Management methods, devices, systems, electronic devices and storage media for edge clusters

Abstract

This disclosure provides a management method, apparatus, system, electronic device, and storage medium for edge clusters, relating to the field of cloud technology, and particularly to edge clusters, big data processing, and the Internet of Things (IoT). The specific implementation scheme is as follows: Obtain the target application associated with a target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster; send the target application to the edge cluster to install the target application on the target edge device. The management method proposed in this disclosure can flexibly realize the construction of edge clusters, enabling them to use their computing power to complete edge computing. Furthermore, this method is universal and can efficiently complete the deployment and management of edge clusters.

Description

Technical Field

[0001] This disclosure relates to the field of cloud technology, and in particular to edge clusters, big data processing, and the Internet of Things. Background Technology

[0002] Edge computing has become a popular computing architecture. This architecture deploys applications on edge devices to reduce data transmission and processing latency. Current edge computing frameworks primarily use containerization technology for deployment; however, for many devices with limited computing power, containerization still consumes excessive resources. Therefore, how to manage edge clusters remains a challenge. Summary of the Invention

[0003] This disclosure provides a method, apparatus, system, electronic device, and storage medium for managing edge clusters.

[0004] According to one aspect of this disclosure, a method for managing an edge cluster is provided, wherein a virtual cluster comprising at least one virtual node is stored in the cloud, the method comprising:

[0005] Retrieve the target application associated with the target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster;

[0006] Send the target application to the edge cluster so that the target application can be installed on the target edge device.

[0007] According to another aspect of this disclosure, a method for managing an edge cluster is provided, comprising:

[0008] Receive the target application associated with the target node sent from the cloud; wherein, the cloud stores a virtual cluster including at least one virtual node, and the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster;

[0009] Install the target application onto the target edge device.

[0010] According to another aspect of this disclosure, an edge cluster management apparatus is provided, wherein a virtual cluster including at least one virtual node is stored in the cloud, the apparatus comprising:

[0011] The acquisition module is used to acquire the target application associated with the target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node can be any virtual node in the virtual cluster;

[0012] The first sending module is used to send the target application to the edge cluster so that the target application can be installed on the target edge device.

[0013] According to another aspect of this disclosure, a management apparatus for an edge cluster is provided, comprising:

[0014] The receiving module is used to receive the target application associated with the target node sent by the cloud; wherein, the cloud stores a virtual cluster including at least one virtual node, and the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster;

[0015] The installation module is used to install the target application onto the target edge device.

[0016] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0017] At least one processor; and

[0018] The memory is communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods of any embodiment of the present disclosure.

[0020] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform a method according to any embodiment of this disclosure.

[0021] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements a method according to any embodiment of this disclosure.

[0022] According to another aspect of this disclosure, a management system for an edge cluster is provided, including the cloud and edge clusters described above.

[0023] In this embodiment of the disclosure, a virtual cluster is stored in the cloud, and each virtual node in the virtual cluster corresponds to a target application, that is, corresponding resources are deployed in the virtual cluster. When it is necessary to deploy the virtual cluster to an edge device to achieve the deployment of an edge cluster, the construction and management of the edge cluster can be flexibly realized.

[0024] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0025] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0026] Figure 1This is a schematic diagram of the architecture of a system for managing an edge cluster according to an embodiment of the present disclosure;

[0027] Figure 2 This is a flowchart illustrating an edge cluster management method according to an embodiment of the present disclosure;

[0028] Figure 3 This is a flowchart illustrating a method for managing an edge cluster according to another embodiment of this disclosure;

[0029] Figure 4 This is a schematic diagram of an edge cluster in process mode according to another embodiment of this disclosure;

[0030] Figure 5 This is a schematic diagram of a container-mode edge cluster according to another embodiment of this disclosure;

[0031] Figure 6 This is a flowchart illustrating a method for managing an edge cluster according to another embodiment of this disclosure;

[0032] Figure 7 This is a schematic diagram of a system for managing an edge cluster according to an embodiment of the present disclosure;

[0033] Figure 8 This is a schematic diagram illustrating the process of constructing and deploying a process-mode edge cluster according to an embodiment of this disclosure;

[0034] Figure 9 This is a schematic diagram illustrating the process of building and deploying a containerized edge cluster according to an embodiment of this disclosure;

[0035] Figure 10 This is a flowchart illustrating the management system for an edge cluster according to an embodiment of the present disclosure;

[0036] Figure 11 This is a schematic diagram of the structure of an edge cluster management device according to an embodiment of the present disclosure;

[0037] Figure 12 This is a schematic diagram of the structure of an edge cluster management device according to another embodiment of the present disclosure;

[0038] Figure 13 This is a block diagram of an electronic device used to implement the edge cluster management method of the embodiments of this disclosure. Detailed Implementation

[0039] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0040] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this disclosure, "multiple" means two or more, unless otherwise explicitly specified.

[0041] This disclosure proposes a system for managing edge clusters, such as... Figure 1 The diagram shows a system architecture for managing an edge cluster according to an embodiment of this disclosure. This application scenario includes a cloud platform 11 and an edge cluster 12, with communication connections between the cloud platform 11 and the edge cluster 12.

[0042] Cloud 11 includes at least one virtual cluster, such as Figure 1 As shown, there are virtual cluster 1 and virtual cluster 2. Each virtual cluster includes at least one virtual node.

[0043] Each virtual cluster can be ported to an edge cluster 12 composed of physical edge devices. Therefore, each virtual cluster also corresponds to an edge cluster 12.

[0044] like Figure 1 As shown, for each edge cluster 12, the edge cluster 12 includes at least one edge device, or edge node. Virtual nodes in the virtual cluster can be deployed to edge devices and exist as edge nodes. Based on the virtual cluster, virtual nodes can be flexibly migrated to any edge device, thereby enabling flexible deployment of the edge cluster 12.

[0045] The processing method for each edge cluster is the same in this embodiment. Taking one edge cluster as an example, the management method of the edge cluster in this embodiment will be described. Figure 2 The diagram shown is a flowchart of the method, including:

[0046] S201, the edge cluster sends an application retrieval request to the target edge device to obtain the target application required by the target edge device.

[0047] The target application can be a process application and / or a container application.

[0048] S202, the cloud uses the application to obtain the request and determines the target node in the virtual cluster that corresponds to the target edge device.

[0049] S203, Obtain the target application associated with the target node from the cloud.

[0050] In other words, within a virtual cluster, each virtual node has a corresponding virtual node identifier, and is associated with the application stored for that virtual node identifier. This application is used to deploy and run on the target edge device corresponding to that virtual node.

[0051] S204, the cloud sends the target application to the edge cluster.

[0052] S205, the edge cluster deploys the target application to the target edge device.

[0053] The above operations are performed on each edge device to complete the deployment of the virtual cluster to the physical edge cluster.

[0054] In some embodiments, the cloud is viewed as the master node, and the edge nodes in the edge cluster are slave nodes. Thus, the cloud directly manages each edge node in the edge cluster. However, this approach requires each edge device to connect to the cloud, which may result in significant network resource consumption.

[0055] In other embodiments, in order to reduce the network resources occupied by network transmission, the edge cluster includes a master node and at least one slave node. The master node communicates with the cloud, the master node communicates with each slave node through a local area network, and the slave nodes communicate with the cloud through the master node.

[0056] like Figure 1 As shown, in this embodiment of the disclosure, the virtual cluster has virtual master nodes and virtual slave nodes. When deployed to an edge device, the edge cluster 12 is configured with master nodes and slave nodes. The master node communicates with the cloud, and the slave nodes connect to the master node via a local area network. Figure 1 The slave nodes shown can include any of the edge nodes 1, ..., n. The master node acts as the control node, responsible for collecting the status information of each slave node in the edge cluster and the master node itself, and reporting it to the cloud so that the cloud can understand and manage the edge cluster 12. Requests from the cloud can be sent to any edge device through the master node.

[0057] In this embodiment of the disclosure, to flexibly manage edge clusters and fully utilize their computing power, each edge cluster can support any of the following modes: process mode, container mode, and hybrid mode. In process mode, each edge device runs a process application. Container mode allows for clustered management of edge nodes, offering good portability, with each edge device running a container application. Hybrid mode combines the advantages of both process and container modes. Users can define the operating mode of virtual nodes in the virtual cluster in the cloud based on their actual needs.

[0058] Therefore, in a process-based edge cluster, both the master and slave nodes run process-based applications; in a container-based edge cluster, both the master and slave nodes run container-based applications; and in a hybrid edge cluster, at least one slave node supports both process-based and container-based applications. In other words, in hybrid mode, some edge devices can run only process-based applications or only container-based applications. Thus, hybrid mode allows for the rational deployment of operating modes based on the performance of each edge device, thereby improving the resource utilization of the edge cluster.

[0059] For ease of understanding, the management methods implemented in the cloud and edge clusters are described in this disclosure embodiment:

[0060] 1. Edge cluster management method implemented in the cloud

[0061] This disclosure proposes a management method for edge clusters applied in the cloud, wherein, as described above, the cloud stores a virtual cluster including at least one virtual node, and the method is as follows: Figure 3 As shown, it includes:

[0062] S301, Obtain the target application associated with the target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster.

[0063] S302 sends the target application to the edge cluster so that the target application can be installed on the target edge device.

[0064] In this embodiment, a virtual cluster is stored in the cloud, and each virtual node in the virtual cluster corresponds to a target application; that is, corresponding resources are deployed in the virtual cluster. When it is necessary to deploy the virtual cluster to an edge device to achieve edge cluster deployment, the cloud can simply distribute the corresponding resources to the edge device. This management method can flexibly realize the construction of edge clusters, enabling the edge clusters to use their computing power to complete edge computing. Moreover, this method is universal and can efficiently complete the deployment of edge clusters.

[0065] In order to save communication resources in the cloud, in this embodiment of the disclosure, the edge cluster includes a master node and at least one slave node. The master node communicates with the cloud, the master node communicates with each slave node through a local area network, and the slave nodes communicate with the cloud through the master node.

[0066] In this embodiment of the disclosure, when the edge cluster includes a master node and at least one slave node, each slave node does not need to access the cloud, thereby saving cloud network resources. Furthermore, the management of slave nodes in the edge cluster can be completed by the master node. Compared to cloud-hosted edge clusters, this frees up cloud resources, enabling cloud management of more edge clusters and improving cloud utilization.

[0067] Based on the foregoing description, the edge cluster in this embodiment supports three operating modes: process-mode edge cluster, container-mode edge cluster, and hybrid-mode edge cluster.

[0068] In the case of a process-mode edge cluster, both the master node and the slave node run process applications.

[0069] In the case of a containerized edge cluster, both the master node and the slave node run containerized applications.

[0070] In the case of a hybrid edge cluster, at least one slave node in the edge cluster supports both process applications and container applications. In hybrid mode, container mode and process mode have independent running directories, thereby preventing data from interfering with each other.

[0071] In this embodiment of the disclosure, the edge cluster can be defined as any desired operating mode based on user needs, thereby enabling the cloud to be compatible with edge clusters in different operating modes and improving the flexibility of remote edge cluster management.

[0072] Regardless of the operating mode of the edge cluster, for any operating mode of the edge cluster, the deployment of the target application of the virtual node to the edge cluster can be carried out in accordance with the following implementation methods, including (1)-(2):

[0073] (1) Obtaining the target application associated with the target node in the virtual cluster can be implemented as follows:

[0074] Step A1: Receive the application acquisition request sent by the master node. The application acquisition request includes the node identifier of the target edge device.

[0075] The node identifier of the target edge device can be either the node identifier of the master node or the node identifier of the slave node; this embodiment of the disclosure does not limit this. That is, in a master-slave edge cluster, the master node can request the required target application from the edge cluster for itself or from the slave node.

[0076] Step A2: Select the virtual node corresponding to the node identifier as the target node, and obtain the application associated with the target node as the target application.

[0077] In this embodiment, upon receiving an application acquisition request from the master node, the cloud can parse the node identifier from the request to determine the virtual node where the application needs to be installed, and thus locate the target application to be deployed on the edge cluster. The cloud implements resource management for both the virtual cluster and the edge cluster, enabling flexible deployment of the target application to edge devices based on the needs of the edge cluster.

[0078] (2) Sending the target application to the edge cluster can be implemented as follows:

[0079] Send the target application to the master node of the edge cluster so that the master node can install the target application on the target edge device.

[0080] In this case, if the master node obtains the target application, the target application can be installed on the target edge device based on the local area network between the master node and the slave node.

[0081] For example, the master node establishes an HTTP (Hypertext Transfer Protocol) connection with the cloud, and sends an application retrieval request to the cloud based on this connection. The cloud receives the application retrieval request and parses out the node identifier. The node identifier is master node 1, so the virtual node 1 corresponding to master node 1 can be used as the target node, and the application associated with virtual node 1 can be sent to master node 1 as the target application.

[0082] For slave nodes, the master node sends an application retrieval request to the cloud. The cloud receives the application retrieval request and parses out the node identifier. If the node identifier is slave node 2, then the virtual node 2 corresponding to slave node 2 can be used as the target node. The application associated with virtual node 2 is then sent to the master node as the target application, and subsequently, the master node 1 can control slave node 2 to install the target application.

[0083] During implementation, users can deploy any application to a virtual cluster in the cloud via a client, and the virtual cluster can then further deploy the application to an edge cluster via a master node.

[0084] In addition, for existing applications, after updating the application, users can upgrade the corresponding application through the client, and then send the upgrade instruction to the master node of the edge cluster. Then, when the master node sends an application retrieval request, the upgraded application is deployed to the corresponding edge device in the edge cluster.

[0085] Therefore, from the cloud's perspective, it only needs to interface with the master node of the edge cluster, without needing to interface with each edge device in the edge cluster separately, making it easier for the cloud to manage the edge cluster. The deployment and updates of applications in the edge cluster (including version upgrades, decommissioning applications, and deploying new applications) can all be achieved through the master node.

[0086] In this embodiment of the disclosure, once the master node obtains the target application, the edge cluster where the master node is located can obtain the target application from the master node, thus saving network resources and improving the management efficiency of the edge cluster.

[0087] To facilitate a further understanding of the management methods for edge clusters in different operating modes according to the embodiments of this disclosure, these will be described separately below.

[0088] 1-1) Management of process-mode edge clusters

[0089] For edge clusters in process mode, users can first build the framework of the edge cluster themselves, that is, the network topology relationship between the master node and the slave node, and then the master node requests the target applications of each edge device from the cloud.

[0090] In other embodiments, the construction of the edge cluster can also rely on the cloud before obtaining the target application associated with the target node in the virtual cluster. For example, the cloud can generate master node installation instructions and slave node installation instructions, which users can execute on the master node and slave node respectively to facilitate the construction of the communication architecture within the edge cluster, as well as the communication architecture between the edge cluster and the cloud. The slave node requests the master node to send an installation request to the cloud based on the slave node installation instructions, and the master node sends an installation request to the cloud based on the master node installation instructions. For a process-mode edge cluster, the construction of any node can be implemented by the cloud as follows:

[0091] In the case of a process-mode edge cluster, the process-mode edge cluster is built in response to the installation request of the edge device sent by the master node.

[0092] In this implementation, the cloud can assist in building the communication architecture of the process-mode edge cluster, enabling the construction of the process-mode edge cluster to be automated to a certain extent, thereby improving the construction efficiency of the process-mode edge cluster and reducing the construction cost.

[0093] The construction of edge clusters in process mode can be divided into the construction of master nodes and the construction of slave nodes.

[0094] In some embodiments, after determining the role of the master node, the master node also needs to possess the functions of a master node. This is used to construct the master node in a process-mode edge cluster, which can be implemented as follows:

[0095] Step B1: If the installation request is for installing the master node, obtain the master node identifier from the installation request.

[0096] Step B2: Obtain the master node installation package associated with the virtual node corresponding to the master node identifier.

[0097] The form of the master node installation package associated with the virtual node corresponding to the master node identifier can be shown in Table 1:

[0098] Table 1

[0099] Master node identifier Virtual Node Master node installation package Master Node 1 Virtual Master Node 1 Master node installation package 1 Master Node 2 Virtual Master Node 2 Master node installation package 2 ... ... ...

[0100] In other words, the virtual cluster has a virtual master node. When an installation request requires the construction of a master node, the corresponding virtual master node will be determined based on the master node identifier, and then the corresponding master node installation package will be determined.

[0101] In this embodiment of the disclosure, the master node executes a master node installation instruction, which can define a master node identifier corresponding to the master node. This master node identifier is associated with a virtual master node identifier in the cloud, so that the master node installation package can be accurately determined even when managing multiple edge clusters in the cloud.

[0102] The cloud-generated master node installation instructions first establish a communication connection between the master node and the cloud, and then send the master node installation request to the cloud so that the cloud can send the master node installation package to the master node. Based on the master node installation package, the master node can possess the functions of a master node, such as how to communicate with the cloud and how to manage slave nodes.

[0103] Step B3: Send the master node installation package to the master node.

[0104] To enable the functionality of the master node, the master node installation package includes: the core application required by the master node, and the object storage service application; the core application is used to manage the slave nodes, and the object storage service application is used to store the processes of the slave nodes.

[0105] like Figure 4 As shown, in a process-mode edge cluster, the master node can allow core applications (BIE-core1) and object storage service applications (minio). BIE-core1 is used to communicate with the cloud, send management requests from the cloud to each slave node, manage the process applications in each slave node, and collect the status information of each node in the edge cluster and report it to the cloud.

[0106] For managing process applications in the edge cluster, the master node's BIE-core1 is used to download process applications from the cloud and store them in minio. Then, the corresponding slave nodes can retrieve process applications from minio, install and run them.

[0107] Correspondingly, to manage edge clusters in process mode, the cloud also has a corresponding object storage service. To distinguish it from the edge cluster's minio, the cloud's object storage service is called minio'. Based on this, after the master node establishes a communication connection with the cloud using the master node installation command, it first obtains the master node installation package from the cloud's minio', installs BIE-core1 and minio using this package, and then starts BIE-core1 on the master node using systemd. After BIE-core1 starts, the master node establishes a connection between BIE-core1 and the cloud and begins collecting data from the master node and reporting it to the cloud. This master node data includes information on hardware resources such as CPU, memory, GPU, and disk information, and may also include the status of the master node's own processes. If there are child processes, the status of these child processes is also reported to the cloud. The master node is also responsible for reporting the status of other slave nodes in its edge cluster. Furthermore, BIE-core1 is also responsible for communicating with slave nodes and downloading the corresponding target applications to minio for them.

[0108] In this embodiment of the disclosure, based on the installation request of the master node, the cloud can realize the core function of assisting in the construction of the master node, so that the construction of the master node is automated to a certain extent and the efficiency of edge cluster construction and deployment can be improved.

[0109] Similarly, in some embodiments, the slave nodes in a process-mode edge cluster can be implemented as follows:

[0110] Step C1: If the installation request is for installing a slave node, obtain the slave node identifier of the slave node from the installation request.

[0111] Step C2: Obtain the installation package of the slave node associated with the virtual node corresponding to the slave node identifier.

[0112] The form of the slave node installation package associated with the virtual node corresponding to the slave node identifier is similar to that of the aforementioned master node, and this embodiment does not limit this aspect.

[0113] Step C3 sends the slave node installation package to the master node, so that the master node can control the slave node to install the slave node installation package.

[0114] The slave node installation package is used to control the establishment of local area network communication between the slave node and the master node, and to control the slave node to obtain the slave node's process application from the master node's object storage service application.

[0115] For example, the installation instructions for a slave node can be copied from the cloud and executed on the slave node. Based on these instructions, the slave node can determine its role in the edge cluster, such as whether it performs the functions of slave node 1 or slave node 2. Similarly, the installation instructions define the slave node's identifier and the identifier of the master node to which the slave node should connect. The slave node executes the installation instructions, establishes local area network communication with the corresponding master node based on the master node identifier, and then sends an installation request carrying the slave node identifier to the master node. The master node then retrieves the slave node's installation package from the cloud based on this request. This completes the construction of the slave node's functionality.

[0116] In this embodiment, the installation package is sent via a local area network between the master and slave nodes, saving network resources. Simultaneously, cloud-assisted deployment of the slave node's communication functions also improves the efficiency of edge cluster construction.

[0117] The master node downloads the slave node installation package from the cloud and uploads it to minio. The slave node then synchronizes the slave node installation package from minio to its local machine, completing the installation of the slave node application. It also uses systemd to start the slave node application so that it can perform the slave node's functions.

[0118] After the slave application starts, it actively connects to the master node and monitors its own node status information in real time, sending this status information to the master node. The master node periodically aggregates the status information sent by each slave node and reports it to the cloud. At this time, the cloud will also be aware that a slave node has been added to the edge cluster, and can determine the slave node's resource usage, application deployment status, etc., based on the slave node's status information, so as to facilitate cloud management of the edge cluster.

[0119] 2) Management of containerized edge clusters

[0120] In some embodiments, a containerized edge cluster can be built first, and then the cloud can distribute specific edge computing applications to the edge cluster for installation and operation. For ease of management, when the edge cluster is in containerized mode, before obtaining the target application associated with the target node in the virtual cluster, the containerized edge cluster will first report the architecture information of the edge cluster so that the cloud can determine which applications to distribute to which edge devices. The cloud can implement this as follows:

[0121] Step D1: In the case of an edge cluster in container mode, receive the device identifier of the master node and the device identifier of the slave node sent by the edge cluster.

[0122] Step D2, in response to the operation of binding virtual nodes to each node, determines the virtual node corresponding to the device identifier of the master node and the virtual node corresponding to the device identifier of the slave node.

[0123] In some embodiments, edge clusters can be managed using K3S / K8S.

[0124] like Figure 5 The diagram illustrates the computing framework of a container-mode edge cluster and the cloud. Similar to the process-mode approach, the container-mode edge cluster includes BIE-core2. BIE-core2 functions similarly to BIE-core1 in the process-mode approach, communicating with the cloud, collecting status information from each edge device in the edge cluster, and reporting it to the cloud. It is also responsible for distributing the target applications from each virtual node in the virtual cluster to the corresponding target edge devices. Figure 5 As shown, the K3S Master acts as the primary node, with BIE-core2 installed. BIE-core2 connects to the API server to manage the slave nodes (K3Sworkers). This includes retrieving status information from each slave node from the API server and summarizing and reporting it to the cloud. It also includes distributing target applications to the slave nodes. Target applications retrieved from the cloud are stored as images in a registry, allowing slave nodes to retrieve, install, and run the corresponding applications from this registry.

[0125] During implementation, a containerized edge cluster can be built in the factory, and its edge computing capabilities can be constructed based on a virtual cluster in the cloud. Testing is then conducted. Once testing is successful, the virtual cluster can be deployed to the customer's edge cluster, thereby improving testing and deployment efficiency.

[0126] In this embodiment of the disclosure, the cloud can perceive each edge device in the container mode edge cluster, so as to manage and allocate the functions of each edge device and improve the flexibility of managing the container mode edge devices.

[0127] 3) Management of hybrid mode edge clusters

[0128] In hybrid mode, the edge cluster supports both container and process modes. Therefore, in hybrid mode, the edge cluster can be further divided into process clusters and container clusters. Process clusters support running in-process applications, while container clusters support running container applications. Both process and container clusters can be managed by a single master node. In terms of management implementation, process and container clusters are isolated.

[0129] Like the process-mode edge clusters described earlier, process clusters can also be built on the cloud. This can be implemented by, before obtaining the target application associated with the target node in the virtual cluster, in the case of a hybrid edge cluster, responding to the edge device installation request sent by the process control node in the master node, constructing a process cluster within the edge cluster for running process applications.

[0130] In this embodiment of the disclosure, in hybrid mode, in response to the installation request of the edge device sent by the process control node in the master node, a process cluster for running process applications is constructed in the edge cluster, which enables the management of the edge cluster and enhances the scalability of the edge cluster.

[0131] Similarly, like process-based edge clusters, process clusters include process control nodes (equivalent to the master node of a process-based edge cluster) and process child nodes (equivalent to the slave nodes of a process-based edge cluster). The construction of these two types of nodes will be explained in detail below.

[0132] Constructing a process control node in a process cluster can be implemented as follows:

[0133] Step E1: If the installation request is for installing a process control node, obtain the identity of the process control node from the installation request.

[0134] Step E2: Obtain the process control node installation package associated with the virtual node corresponding to the identity identifier of the process control node.

[0135] Step E3: Send the process control node installation package to the process control node.

[0136] The process control node installation package includes: the core application required by the process control node, and the object storage service application; the core application is used to manage the process child nodes of the running process application, and the object storage service application is used to store the process application.

[0137] In other words, after determining the role of the process control node, the functionality of the process master node needs to be implemented by obtaining the process control node installation package from the cloud. Similar to what was described earlier, the process control node runs the master node installation command. Running this command determines the connection method with the cloud and the node identifier of the process master node, so as to send an installation request to the cloud and obtain the process control node installation package from the cloud. After installing and starting the application, BIE-core1 and minio are obtained in the process cluster. Then, a connection is established between BIE-core1 and the cloud to execute the functions of the process control node.

[0138] In this embodiment, upon obtaining the process control node installation package associated with the virtual node corresponding to the process control node's identity identifier, the process control node installation package is then sent to the process control node, thus accurately pushing the process control node installation package to the process control node. This defines the functionality of the process control node and enables flexible construction of process clusters.

[0139] Similarly, the process child nodes of a running application in a process cluster can be implemented as follows:

[0140] Step F1: If the installation request is for installing process child nodes, obtain the identity of the process child node from the installation request.

[0141] Step F2: Obtain the process child node installation package associated with the virtual node corresponding to the identity identifier of the process child node.

[0142] Step F3 sends the process sub-node installation package to the process control node, so that the process control node controls the process sub-node to install the process sub-node installation package.

[0143] The process sub-node installation package is used to control the process sub-node to establish local area network communication with the process control node, and to control the process sub-node to obtain the process application of the process sub-node from the object storage service of the process control node.

[0144] Similarly, the process child node executes the slave node installation command to obtain the process child node installation package from the cloud via the process master node. After the process child node installation package is installed and started, the slave node application connects to the process control node to obtain the corresponding process application from the process control node's minio.

[0145] In this embodiment, the installation package of the process child node can be accurately sent based on this identity identifier to the process control node in the edge cluster where the process child node resides, laying the foundation for the subsequent acquisition of the installation package by the process child node. Furthermore, it can automatically build slave nodes to a certain extent, improving the deployment efficiency of the edge cluster.

[0146] In some embodiments, similar to containerized edge clusters, container clusters can be built first and then report the status of the edge clusters to the cloud, where the cloud is responsible for deploying specific edge computing applications.

[0147] In the case of a hybrid edge cluster, where the master node includes both container master nodes and container slave nodes that support container applications, before obtaining the target application associated with the target node in the virtual cluster, it can also be implemented as follows:

[0148] Step G1: In the case of a hybrid edge cluster, receive the device identifier of the container master node and the device identifier of the container slave node sent by the edge cluster.

[0149] Step G2, in response to the operation of binding virtual nodes to each node, determines the virtual node corresponding to the device identifier of the container master node and the virtual node corresponding to the device identifier of the container slave node.

[0150] Users can customize the mapping between virtual nodes and edge devices in the edge cluster, which can improve the flexibility of edge cluster deployment.

[0151] In this embodiment of the disclosure, the virtual nodes corresponding to each node can be quickly and accurately determined based on the device identifier, so as to provide a data foundation for subsequent application deployment of the cluster.

[0152] In this embodiment of the disclosure, the same edge device may deploy both container applications and process applications. Therefore, the resource usage of the edge device may be reported repeatedly. In order to handle resources, in this embodiment of the disclosure, when the edge cluster is a hybrid edge cluster and duplicate status information is received for any edge device in the edge cluster, a shutdown instruction is sent to the master node in the edge cluster to disable the sending of status information in process mode or the sending of status information in container mode.

[0153] In this embodiment of the disclosure, by instructing the cloud to disable the uploading of duplicate status information in a certain mode, the problem of duplicate reporting can be effectively solved, thereby saving network resources occupied by the communication between the edge cluster and the cloud.

[0154] 4) Other management matters for edge clusters

[0155] Regardless of the operating mode supported by the edge cluster, in this embodiment of the disclosure, the cloud can update the target application associated with the target node in response to an update request for the target application associated with the target node.

[0156] For example, the update may include deleting some applications, upgrading some applications, or adding new applications. Even applications that were originally running on node 1 can be redeployed to node 2. In this way, the same virtual cluster can be adapted to edge clusters with different hardware performance.

[0157] In this embodiment of the disclosure, the application deployment of the virtual cluster can be flexibly managed in order to meet different application needs.

[0158] 2. Related operations of edge clusters

[0159] In this embodiment of the disclosure, based on the same technical concept, a management method for edge clusters is proposed, such as... Figure 6 As shown, it includes:

[0160] This disclosure proposes a management method for edge clusters, such as... Figure 4 As shown, it includes:

[0161] S601, receive the target application associated with the target node sent by the cloud; wherein, the cloud stores a virtual cluster including at least one virtual node, and the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster.

[0162] S602 installs the target application onto the target edge device.

[0163] In this embodiment, a virtual cluster is stored in the cloud, and each virtual node in the virtual cluster corresponds to a target application; that is, corresponding resources are deployed in the virtual cluster. When it is necessary to deploy the virtual cluster to an edge device to achieve edge cluster deployment, the cloud can simply distribute the corresponding resources to the edge device. This management method can flexibly realize the construction of edge clusters, enabling the edge clusters to use their computing power to complete edge computing. Moreover, this method is universal and can efficiently complete the deployment of edge clusters.

[0164] In some embodiments, the edge cluster includes a master node and at least one slave node. The master node communicates with the cloud, the master node communicates with each slave node via a local area network, and the slave nodes communicate with the cloud through the master node.

[0165] In this embodiment of the disclosure, when the edge cluster includes a master node and at least one slave node, each slave node does not need to access the cloud, thereby saving cloud network resources. Furthermore, the management of the edge cluster slave nodes can be completed by the master node. Compared to cloud-hosted edge clusters, this frees up cloud resources, enabling cloud management of more edge clusters and improving cloud utilization.

[0166] In some embodiments, when the edge cluster is a process-mode edge cluster, both the master node and the slave node run process applications; when the edge cluster is a container-mode edge cluster, both the master node and the slave node run container applications; when the edge cluster is a hybrid-mode edge cluster, at least one slave node in the edge cluster supports both process applications and container applications.

[0167] In this embodiment of the disclosure, the edge cluster can be defined as any desired operating mode based on user needs, thereby enabling the cloud to be compatible with edge clusters in different operating modes and improving the flexibility of remote edge cluster management.

[0168] In some embodiments, the cloud can communicate with each edge device, and the edge device can send an application retrieval request to the cloud in order to retrieve the corresponding target application from the cloud.

[0169] In other embodiments, under the master node and slave node architecture as described above, before receiving the target application associated with the target node sent by the cloud, an application acquisition request can be sent to the cloud based on the master node. The application acquisition request includes the node identifier of the target edge device; the node identifier is used by the cloud to determine the target node.

[0170] The node identifier of the target edge device can be either the node identifier of a slave node or the node identifier of a master node. Therefore, the master node can obtain the corresponding application for itself or for the slave nodes.

[0171] Therefore, the nodes of the target application need to send their node identifiers to the cloud in the application acquisition request, so that the cloud can determine which virtual node is associated with the target application and thus accurately deploy the application on each node.

[0172] In some embodiments, installing the target application onto the target edge device can be implemented as follows:

[0173] When the target edge device is a slave node, the target application is stored in the storage application of the master node.

[0174] The master node instructs the slave node to retrieve the target application from the storage application.

[0175] In this embodiment of the disclosure, the management of slave nodes is based on the master node, which can save network resources and thus realize the management of slave nodes in a cluster manner, thereby saving resources.

[0176] In some embodiments, when the edge cluster is a process-mode edge cluster, the storage application is an object storage service application.

[0177] In the case of an edge cluster that is in container mode, the storage application is an image repository.

[0178] In the case of a hybrid edge cluster, the storage applications include object storage service applications and image repositories.

[0179] Among them, the object storage service application is used to store process applications, and the image repository is used to store images of container applications.

[0180] In other words, process-mode edge clusters distribute process applications to the corresponding edge devices through object storage service applications (minio). Container-mode edge clusters distribute container application images to the corresponding edge devices through an image repository. In contrast, good-mode edge clusters are divided into process clusters and container clusters, which are managed separately. Process clusters are managed in the same way as process-mode edge clusters, and container clusters are managed in the same way as container-mode edge clusters.

[0181] The master node in a hybrid edge cluster includes a process control node for managing the process cluster. It may also include a container control node for managing the container cluster. The container control node and the process control node can be deployed on the same edge device or on different edge devices.

[0182] In this embodiment of the disclosure, the storage method for applications is different under different operating modes. In view of this, the operating mode of the edge cluster can be determined based on the user-defined method, and then the storage method for applications in each edge device can be determined. This can achieve that different operating modes do not interfere with each other, and clusters in different operating modes are managed separately.

[0183] The edge clusters for each operating mode are further explained below.

[0184] 1) Management of process-mode edge clusters

[0185] In some embodiments, before receiving the target application associated with the target node sent from the cloud, the method further includes:

[0186] In the case of a process-mode edge cluster, the process-mode edge cluster is built based on the installation requests of edge devices sent by the master node.

[0187] The installation request for the master node carries the node identifier of the master node and the information of the target application that needs to be obtained, so that the cloud can quickly locate the virtual node corresponding to the target application.

[0188] In this embodiment, the edge cluster can be built based on the installation request of the master node, enabling the edge cluster to run the target application and achieving the portability of the target application. Furthermore, the construction of the process-mode edge cluster can be automated to a certain extent, improving the construction efficiency of the edge cluster.

[0189] The construction of a process-mode edge cluster can be divided into building the master node and building the slave node. The following is a detailed explanation of the two node construction methods:

[0190] In some embodiments, the master node in a process-mode edge cluster can be implemented as follows:

[0191] Step H1: If the master node is installed, generate an installation request containing the master node identifier.

[0192] Step H2: Send an installation request.

[0193] Step H3: Receive the master node installation package returned by the cloud in response to the installation request.

[0194] Step H4: Build the master node based on the master node installation package.

[0195] The master node installation package includes: the core application required by the master node, and the object storage service application; the core application is used to manage slave nodes, and the object storage service application is used to store the process application of the slave nodes.

[0196] In this embodiment of the disclosure, the master node installation package of the master node can be quickly located based on the master node identifier in the installation request, so as to accurately build the function of the master node and complete the construction of the process mode master node.

[0197] In some embodiments, the slave nodes in a process-mode edge cluster include:

[0198] Step I1: In the case of installing a slave node, generate an installation request containing the slave node's identifier.

[0199] Step I2: Send the installation request.

[0200] In this embodiment of the disclosure, the slave node can generate an installation request containing the slave node identifier and send it to the master node. The master node then sends it to the cloud. Alternatively, the slave node can notify the master node that it needs to send an installation request to the cloud, and the master node will then generate the installation request corresponding to that slave node and send it to the cloud.

[0201] Step I3: Receive the slave node installation package returned by the cloud in response to the installation request.

[0202] Step I4: Build the slave node based on the slave node installation package.

[0203] The slave node installation package is used to control the establishment of local area network communication between the slave node and the master node, and to control the slave node to obtain the slave node's process application from the master node's object storage service application.

[0204] In this embodiment of the disclosure, the slave node installation package can be accurately obtained from the cloud based on the slave node's node identifier, thereby accurately deploying the slave node's functionality.

[0205] In this embodiment of the disclosure, the master node can be installed based on the master node installation instructions described above, and the slave node can be installed based on the slave node installation instructions described above. These details will not be repeated here.

[0206] 2) Management of containerized edge clusters

[0207] Similarly, a containerized edge cluster can be built first, and then its device information can be reported to the cloud so that the cloud can allocate functions to each edge device. In some embodiments, before receiving the target application associated with the target node sent by the cloud, in the case of a containerized edge cluster, the device identifiers of the master node and the slave nodes of the edge cluster are sent.

[0208] In some embodiments, the device identifiers of the master node and the slave node are sent to the cloud to facilitate the flexible deployment of the correspondence between virtual nodes and edge devices in the cloud, laying the foundation for the automatic deployment and distribution of applications to each edge device.

[0209] After establishing the mapping between edge devices and virtual nodes, it is possible to base it on... Figure 6 The process shown completes the application deployment for each edge device.

[0210] 3) Management of hybrid mode edge clusters

[0211] Consistent with the cloud-based description, hybrid edge clusters include both process clusters and container clusters.

[0212] The process cluster can be built with cloud assistance. In some embodiments, before receiving the target application associated with the target node sent by the cloud, the method further includes: in the case of a hybrid edge cluster, building a process cluster in the edge cluster for running the process application based on the installation request of the edge device sent by the process control node in the master node.

[0213] In this embodiment of the disclosure, the process cluster can complete the construction of the functions of each node in the process cluster based on the interaction between the master node and the cloud, thereby realizing the automatic deployment of the cluster.

[0214] A process cluster includes a process control node and process child nodes. Constructing the process control node in a process cluster includes:

[0215] Step J1: If the installation request is for installing a process control node, generate an installation request containing the identity of the process control node.

[0216] Step J2: Send an installation request.

[0217] Step J3: Receive the process control node installation package returned by the cloud in response to the installation request.

[0218] Step J4: Build the process control node based on the process control node installation package.

[0219] The process control node installation package includes: the core application required by the process control node, and the object storage service application; the core application is used to manage the process child nodes of the running process application, and the object storage service application is used to store the process application.

[0220] Consistent with the previous text, the process control node, as the master node of the process cluster, runs the master node installation instructions to establish a connection with the cloud and inform the cloud of its identity as the process control node so that the cloud can perform subsequent operations.

[0221] In this embodiment of the disclosure, the process control node notifies the cloud of its role through its identity identifier, enabling the cloud to accurately determine the resources of the process control node (i.e., the process control node installation package) in order to improve the functionality of the process control node.

[0222] In some embodiments, constructing process child nodes of running process applications in a process cluster includes:

[0223] Step K1: In the case of installing process sub-nodes, generate an installation request containing the identity of the process sub-nodes.

[0224] Step K2: Send the installation request.

[0225] The installation request can be generated by the process sub-node and sent to the cloud by the master node, or the process sub-node can notify the process control node, which will then generate and send the installation request to the cloud.

[0226] Step K3: Receive the process sub-node installation package returned by the cloud in response to the installation request.

[0227] Step K4: Build process sub-nodes based on the process sub-node installation package.

[0228] The process sub-node installation package is used to control the process sub-node to establish local area network communication with the process control node, and to control the process sub-node to obtain the process application of the process sub-node from the object storage service of the process control node.

[0229] In this embodiment of the disclosure, the identity of the process sub-node is clearly identified through the installation request, so that the cloud can accurately deliver the process sub-node installation package, thereby improving the functionality of the process sub-node.

[0230] Similar to container mode, in the case of a hybrid edge cluster, the master node includes a container master node that supports container applications and container slave nodes that also support container applications. Before receiving the target application associated with the target node sent from the cloud, it can also be implemented as follows:

[0231] Send the device identifier of the container master node that supports container mode and the device identifier of the container slave node that supports container mode.

[0232] In this embodiment of the disclosure, the use of the device identifier of the container master node and the device identifier of the container slave node that supports container mode enables the cloud to quickly understand the architecture of the container cluster and determine the correspondence between virtual nodes and edge devices, laying the foundation for easy deployment of container applications.

[0233] 4) Other management matters

[0234] Regardless of the operating mode supported by the edge cluster, this embodiment of the disclosure also includes: obtaining the status information of each edge device in the edge cluster based on the master node; and sending the status information to the cloud.

[0235] In this embodiment of the disclosure, the status information of each node is sent to the cloud so that the cloud can understand the status of each node in real time, thereby enabling edge cloud to manage the edge cluster.

[0236] In addition, in hybrid mode, the edge cluster may repeatedly report the status information of the same edge device. In order to save network resources, the edge cluster master node can receive a shutdown instruction sent by the cloud to disable the sending of status information in process mode or container mode.

[0237] Therefore, when the same status information is repeatedly reported in both process mode and container mode, the problem of repeated reporting can be solved by using a shutdown instruction in the cloud.

[0238] In summary, the operating modes of the edge cluster in this embodiment can be divided into process mode, container mode, and hybrid mode. The specific implementation of each mode is described in detail below:

[0239] like Figure 7As shown, the BIE Cloud module is deployed in the cloud to manage the edge cluster. For the process-mode edge cluster, BIE-Core1 and minio are deployed on the master node master1. For the container-mode edge cluster, BIE-Core2 and the image repository are deployed on the master node master2. In each operating mode of the edge cluster, the slave nodes are all BIE Workers, or simply workers. The worker in the process-mode cluster is labeled BIEWorker1, and the worker in the container-mode cluster is labeled BIEWorker2. The hybrid mode includes both process-mode and container-mode clusters.

[0240] The BIE Cloud module provides an open API that users can call, allowing them to create virtual nodes and update applications in the cloud.

[0241] BIE-Core1, also known as master1, actively establishes connections with the cloud and synchronizes application changes there. It monitors the running status of deployed applications and node status. It creates a pair of RPC servers and clients to facilitate LAN communication with slave nodes. The master acts as the RPC server, and the worker acts as the RPC client. It receives resource reports from workers, aggregates and uploads them, and synchronizes application processes to the workers. In process mode, the miniio service started by BIE-core1 stores the process packages of applications deployed from the cloud, allowing workers on the LAN to access and run these packages.

[0242] BIE Worker1: Connects to the master node, periodically reports data, synchronizes applications, and executes on the current device.

[0243] BIE-Core2 (master2) actively establishes a connection with the cloud and synchronizes application changes there. It monitors the running status and node status of edge devices with deployed applications in real time and reports the data accordingly. It connects to the API server controller in the edge cluster to manage slave nodes and manage cloud-deployed applications within the edge cluster. The image repository service started by BIE-core2 stores images of cloud-deployed applications, allowing BIEWorker1 within the local area network to access these images.

[0244] BIE Worker2: Connects to the master2 node, periodically reports data, synchronizes applications, and executes on the current device.

[0245] 1) In the case of an edge cluster operating in process mode, both the master node and the slave node run process applications.

[0246] The construction and deployment of applications in process-mode edge clusters can be implemented as follows: Figure 8 As shown:

[0247] S801: The master node executes the master node installation instructions copied from the cloud and establishes a connection with the cloud.

[0248] S802: Based on the master node installation instructions, the master node generates a master node installation request containing the master node identifier and sends the installation request to the cloud.

[0249] S803, the cloud obtains the master node identifier of the master node from the master node installation request.

[0250] S804: Obtain the master node installation package associated with the virtual node corresponding to the master node identifier from the cloud; and send the master node installation package to the master node.

[0251] In S805, the master node receives the master node installation package returned by the cloud in response to the installation request; and builds the master node based on the master node installation package. After the construction is complete, the master node reports its own status information to the cloud.

[0252] The master node installation package includes: the core application required by the master node, and the object storage service application; the core application is used to manage slave nodes, and the object storage service application is used to store the process application of the slave nodes.

[0253] That is, the master node completes the installation of BIE-CORE1 and minio. After installation, BIE-CORE1 establishes a communication connection with the cloud.

[0254] S806 executes the slave node installation instructions from the cloud, generates a slave node installation request containing the slave node's identifier, and sends it to the master node.

[0255] S807, the master node sends a slave node installation request to the cloud.

[0256] S808, the cloud obtains the slave node identifier of the slave node from the slave node installation request.

[0257] S809 retrieves the slave node installation package associated with the virtual node corresponding to the slave node identifier from the cloud, and then sends the slave node installation package to the master node.

[0258] In the S810, the master node receives the slave node installation package returned by the cloud in response to the installation request, and stores the slave node installation package in minio.

[0259] S811, the master node controls the slave node to obtain and install the slave node installation package from minio. This completes the construction of the slave node. The slave node will report status information to the master node.

[0260] In other words, the slave node installs and starts BIE worker1 based on the slave node installation package. BIE worker1 communicates with the master node via the local area network, can obtain process applications from the master node, and report its own status to the master node. Requests issued from the cloud will also be forwarded to BIE worker1 on the slave node through the master node.

[0261] Therefore, based on the slave node installation package, it is possible to control the establishment of local area network communication between the slave node and the master node, and to control the slave node to obtain the slave node's process application from the master node's object storage service application.

[0262] In S812, all application requests generated by the master node or slave node are sent to the cloud by the master node.

[0263] S813 receives application acquisition requests sent by the master node in the cloud. The application acquisition requests include the node identifier of the target edge device.

[0264] S814, the cloud will use the virtual node corresponding to the node identifier as the target node, and obtain the application associated with the target node as the target application.

[0265] S815: The cloud sends the target application to the master node of the edge cluster.

[0266] S816, the master node stores the target application in minio.

[0267] S817: When the target edge device is a slave node, the master node controls the slave node to obtain the target application from minio.

[0268] For example, a process package for the target application (hereinafter referred to as app1) is created in the cloud and uploaded to the cloud's minio. Simultaneously, app1 is bound to the edge node (worker1) of the edge cluster (cluster1). At the same time, the cloud sends a message to the master node in (cluster1). The message includes: resource type, edge cluster name, name of the edge node (i.e., the target edge device) in the edge cluster, the edge cluster's operating mode (the first field, such as native, indicates process mode, and the second field, such as kube, indicates container mode), the minio address of the process package (cloud address), and the entry point of the program within the process package.

[0269] After receiving the message, the master node will download app1 from the cloud address and upload it to the master node's minio. At the same time, it will modify the address of app1 in the message to the minio address within the local area network and pass the message to worker1.

[0270] After receiving the message, worker1 downloads app1 using the local network's miniio and starts app1 using systemd. It then transmits the program's running status to the master node, which in turn reports it to the cloud. The cloud can thus be aware of the running status of app1 on worker1 within the edge cluster.

[0271] 2) In the case of a containerized edge cluster, both the master node and the slave node run containerized applications.

[0272] In container mode, users first build an edge cluster on edge devices, and a corresponding virtual cluster is created in the cloud. The constructed edge cluster is still managed using K3S or K8S, with the master node reporting the status of the edge cluster to the cloud.

[0273] The master node sends the node identifier of the BIE worker2 that needs to run in container mode to the cloud. If the BIE-core2 image is pulled from the cloud, a BIE-core2 that supports container mode is started and connected to the edge K3Sapi server. This method enables the edge node to connect to the cloud.

[0274] For persistence, container mode can map storage resources within the container to edge devices. Container mode relies entirely on K3S / K8S to manage the edge cluster; BIE-core is only responsible for collecting and reporting, while K3S handles the startup information for application deployment.

[0275] In a containerized edge cluster that includes at least one slave node, it is necessary to obtain the image application from the slave node. Therefore, it is necessary to start an image repository to download the container specified by the containerized application from the cloud and upload the image.

[0276] like Figure 9 The diagram illustrates the management method for containerized edge clusters, including:

[0277] S901, the master node sends the device identifiers of the master node and the slave nodes of the edge cluster to the cloud.

[0278] S902, the cloud receives the device identifiers of the master node and slave node sent by the edge cluster.

[0279] S903, in response to the cloud's operation of binding virtual nodes to each node, determines the virtual node corresponding to the device identifier of the master node and the virtual node corresponding to the device identifier of the slave node.

[0280] At this point, the cloud can distribute the BIE-core2 image, install it on the master node, and start it. After starting, BIE-core2 establishes a connection with the cloud, reports its own status information, and manages the image repository. BIE-core2 also establishes a connection with the API server to manage slave nodes.

[0281] S904, the master node sends an application retrieval request to the cloud, and the application retrieval request includes the node identifier of the target edge device.

[0282] The S905 receives an application acquisition request sent by the master node in the cloud. The application acquisition request includes the node identifier of the target edge device.

[0283] S906, the cloud will use the virtual node corresponding to the node identifier as the target node, and obtain the application associated with the target node as the target application.

[0284] S907: The cloud sends the target application to the master node of the edge cluster.

[0285] S908: The master node stores the image of the target application in the master node's image repository. It then controls the slave nodes to retrieve the target application from this image repository and install it.

[0286] 3) In the case of a hybrid edge cluster, at least one slave node in the edge cluster supports both process applications and container applications.

[0287] Since hybrid mode slave nodes can support both process-based and container-based applications, the execution method for slave nodes supporting process-based applications is the same as that of the process mode, and the execution method for slave nodes supporting container-based applications is the same as that of the container mode. Further details will not be provided here.

[0288] In the hybrid mode, container mode is mapped to the first directory, while process mode uses the second directory as the root directory. The runtime directories for container clusters and process clusters are separate.

[0289] Based on the same technical concept, embodiments of this disclosure also provide a management system for edge clusters, wherein, as Figure 10 As shown, the cloud storage contains a virtual cluster including at least one virtual node. The system includes:

[0290] S1001, obtain the target application associated with the target node in the virtual cluster from the cloud; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster.

[0291] S1002, the cloud sends the target application to the edge cluster so that the target application can be installed on the target edge device.

[0292] S1003, the edge cluster receives the target application associated with the target node sent from the cloud.

[0293] S1004, the edge cluster installs the target application onto the target edge device.

[0294] It should be noted that the operations performed by the cloud and edge clusters have been explained above and will not be repeated here.

[0295] Based on the same technical concept, this disclosure also provides an edge cluster management device 1100, wherein a virtual cluster including at least one virtual node is stored in the cloud, and the device is as follows: Figure 11 As shown, it includes:

[0296] The acquisition module 1101 is used to acquire the target application associated with the target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster;

[0297] The first sending module 1102 is used to send the target application to the edge cluster so that the target application can be installed on the target edge device.

[0298] In some embodiments, the edge cluster includes a master node and at least one slave node. The master node communicates with the cloud, the master node communicates with each slave node via a local area network, and the slave nodes communicate with the cloud through the master node.

[0299] In some embodiments, when the edge cluster is a process-mode edge cluster, both the master node and the slave node run process applications.

[0300] In the case of a containerized edge cluster, both the master node and the slave node run containerized applications.

[0301] In the case of a hybrid edge cluster, at least one slave node in the edge cluster supports both process applications and container applications.

[0302] In some embodiments, the acquisition module is configured to:

[0303] Receive the application acquisition request sent by the master node. The application acquisition request includes the node identifier of the target edge device.

[0304] The virtual node corresponding to the node identifier is taken as the target node, and the application associated with the target node is taken as the target application.

[0305] In some embodiments, the first transmitting module is configured to:

[0306] Send the target application to the master node of the edge cluster so that the master node can install the target application on the target edge device.

[0307] In some embodiments, before obtaining the target application associated with the target node in the virtual cluster, a first construction module is further included, for:

[0308] In the case of a process-mode edge cluster, the process-mode edge cluster is built in response to the installation request of the edge device sent by the master node.

[0309] In some embodiments, in response to an installation request for an edge device sent by the master node, a first construction module is configured to build the master node in a process-mode edge cluster, wherein the first construction module is used to:

[0310] If the installation request is for installing the master node, obtain the master node identifier from the installation request;

[0311] Obtain the master node installation package associated with the virtual node corresponding to the master node identifier;

[0312] Send the master node installation package to the master node;

[0313] The master node installation package includes: the core application required by the master node, and the object storage service application; the core application is used to manage slave nodes, and the object storage service application is used to store the process application of the slave nodes.

[0314] In some embodiments, in response to an installation request for an edge device sent by the master node, a first building module is configured to build slave nodes in a process-mode edge cluster, wherein the first building module is used to:

[0315] If the installation request is for installing a slave node, obtain the slave node identifier from the installation request;

[0316] Retrieve the installation package of the slave node associated with the virtual node corresponding to the slave node identifier;

[0317] The slave node installation package is sent to the master node so that the master node can control the slave node to install the slave node installation package.

[0318] The slave node installation package is used to control the establishment of local area network communication between the slave node and the master node, and to control the slave node to obtain the slave node's process application from the master node's object storage service application.

[0319] In some embodiments, before obtaining the target application associated with the target node in the virtual cluster, a second building module is further included, which is also used for:

[0320] In the case of a hybrid edge cluster, in response to the installation request of the edge device sent by the process control node in the master node, a process cluster for running process applications is built in the edge cluster.

[0321] In some embodiments, in response to an installation request for an edge device sent by the process control node in the master node, a second construction module is constructed to build the process control node in the process cluster, and is configured to:

[0322] If the installation request is for installing a process control node, obtain the identity of the process control node from the installation request;

[0323] Obtain the process control node installation package associated with the virtual node corresponding to the process control node's identity identifier;

[0324] Send the process control node installation package to the process control node;

[0325] The process control node installation package includes: the core application required by the process control node, and the object storage service application; the core application is used to manage the process child nodes of the running process application, and the object storage service application is used to store the process application.

[0326] In some embodiments, in response to an installation request for an edge device sent by the process control node in the master node, a process child node for a running application in the process cluster is constructed. The second construction module is configured to:

[0327] If the installation request is for installing process child nodes, obtain the identity of the process child node from the installation request;

[0328] Obtain the installation package of the process child node associated with the virtual node corresponding to the identity identifier of the process child node;

[0329] Send the process child node installation package to the process control node so that the process control node controls the process child node to install the process child node installation package.

[0330] The process sub-node installation package is used to control the process sub-node to establish local area network communication with the process control node, and to control the process sub-node to obtain the process application of the process sub-node from the object storage service of the process control node.

[0331] In some embodiments, before obtaining the target application associated with the target node in the virtual cluster, a first identifier determination module is further included, configured to:

[0332] In the case of a containerized edge cluster, receive the device identifier of the master node and the device identifier of the slave node sent by the edge cluster;

[0333] In response to the operation of binding virtual nodes to each node, determine the virtual node corresponding to the device identifier of the master node and the virtual node corresponding to the device identifier of the slave node.

[0334] In some embodiments, when the edge cluster is a hybrid edge cluster, the master node includes a container master node that supports container applications and a container slave node that supports container applications. Before obtaining the target application associated with the target node in the virtual cluster, a second identifier determination module is further included, which is also used for:

[0335] In the case of a hybrid edge cluster, the device identifier of the container master node and the device identifier of the container slave node are received from the edge cluster.

[0336] In response to the operation of binding virtual nodes to each node, determine the virtual node corresponding to the device identifier of the container master node and the virtual node corresponding to the device identifier of the container slave node.

[0337] In some embodiments, an update module is also included, for:

[0338] In response to an update request for the target application associated with the target node, update the target application associated with the target node.

[0339] In some embodiments, the first transmitting module is further configured to:

[0340] In the case of a hybrid edge cluster where duplicate status information is received for any edge device in the edge cluster, a shutdown instruction is sent to the master node in the edge cluster to disable the sending of status information in process mode or container mode.

[0341] Based on the same technical concept, this disclosure also provides a management device 1200 for an edge cluster, which, as follows: Figure 12 As shown, it includes:

[0342] The receiving module 1201 is used to receive the target application associated with the target node sent by the cloud; wherein, the cloud stores a virtual cluster including at least one virtual node, and the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster.

[0343] Installation module 1202 is used to install the target application onto the target edge device.

[0344] In some embodiments, the edge cluster includes a master node and at least one slave node. The master node communicates with the cloud, the master node communicates with each slave node via a local area network, and the slave nodes communicate with the cloud through the master node.

[0345] In some embodiments, when the edge cluster is a process-mode edge cluster, both the master node and the slave node run process applications.

[0346] In the case of a containerized edge cluster, both the master node and the slave node run containerized applications.

[0347] In the case of a hybrid edge cluster, at least one slave node in the edge cluster supports both process applications and container applications.

[0348] In some embodiments, before receiving the target application associated with the target node sent from the cloud, a second sending module is further included, for:

[0349] The master node sends an application retrieval request to the cloud, which includes the node identifier of the target edge device. The node identifier is used by the cloud to identify the target node.

[0350] In some embodiments, the installation module is used for:

[0351] When the target edge device is a slave node, the target application is stored in the storage application of the master node;

[0352] The master node instructs the slave node to retrieve the target application from the storage application.

[0353] In some embodiments, when the edge cluster is a process-mode edge cluster, the storage application is an object storage service application;

[0354] In the case of an edge cluster that is in container mode, the storage application is an image repository;

[0355] In the case of a hybrid edge cluster, the storage applications include object storage service applications and image repositories;

[0356] Among them, object storage service applications are used for storage process applications;

[0357] An image repository is used to store images of container applications.

[0358] In some embodiments, before receiving the target application associated with the target node sent from the cloud, a third building module is further included, for:

[0359] In the case of a process-mode edge cluster, the process-mode edge cluster is built based on the installation requests of edge devices sent by the master node.

[0360] In some embodiments, the third building module, which builds the master node in a process-mode edge cluster, is further configured to:

[0361] If the master node is to be installed, an installation request containing the master node's identifier is generated;

[0362] Send installation request;

[0363] Receive the master node installation package returned by the cloud in response to the installation request;

[0364] Build the master node based on the master node installation package;

[0365] The master node installation package includes: the core application required by the master node, and the object storage service application; the core application is used to manage slave nodes, and the object storage service application is used to store the process application of the slave nodes.

[0366] In some embodiments, the third building module, which is a slave node in a process-mode edge cluster, is further configured to:

[0367] If a slave node is to be installed, an installation request containing the slave node's identifier is generated;

[0368] Send installation request;

[0369] Receive the slave node installation package returned by the cloud in response to the installation request;

[0370] Build a slave node based on the slave node installation package;

[0371] The slave node installation package is used to control the establishment of local area network communication between the slave node and the master node, and to control the slave node to obtain the slave node's process application from the master node's object storage service application.

[0372] In some embodiments, before receiving the target application associated with the target node sent from the cloud, a fourth building module is further included, which is also used for:

[0373] In the case of a hybrid edge cluster, a process cluster for running process applications is constructed in the edge cluster based on the installation request of the edge device sent by the process control node in the master node.

[0374] In some embodiments, the fourth construction module, which constructs the process control node in the process cluster, is further configured to:

[0375] If the installation request is for installing a process control node, generate an installation request containing the identity of the process control node;

[0376] Send installation request;

[0377] Receive the installation package from the process control node returned by the cloud in response to the installation request;

[0378] Build a process control node based on the process control node installation package;

[0379] The process control node installation package includes: the core application required by the process control node, and the object storage service application; the core application is used to manage the process child nodes of the running process application, and the object storage service application is used to store the process application.

[0380] In some embodiments, the fourth construction module, which constructs the process child nodes of the running application in the process cluster, is further configured to:

[0381] When installing process child nodes, generate an installation request containing the identity of the process child nodes;

[0382] Send installation request;

[0383] Receive the installation package for the process child node returned by the cloud in response to the installation request;

[0384] Build process sub-nodes based on the process sub-node installation package;

[0385] The process sub-node installation package is used to control the process sub-node to establish local area network communication with the process control node, and to control the process sub-node to obtain the process application of the process sub-node from the object storage service of the process control node.

[0386] In some embodiments, before receiving the target application associated with the target node sent from the cloud, a third sending module is further included, for:

[0387] In the case of an edge cluster in container mode, send the device identifier of the master node and the device identifier of the slave node of the edge cluster.

[0388] In some embodiments, when the edge cluster is a hybrid edge cluster, the master node includes a container master node that supports container applications and container slave nodes that support container applications. Before receiving the target application associated with the target node sent from the cloud, a fourth sending module is also included, for:

[0389] Send the device identifier of the container master node that supports container mode and the device identifier of the container slave node that supports container mode.

[0390] In some embodiments, a status acquisition module is further included, for:

[0391] Obtain the status information of each edge device in the edge cluster based on the master node;

[0392] Send status information to the cloud.

[0393] The specific functions and examples of each module and submodule of the apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0394] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0395] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0396] Figure 13 A schematic block diagram of an example electronic device 1300 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0397] like Figure 13 As shown, device 1300 includes a computing unit 1301, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1302 or a computer program loaded from storage unit 1308 into random access memory (RAM) 1303. The RAM 1303 may also store various programs and data required for the operation of device 1300. The computing unit 1301, ROM 1302, and RAM 1303 are interconnected via bus 1304. Input / output (I / O) interface 1305 is also connected to bus 1304.

[0398] Multiple components in device 1300 are connected to I / O interface 1305, including: input unit 1306, such as keyboard, mouse, etc.; output unit 1307, such as various types of monitors, speakers, etc.; storage unit 1308, such as disk, optical disk, etc.; and communication unit 1309, such as network card, modem, wireless transceiver, etc. Communication unit 1309 allows device 1300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0399] The computing unit 1301 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1301 performs the various methods and processes described above, such as edge cluster management methods. For example, in some embodiments, the edge cluster management method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1308. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1300 via ROM 1302 and / or communication unit 1309. When the computer program is loaded into RAM 1303 and executed by the computing unit 1301, one or more steps of the edge cluster management method described above may be performed. Alternatively, in other embodiments, computing unit 1301 may be configured to perform edge cluster management methods by any other suitable means (e.g., by means of firmware).

[0400] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0401] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0402] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0403] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0404] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0405] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0406] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0407] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method of managing an edge cluster, wherein, The edge cluster includes a master node and at least one slave node, the master node communicates with the cloud, the master node communicates with each slave node through a local area network, and the slave node communicates with the cloud through the master node; The cloud stores a virtual cluster including at least one virtual node, and the method is applied to the cloud and includes: Obtaining a target application associated with a target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster; Sending the target application to the edge cluster to enable the edge cluster to install the target application to the target edge device; In the case that the edge cluster is a process mode edge cluster, the applications running on the master node and the slave node are process applications; before obtaining the target application associated with the target node in the virtual cluster, the method further includes constructing the process mode edge cluster based on the following method: The cloud generates a master node installation instruction and sends it to the master node to enable the master node to send a master node installation request to the cloud after establishing a communication connection with the cloud; Based on the master node installation request, a master node installation package is sent to the master node; the master node installation package includes: a core application required by the master node, and an object storage service application; the core application is used to manage the slave node, and the object storage service application is used to store the process application of the slave node; The cloud also generates a slave node installation instruction, which is copied by the master node and then sent to the slave node for execution, wherein the slave node installation instruction defines a slave node identifier and a master node identifier, to enable the slave node to establish a local area network communication with the master node based on the master node identifier, and send an installation request carrying the slave node identifier to the master node to obtain a slave node installation package of the slave node from the cloud based on the installation request.

2. The method of claim 1, wherein, in the case that the edge cluster is a container mode edge cluster, the applications running on the master node and the slave node are container applications; In the case that the edge cluster is a hybrid mode edge cluster, at least one slave node in the edge cluster supports process applications and container applications.

3. The method of claim 2, wherein, Obtaining the target application associated with the target node in the virtual cluster includes: Receiving an application acquisition request sent by the master node, the application acquisition request including a node identifier of the target edge device; The virtual node corresponding to the node identifier is taken as the target node, and the application associated with the target node is taken as the target application.

4. The method of claim 3, wherein, Sending the target application to the edge cluster includes: Sending the target application to the master node of the edge cluster to enable the master node to install the target application in the target edge device.

5. The method of claim 1, wherein, Based on the master node installation request, a master node installation package is sent to the master node, including: Obtaining a master node identifier of the master node from the installation request of the master node; Obtaining a master node installation package associated with the virtual node corresponding to the master node identifier; Sending the master node installation package to the master node.

6. The method of claim 1, further comprising constructing a slave node in the edge cluster of the process mode based on the following method: in response to an installation request of an edge device sent by the master node, obtaining a slave node identifier of the slave node from the installation request in a case that the installation request is for installing the slave node; obtaining a slave node installation package associated with a virtual node corresponding to the slave node identifier; sending the slave node installation package to the master node, so that the master node controls the slave node to install the slave node installation package; the slave node installation package is used to control the slave node and the master node to establish local area network communication, and control the slave node to obtain a process application of the slave node from an object storage service application of the master node.

7. The method of claim 2, before obtaining the target application associated with the target node in the virtual cluster, further comprising: in a case that the edge cluster is a hybrid mode edge cluster, constructing a process cluster in the edge cluster for running a process application in response to an installation request of an edge device sent by a process control node in the master node.

8. The method of claim 7, wherein, constructing a process control node in the process cluster in response to an installation request of an edge device sent by a process control node in the master node, comprising: in a case that the installation request is for installing the process control node, obtaining an identity of the process control node from the installation request; obtaining a process control node installation package associated with a virtual node corresponding to the identity of the process control node; sending the process control node installation package to the process control node; wherein the process control node installation package comprises a core application required by the process control node and an object storage service application; the core application is used to manage a process child node running a process application, and the object storage service application is used to store a process application.

9. The method of claim 8, wherein, constructing a process child node running a process application in the process cluster in response to an installation request of an edge device sent by a process control node in the master node, comprising: in a case that the installation request is for installing the process child node, obtaining an identity of the process child node from the installation request; obtaining a process child node installation package associated with a virtual node corresponding to the identity of the process child node; sending the process child node installation package to the process control node, so that the process control node controls the process child node to install the process child node installation package; the process child node installation package is used to control the process child node and the process control node to establish local area network communication, and control the process child node to obtain a process application of the process child node from an object storage service of the process control node.

10. The method of claim 2, before obtaining the target application associated with the target node in the virtual cluster, further comprising: in a case that the edge cluster is a container mode edge cluster, receiving a device identifier of the master node and a device identifier of the slave node sent by the edge cluster; In response to the operation of binding the virtual nodes to each node, the virtual node corresponding to the device identifier of the master node and the virtual node corresponding to the device identifier of the slave node are determined.

11. The method of claim 2, in the case where the edge cluster is a hybrid mode edge cluster, the master node includes a container master node supporting a container application and a container slave node supporting a container application, and before the target application associated with the target node in the virtual cluster is acquired, the method further comprises: In the case where the edge cluster is a hybrid mode edge cluster, the device identifier of the container master node and the device identifier of the container slave node sent by the edge cluster are received; In response to the operation of binding the virtual nodes to each node, the virtual node corresponding to the device identifier of the container master node and the virtual node corresponding to the device identifier of the container slave node are determined.

12. The method of any one of claims 1-11, further comprising: In response to an update request for the target application associated with the target node, the target application associated with the target node is updated.

13. The method of claim 2, further comprising: In the case where the edge cluster is a hybrid mode edge cluster and duplicate state information is received for any edge device in the edge cluster, a shutdown instruction is sent to the master node in the edge cluster to disable the process mode to send the state information or disable the container mode to send the state information.

14. A management method of an edge cluster, applied to an edge cluster, the edge cluster including a master node and at least one slave node, the master node being in communication with a cloud, the master node being in communication with each slave node through a local area network, and the slave node being in communication with the cloud through the master node; the method comprising: Receiving a target application associated with a target node sent by the cloud; wherein the cloud stores a virtual cluster including at least one virtual node, and the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster; Installing the target application to the target edge device; In the case where the edge cluster is a process mode edge cluster, the applications running on the master node and the slave node are all process applications; before receiving the target application associated with the target node sent by the cloud, the edge cluster is constructed based on the following method: The cloud generates a master node installation instruction and sends it to the master node, so that after the master node and the cloud establish a communication connection, the master node sends an installation request of the master node to the cloud; Based on the installation request of the master node, a master node installation package is sent to the master node; the master node installation package includes: a core application required by the master node, and an object storage service application; the core application is used to manage the slave node, and the object storage service application is used to store the process application of the slave node; The cloud end also generates a slave node installation instruction, which is sent to the slave node for execution after being copied by the master node, wherein the slave node installation instruction defines a slave node identifier and a master node identifier, so that the slave node establishes local area network communication with the master node based on the master node identifier, and sends an installation request carrying the slave node identifier to the master node, so that the master node obtains a slave node installation package for the slave node from the cloud end based on the installation request.

15. The method of claim 14, in the case of the edge cluster being a container mode edge cluster, the application running on the master node and the slave node is a container application; in the case of the edge cluster being a hybrid mode edge cluster, at least one slave node in the edge cluster supports process applications and container applications.

16. The method of claim 15, wherein, Before receiving the target application associated with the target node sent by the cloud end, the method further comprises: sending an application acquisition request to the cloud end based on the master node, wherein the application acquisition request comprises a node identifier of the target edge device; and the node identifier is used for determining the target node by the cloud end.

17. The method of claim 16, wherein installing the target application to the target edge device comprises: in the case of the target edge device being a slave node, storing the target application in a storage application of the master node; based on the master node instructing the slave node to acquire the target application from the storage application.

18. The method of claim 17, wherein, in the case of the edge cluster being a process mode edge cluster, the storage application is an object storage service application; in the case of the edge cluster being a container mode edge cluster, the storage application is an image repository; in the case of the edge cluster being a hybrid mode edge cluster, the storage application comprises the object storage service application and the image repository; wherein the object storage service application is used to store process applications; the image repository is used to store images of container applications.

19. The method of claim 14, wherein, constructing a master node in the edge cluster in a process mode comprises: in the case of installing the master node, generating an installation request comprising a master node identifier of the master node; sending the installation request; receiving a master node installation package returned by the cloud end for the installation request; constructing the master node based on the master node installation package; wherein the master node installation package comprises: a core application required by the master node, and an object storage service application; the core application is used to manage the slave node, and the object storage service application is used to store process applications of the slave node.

20. The method of claim 14, wherein, constructing a slave node in the edge cluster in a process mode comprises: in the case of installing the slave node, generating an installation request comprising a slave node identifier of the slave node; sending the installation request; receiving a slave node installation package returned by the cloud end for the installation request; constructing the slave node based on the slave node installation package; the slave node installation package is used to control the slave node and the master node to establish local area network communication, and control the slave node to acquire process applications of the slave node from the object storage service application of the master node.

21. The method of claim 15, before receiving the target application associated with the target node sent by the cloud, further comprising: in a case where the edge cluster is a hybrid mode edge cluster, constructing a process cluster for running a process application in the edge cluster based on an installation request of an edge device sent by a process control node in the master node.

22. The method of claim 21, wherein, constructing the process control node in the process cluster, comprising: in a case where the installation request is for installing the process control node, generating an installation request containing an identity of the process control node; sending the installation request; receiving a process control node installation package returned by the cloud for the installation request; constructing the process control node based on the process control node installation package; wherein the process control node installation package comprises a core application required by the process control node and an object storage service application, the core application being used for managing a process child node running a process application, and the object storage service application being used for storing process applications.

23. The method of claim 22, wherein, constructing the process child node running a process application in the process cluster, comprising: in a case where the process child node is installed, generating an installation request containing an identity of the process child node; sending the installation request; receiving a process child node installation package returned by the cloud for the installation request; constructing the process child node based on the process child node installation package; the process child node installation package being used for controlling the process child node and the process control node to establish local area network communication, and controlling the process child node to obtain a process application of the process child node from an object storage service of the process control node.

24. The method of claim 15, before receiving the target application associated with the target node sent by the cloud, further comprising: in a case where the edge cluster is a container mode edge cluster, sending a device identity of the master node of the edge cluster and a device identity of the slave node.

25. The method of claim 15, in a case where the edge cluster is a hybrid mode edge cluster, the master node comprising a container master node supporting a container application and a container slave node supporting a container application, before receiving the target application associated with the target node sent by the cloud, further comprising: sending a device identity of the container master node supporting a container mode and a device identity of the container slave node supporting a container mode.

26. The method of any one of claims 14-25, further comprising: obtaining state information of each edge device in the edge cluster based on the master node; sending the state information to the cloud.

27. A management apparatus of an edge cluster, wherein, the edge cluster comprising a master node and at least one slave node, the master node being in communication with the cloud, the master node being in communication with each slave node through a local area network, and the slave node being in communication with the cloud through the master node; the cloud storing a virtual cluster comprising at least one virtual node, the apparatus being applied to the cloud, comprising: The acquisition module is configured to acquire a target application associated with a target node in the virtual cluster; the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster; The first sending module is configured to send the target application to the edge cluster to install the target application to the target edge device by the edge cluster; In a case where the edge cluster is a process mode edge cluster, the applications running on the master node and the slave nodes are process applications; the method further comprises a first construction module configured to, before acquiring the target application associated with the target node in the virtual cluster, construct the process mode edge cluster based on the following method: The cloud generates a master node installation instruction and sends the master node installation instruction to the master node, so that the master node and the cloud establish a communication connection, and the master node sends an installation request of the master node to the cloud after the communication connection is established; Based on the installation request of the master node, a master node installation package is sent to the master node; the master node installation package includes: a core application required by the master node, and an object storage service application; the core application is used to manage the slave nodes, and the object storage service application is used to store process applications of the slave nodes; The cloud also generates a slave node installation instruction, the slave node installation instruction is copied by the master node and then sent to the slave node for execution, wherein the slave node installation instruction defines a slave node identifier and a master node identifier, so that the slave node establishes a local area network communication with the master node based on the master node identifier, and sends an installation request carrying the slave node identifier to the master node, so that the master node acquires a slave node installation package of the slave node from the cloud based on the installation request.

28. An edge cluster management apparatus applied to an edge cluster, the edge cluster comprising a master node and at least one slave node, the master node being in communication with a cloud, the master node being in communication with each slave node through a local area network, and the slave nodes being in communication with the cloud through the master node; the apparatus comprising: The receiving module is configured to receive a target application associated with a target node sent by the cloud; wherein the cloud stores a virtual cluster comprising at least one virtual node, the target node has a corresponding target edge device in the edge cluster; the target node is any virtual node in the virtual cluster; The installation module is configured to install the target application to the target edge device; In a case where the edge cluster is a process mode edge cluster, the applications running on the master node and the slave nodes are process applications; before receiving the target application associated with the target node sent by the cloud, the edge cluster is constructed based on the following method: The cloud generates a master node installation instruction and sends the master node installation instruction to the master node, so that the master node and the cloud establish a communication connection, and the master node sends an installation request of the master node to the cloud after the communication connection is established; Based on the installation request of the master node, a master node installation package is sent to the master node; the master node installation package includes: a core application required by the master node, and an object storage service application; the core application is used to manage the slave nodes, and the object storage service application is used to store process applications of the slave nodes; The cloud further generates a slave node installation instruction, which is sent to the slave node for execution after being copied by the master node, wherein the slave node installation instruction defines a slave node identifier and a master node identifier, so that the slave node establishes a local area network communication with the master node based on the master node identifier, and sends an installation request carrying the slave node identifier to the master node, so that the master node obtains a slave node installation package of the slave node from the cloud based on the installation request.

29. A management system of an edge cluster, comprising: a cloud, configured to execute the method of any one of claims 1-13; an edge cluster, configured to execute the method of any one of claims 14-26.

30. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method of any one of claims 1-26.

31. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to execute the method of any one of claims 1-26.

32. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-26.

Images