The first central cloud region management node selects a target path from the at least one candidate path, the target path including: The second path between the first central cloud regional management node and the second edge cloud, and the third path between the second edge cloud and the first edge cloud; or, The fourth path between the first central cloud region management node and the second central cloud region management node, the fifth path between the second central cloud region management node and the second edge cloud, and the third path; or, The fourth path, and the sixth path between the second central cloud regional management node and the first edge cloud; The first central cloud area management node constructs the target path and manages the first edge cloud through the target path.

2. The method according to claim 1, characterized in that, The second path has been constructed, and the second edge cloud is managed by the first central cloud regional management node through the second path. The construction of the target path by the first central cloud regional management node includes: The first central cloud region management node constructs the third path, thereby obtaining the target path constructed based on the second path and the third path.

3. The method according to claim 1, characterized in that, The fourth and fifth paths have been constructed. The second edge cloud is managed by the second central cloud regional management node through the fifth path. The first central cloud regional node constructs the target path as follows: The first central cloud region management node constructs the third path, thereby obtaining the target path constructed based on the fourth path, the fifth path, and the third path.

4. The method according to claim 1, characterized in that, The fourth path has been constructed, and the construction of the target path by the first central cloud regional management node includes: The first central cloud region management node constructs the sixth path, thereby obtaining the target path constructed based on the fourth path and the sixth path.

5. The method according to any one of claims 1 to 4, characterized in that, The target path is the path with the minimum network latency among the at least one candidate path, or the target path is the path with the minimum number of hops among the at least one candidate path, or the target path is the path with the maximum network bandwidth among the at least one candidate path.

6. The method according to any one of claims 1 to 4, characterized in that, The first path is a direct path.

7. A path construction method based on cloud services, characterized in that, The first edge cloud is managed by the first central cloud regional management node through the first path, and the method includes: In the event of a failure of the first central cloud area management node or a failure of the path between the first central cloud area management node and the client, the second central cloud area management node receives the access request for the first edge cloud sent by the client. The second central cloud region management node obtains at least one candidate path between the second central cloud region management node and the first edge cloud based on the access request; The second central cloud region management node selects a target path from the at least one candidate path; The second central cloud regional management node constructs the target path and manages the first edge cloud through the target path.

8. The method according to claim 7, characterized in that, The target path is the sixth path between the second central cloud region management node and the first edge cloud. The second central cloud region management node constructs the target path by: The second central cloud region management node constructs the sixth path.

9. The method according to claim 7, characterized in that, A second path has been established between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through the second path. The target path includes a fifth path between the second central cloud region management node and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. The second central cloud region management node constructs the target path by: The second central cloud region management node constructs the fifth path and the third path to obtain the target path constructed based on the fifth path and the third path.

10. The method according to claim 7, characterized in that, A fourth path has been established between the second central cloud region management node and the first central cloud region. The target path includes the fourth path and the first path. The second central cloud region management node constructs the target path by: The second central cloud region management node directly obtains the target path constructed based on the fourth path and the first path.

11. The method according to claim 7, characterized in that, A second path has been established between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through the second path. A fourth path has been established between the second central cloud region management node and the first central cloud region. The target path includes the fourth path, the second path, and a third path between the second edge cloud and the first edge cloud. The second central cloud region management node constructs the target path by: The second central cloud region management node constructs the third path, thereby obtaining the target path constructed based on the fourth path, the second path, and the third path.

12. The method according to any one of claims 7 to 11, characterized in that, The target path is the path with the minimum network latency among the at least one candidate path, or the target path is the path with the minimum number of hops among the at least one candidate path, or the target path is the path with the maximum network bandwidth among the at least one candidate path.

13. The method according to any one of claims 7 to 11, characterized in that, The first path is a direct path.

14. A path construction device based on cloud services, characterized in that, The first edge cloud is managed by the device via a first path, the device comprising: The acquisition module is used to determine that the first path is faulty and acquire at least one candidate path between the device and the first edge cloud, other than the first path. A selection module is configured to select a target path from the at least one candidate path, the target path comprising: The second path between the device and the second edge cloud, and the third path between the second edge cloud and the first edge cloud; or, The fourth path between the device and the second central cloud regional management node, the fifth path between the second central cloud regional management node and the second edge cloud, and the third path; or, The fourth path, and the sixth path between the second central cloud regional management node and the first edge cloud; A building module is used to build the target path and manage the first edge cloud through the target path.

15. The apparatus according to claim 14, characterized in that, The second path has been constructed, and the second edge cloud is managed by the device through the second path. The construction module is specifically used to construct the third path, thereby obtaining the target path constructed based on the second path and the third path.

16. The apparatus according to claim 14, characterized in that, The fourth and fifth paths have been constructed, and the second edge cloud is managed by the second central cloud regional management node through the fifth path. The construction module is specifically used to construct the third path, thereby obtaining the target path constructed based on the fourth, fifth, and third paths.

17. The apparatus according to claim 14, characterized in that, The fourth path has been constructed. The construction module is specifically used to construct the sixth path, thereby obtaining the target path constructed based on the fourth path and the sixth path.

18. The apparatus according to any one of claims 14 to 17, characterized in that, The target path is the path with the minimum network latency among the at least one candidate path, or the target path is the path with the minimum number of hops among the at least one candidate path, or the target path is the path with the maximum network bandwidth among the at least one candidate path.

19. The apparatus according to any one of claims 14 to 17, characterized in that, The first path is a direct path.

20. A path construction device based on cloud services, characterized in that, The first edge cloud is managed by the first central cloud regional management node through a first path, and the device includes: The receiving module is configured to receive an access request for the first edge cloud sent by the client in the event that the first central cloud area management node fails or the path between the first central cloud area management node and the client fails. The determination module is used to obtain at least one candidate path between the device and the first edge cloud based on the access request; A selection module is used to select a target path from the at least one candidate path; A building module is used to build the target path and manage the first edge cloud through the target path.

21. The apparatus according to claim 20, characterized in that, The target path is the sixth path between the device and the first edge cloud, and the construction module is specifically used to construct the sixth path.

22. The apparatus according to claim 20, characterized in that, A second path has been constructed between the first central cloud regional management node and the second edge cloud. The second edge cloud is managed by the first central cloud regional management node through the second path. The target path includes a fifth path between the device and the second edge cloud, and a third path between the second edge cloud and the first edge cloud. The construction module is specifically used to construct the fifth path and the third path to obtain the target path constructed based on the fifth path and the third path.

23. The apparatus according to claim 20, characterized in that, A fourth path has been constructed between the device and the first central cloud region. The target path includes the fourth path and the first path. The construction module is specifically used to directly obtain the target path constructed based on the fourth path and the first path.

24. The apparatus according to claim 20, characterized in that, A second path has been constructed between the first central cloud region management node and the second edge cloud. The second edge cloud is managed by the first central cloud region management node through the second path. A fourth path has been constructed between the device and the first central cloud region. The target path includes the fourth path, the second path, and a third path between the second edge cloud and the first edge cloud. The construction module is specifically used to construct the third path, thereby obtaining the target path constructed based on the fourth path, the second path, and the third path.

25. The apparatus according to any one of claims 20 to 24, characterized in that, The target path is the path with the minimum network latency among the at least one candidate path, or the target path is the path with the minimum number of hops among the at least one candidate path, or the target path is the path with the maximum network bandwidth among the at least one candidate path.

26. The apparatus according to any one of claims 20 to 24, characterized in that, The first path is a direct path.

27. A network device, characterized in that, The network device includes a memory and a processor; The memory stores code, and the processor is configured to execute the code. When the code is executed, the network device performs the method as described in any one of claims 1 to 13.

28. A computer storage medium, characterized in that, The computer storage medium stores a computer program that, when executed by a computer, causes the computer to perform the method described in any one of claims 1 to 13.

29. A computer program product, characterized in that, The computer program product stores instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1 to 13.

Images