CN117931379A - Cluster deployment method, device, system, equipment and medium for edge computing - Google Patents

Abstract

The disclosure relates to the technical field of edge computing, and discloses a cluster deployment method, device, system, equipment and medium for edge computing. The cluster deployment method for edge computing comprises the following steps: in response to the container group creation request, creating a target container group in the tenant cluster; screening target clusters for bearing the target container group from a plurality of preset edge clusters, wherein the plurality of preset edge clusters are respectively connected with tenant clusters in a communication mode; selecting a target node corresponding to a target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters; and configuring the node field of the target container group according to the node name of the target node so as to schedule the target container group to the target node. Compared with the related art, the method and the device have the advantages that the resource utilization rate is obviously improved, and the user experience is improved.

Description

Cluster deployment method, device, system, equipment and medium for edge computing

Technical Field

The disclosure relates to the technical field of edge computing, in particular to a cluster deployment method, a device, a system, equipment and a medium for edge computing.

Background

Kubernetes (K8 s for short) is an industrial container arrangement platform, which is mainly used for providing service deployment, elastic capacity expansion, management and the like, and has various functions of deploying application, realizing expansion mechanism, maintaining and the like. The traditional Kubernetes architecture generally includes a Master Node (Master) that is primarily responsible for the management and scheduling of the entire cluster, and a worker Node (Node) that is used to run containers and interact with the Master Node.

In the edge computing scenario, the cloud computing vendor may provide a Kubernetes hosting scheme for the user who needs to purchase the virtual machine or Node separately, so in the traditional scheme, the Node is held by the user. When a user uses Kubernetes to manage a container, the application scale must be reasonably planned, for example, the sum of resources to be occupied by all applications to be deployed is planned, and the specification of a virtual machine or Node to be sold by a manufacturer is generally fixed, so that the situation that the purchased Node resources are obviously larger than the required sum of resources often occurs for the user, the resource waste situation occurs, and the problem of low resource utilization is caused.

Disclosure of Invention

In view of this, the present disclosure provides a cluster deployment method, device, system, device and medium for edge computing, so as to solve the problem of low Node resource utilization rate that is easily occurred in the related art in a manner that a user holds a Node.

In a first aspect, the present disclosure provides a cluster deployment method for edge computing, the method comprising:

In response to the container group creation request, creating a target container group in the tenant cluster;

screening target clusters for bearing the target container group from a plurality of preset edge clusters, wherein the plurality of preset edge clusters are respectively connected with tenant clusters in a communication mode;

selecting a target node corresponding to a target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters;

and configuring the node field of the target container group according to the node name of the target node so as to schedule the target container group to the target node.

According to the cluster deployment method for edge computing, the target container group is created in the tenant cluster, and the target container group is borne by the edge cluster in communication connection with the tenant cluster and is scheduled to the virtual Node in the tenant cluster, so that in the scheme of the invention, a user only needs to hold the target container group in the virtual Node and does not need to hold a real Node in the edge cluster, so that the user can purchase resources with the container group as granularity, and can purchase resources closer to the actual capacity demand according to the planning of the application scale, thereby better overcoming the problems of low Node resource utilization rate and the like which are easy to occur in the mode that the user holds the Node in the related technology, and obviously reducing the cost input of the user. Therefore, the method and the device obviously improve the Node resource utilization rate, truly realize that the user purchases the Node resource according to the needs, and improve the user experience and satisfaction.

In a second aspect, the present disclosure provides a cluster deployment apparatus for edge computing, the apparatus comprising:

A creation module for creating a target container group in the tenant cluster in response to a container group creation request;

the screening module is used for screening target clusters for bearing the target container group from a plurality of preset edge clusters, and the plurality of preset edge clusters are respectively connected with the tenant clusters in a communication mode;

the selection module is used for selecting a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters;

And the updating module is used for configuring the node field of the target container group according to the node name of the target node so as to schedule the target container group to the target node.

In a third aspect, the present disclosure provides a cluster deployment system for edge computing, the system comprising a tenant cluster and a target cluster, the tenant cluster comprising an application programming interface server and a scheduler;

An application programming interface server for creating a target container group in the tenant cluster in response to the container group creation request;

The dispatcher is used for screening out target clusters for bearing the target container group from a plurality of preset edge clusters, selecting target nodes corresponding to the target clusters from a plurality of virtual nodes in the tenant clusters, and configuring node fields of the target container group according to node names of the target nodes so as to dispatch the target container group to the target nodes;

the plurality of preset edge clusters are respectively in communication connection with the tenant clusters, and the plurality of virtual nodes are in one-to-one correspondence with the plurality of preset edge clusters.

In a fourth aspect, the present disclosure provides a computer device comprising: the processor is in communication connection with the memory, and the memory stores computer instructions, so that the processor executes the computer instructions to perform the cluster deployment method for edge computing according to the first aspect or any implementation manner corresponding to the first aspect.

In a fifth aspect, the present disclosure provides a computer readable storage medium having stored thereon computer instructions for causing a computer to perform the cluster deployment method for edge computing of the first aspect or any of its corresponding embodiments.

Drawings

In order to more clearly illustrate the embodiments of the present disclosure or the prior art, the drawings that are required in the detailed description or the prior art will be briefly described, it will be apparent that the drawings in the following description are some embodiments of the present disclosure, and other drawings may be obtained according to the drawings without inventive effort for a person of ordinary skill in the art.

FIG. 1 is a flow diagram of a cluster deployment method for edge computing according to an embodiment of the present disclosure;

FIG. 2 is a flow diagram of another cluster deployment method for edge computing, according to an embodiment of the disclosure;

FIG. 3 is a flow diagram of yet another cluster deployment method for edge computing, according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of a cluster deployment system for edge computing, in accordance with an embodiment of the present disclosure;

FIG. 5 is a flow diagram of a container group creation process according to an embodiment of the present disclosure;

FIG. 6 is a block diagram of a cluster deployment apparatus for edge computing, according to an embodiment of the disclosure;

fig. 7 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present disclosure.

Detailed Description

For the purposes of making the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some embodiments of the present disclosure, but not all embodiments. Based on the embodiments in this disclosure, all other embodiments that a person skilled in the art would obtain without making any inventive effort are within the scope of protection of this disclosure.

The Kubernetes is taken as a container arrangement system of an open source, the functions of the Kubernetes comprise service discovery and load balancing, service automatic boxing, container storage arrangement, container fault recovery, automatic release and rollback, configuration and key storage, service level expansion and contraction, batch execution, daemon tasks and the like, the container arrangement system has become a container arrangement standard in the field of cloud primordia, a user can reduce operation and maintenance cost and improve operation and maintenance efficiency by using the Kubernetes, and locking of cloud manufacturers can be effectively avoided by using a standardized API (Application Programming Interface ) provided by the Kubernetes. For users, the Kubernetes can provide standard Kubernetes interfaces, and can be deployed by mirroring and containerization, and the deployment efficiency of PAAS (Platform AS A SERVICE ) application is improved by combining CI (Continuous Integration )/CD (Continuous Delivery) processes, and the Node resource utilization rate can be improved by reasonable scheduling, so that the use cost is reduced.

In an edge computing scenario, cloud computing vendors may provide classical Kubernetes hosting schemes, typically purchasing virtual machines for users, cloud computing vendors provide management of Kubernetes lifecycles, e.g., kubernetes installation, deployment, upgrades, etc., with nodes held by users. However, the existing solutions have the following problems: when a user uses Kubernetes to manage a container, the user must reasonably plan the application scale capacity of the user, for example, plan the resources (such as computing resources or storage resources) occupied by each application, and the purchased Node resources need to be greater than the sum of the resources required by all applications; in addition, due to the problem of resource fragmentation, node resources are definitely larger than resources required by all applications, so that certain resource waste exists in the related technology. It can be seen that in the traditional Kubernetes hosting scheme, a user needs to have accurate resource planning on own service scale, and purchase virtual machines or Node equipment nanotubes into Kubernetes as required; however, due to the problems of resource fragmentation and the like, node resources cannot tightly cover the sum of resources to be occupied by all applications to be deployed, so that the problems of resource waste and low resource utilization rate occur; moreover, the service has the situations of wave crests and wave troughs, the occupation of resources is relatively large when the service wave crests exist, and the occupation of resources is relatively small when the service wave troughs exist, so that users are required to purchase enough virtual machines in advance to cope with the service wave crests (namely, traffic flood peaks), and parts of the virtual machines are required to be cleared and retired when the service wave troughs exist, the expansion or the contraction of the resources under Node granularity is often in a minute level, and the time occupation of users in a minute level nano-tube process is relatively long; in addition, the Node in the related technology is self-owned by the user, and the user is required to maintain the life cycle of the Node, such as Node component upgrade, node monitoring, alarm management and the like, so that the user experience is poor.

In accordance with the disclosed embodiments, a cluster deployment method embodiment for edge computing is provided, it being noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system, such as a set of computer executable instructions, and, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order other than that illustrated herein.

The embodiment provides a cluster deployment method for edge computing, which can be used for tenant clusters, and particularly can be used for schedulers (edge-cluster-schedulers) in tenant k8s clusters.

FIG. 1 is a flow chart of a cluster deployment method for edge computing, as shown in FIG. 1, according to an embodiment of the disclosure, the flow comprising the steps of:

Step S101, in response to the container group creation request, creates a target container group in the tenant cluster.

Among them, a "POD group" is a POD, in which a plurality of PODs can be deployed. The tenant cluster in this embodiment is a tenant k8s cluster, and the Kubernetes cluster accessed by the tenant does not manage any machine, and replaces a real k8s Node by a virtual Node; wherein the target container group is a virtual POD.

The tenant clusters of the disclosed embodiments may be deployed on a computer device, which may be, for example, a user device.

In some alternative embodiments, prior to responding to the container group creation request, the method further comprises: creating a plurality of virtual nodes corresponding to the preset edge clusters one by one in the tenant clusters, wherein the node names of the virtual nodes are the same as the cluster names of the preset edge clusters one by one.

The plurality of preset edge clusters in the embodiment of the present disclosure are all edge clusters in the resource pool, the edge clusters belong to k8s clusters of the edge data plane, and the k8s clusters are used as a management mode for managing edge computing resources and are standard k8s clusters; the granularity of one Kubernetes cluster in one machine room is adopted in the embodiment, for example, the Kubernetes cluster in Shanghai telecom is used for managing machines in Shanghai telecom machine rooms, the Kubernetes cluster in Hangzhou telecom is used for managing Kubernetes clusters in Hangzhou telecom machine rooms, and other machine rooms and the like. In this embodiment, the Virtual nodes are Virtual nodes, and different Virtual nodes correspond to different areas, for example, an area corresponding to one Virtual Node is Shanghai, and an area corresponding to another Virtual Node is Hangzhou.

As shown in fig. 4, the preset edge cluster is accessed to the tenant cluster in the form of Virtual Node, the POD in the tenant cluster is synchronized to the designated preset edge cluster through the corresponding Virtual Node, and no real Virtual machine resource is set in the tenant cluster, so that the purpose of starting the POD as required is achieved. For example, in the tenant cluster in fig. 4, a Virtual Node (Virtual Node 20) and a corresponding target container group (POD 10) and a corresponding target container group (POD 11), a Virtual Node (Virtual Node 21) and a corresponding target container group (POD 12), a Virtual Node (Virtual Node 22) and a corresponding target container group (POD 13), a preset edge cluster (Kubernetes 61, shanghai telecom), a preset edge cluster (Kubernetes 62, hangzhou telecom), a preset edge cluster (Kubernetes 63, and others) may be set, a Node proxy (kubelet) is set in the edge cluster Kubernetes61, and a real container group (POD 40) corresponding to the target container group (POD 10), a real container group (POD 41) corresponding to the target container group (POD 11), and a cluster management Node (Kubernetes master) is set; a node agent (kubelet) and a real container group (POD 42) corresponding to the target container group (POD 12) are arranged in the edge cluster Kubernetes62, and a cluster management node (Kubernetes master) is arranged; in the edge cluster Kubernetes63, a node proxy (kubelet) is provided, and a real container group (POD 43) corresponding to the target container group (POD 13) and a cluster management node (Kubernetes master) is provided.

According to the embodiment, the mode that the node names of the virtual nodes are in one-to-one correspondence with the cluster names of the preset edge clusters can be used for ensuring that the scheme can accurately access the corresponding preset edge clusters through the virtual nodes, and providing resource support for the user to create the POD.

Step S102, a target cluster used for bearing the target container group is screened out from a plurality of preset edge clusters, and the plurality of preset edge clusters are respectively connected with the tenant clusters in a communication mode.

Specifically, the plurality of preset edge clusters in the embodiment may be all edge clusters in the resource pool, and the embodiment can select an edge cluster meeting the requirement according to a preset configuration rule in all edge clusters in the resource pool.

In some optional embodiments, a preset label is set in the container group creation request, where the preset label is used to characterize the configuration requirement of the target container group.

The screening the target cluster for carrying the target container group from the plurality of preset edge clusters includes: and screening target clusters from the plurality of preset edge clusters based on the preset labels and the resource allowance of the plurality of preset edge clusters.

Step S103, selecting a target node corresponding to a target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters.

Under the condition of determining the target cluster, the one-to-one correspondence between the preset edge cluster and the virtual node is configured in advance, so that the virtual node corresponding to the target cluster can be identified according to the currently determined target cluster.

In some alternative embodiments, the cluster name of the target cluster is the same as the node name of the target node; selecting a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, including:

step a1, obtaining a cluster name of a target cluster.

In this embodiment, each preset edge cluster has a name, and the node name of each target node is the same as the name of the preset edge cluster in a one-to-one correspondence.

Step a2, selecting a virtual node with the same node name as the cluster name from a plurality of virtual nodes in the tenant cluster as a target node.

For a plurality of virtual nodes automatically registered in the tenant cluster, the embodiment can find the target node from the plurality of virtual nodes according to the acquired cluster name of the target cluster, thereby providing a reliable basis for on-demand scheduling of the container group.

Step S104, the node fields of the target container group are configured according to the node names of the target nodes so as to schedule the target container group to the target nodes.

The node field of the target container group represents the name of the node where the target container group operates, and in this embodiment, the target container group is scheduled to the target node by configuring the node field of the target container group.

In some alternative embodiments, step S104 includes: the value of the Node field of the target container group is set to the Node Name (Node Name) of the target Node.

The Node Name of the target Node can be set as the Node Name of the corresponding virtual Node in the target POD.

Based on the setting of the value of the node field of the target container group, the present embodiment can realize the function of accurate scheduling of the target container group.

According to the cluster deployment method for edge computing, the target container group is created in the tenant cluster, the target container group is borne by the edge cluster in communication connection with the tenant cluster, and is scheduled to the virtual Node in the tenant cluster, so that in the scheme of the embodiment, a user only needs to hold the target container group in the virtual Node and does not need to hold the real Node in the edge cluster, so that the user can purchase resources with the container group as granularity, the user only needs to pay for each POD purchased, and resources which are closer to the actual capacity demand can be purchased according to the planning of the application scale, and the problem that Node resource utilization rate is low and the like easily occurs in a mode of holding the Node by the user in the related technology is well overcome, and cost investment of the user is obviously reduced. In addition, in this embodiment, when the user performs resource expansion or resource contraction, the Node minute level nanotube process in the related technology can be replaced by the POD second level elastic expansion process for the ready (ready) edge cluster, so that user experience and satisfaction are greatly improved. In addition, in this embodiment, the user does not need to manage the Node, so that the difficulty of using the POD by the user is reduced, and the start speed of the POD is also faster. Therefore, the embodiment obviously improves the Node resource utilization rate, realizes that the user purchases the Node resource according to the requirement by using the POD granularity, and further improves the user experience and satisfaction.

In this embodiment, a cluster deployment method for edge computing is provided, which may be used for tenant clusters, and fig. 2 is a flowchart of the cluster deployment method for edge computing according to an embodiment of the disclosure, as shown in fig. 2, where the flowchart includes the following steps:

Step S201, in response to the container group creation request, creates a target container group in the tenant cluster. Please refer to step S101 in the embodiment shown in fig. 1 in detail, which is not described herein.

Step S202, a target cluster for bearing the target container group is screened out from a plurality of preset edge clusters, and the plurality of preset edge clusters are respectively connected with the tenant clusters in a communication mode.

Specifically, the preset label includes at least one of an area label, a machine room label, an operator label, and a city label, and in combination with the foregoing embodiment, based on a resource allowance of the preset label and a plurality of preset edge clusters, the screening of the target cluster from the plurality of preset edge clusters includes:

In step S2021, a preset edge cluster conforming to the preset label is selected from the plurality of preset edge clusters.

The zone (zone) tag represents a region to which the edge cluster belongs, the machine room (idc) tag represents a machine room to which the edge cluster belongs, the operator (isp) tag represents an operator to which the edge cluster belongs, and the city (city) tag represents a city to which the edge machine room belongs.

Specifically, the preset edge clusters may be sequentially screened according to dimensions of a zone (zone), a machine room (idc), an operator (isp), a city (city), etc., for example, the edge clusters of the east China telecom are selected according to the zone, and then the Hangzhou cluster is selected from the edge clusters of the east China telecom according to the city.

More specifically, the plurality of preset tags may be configured in a specified configuration file (e.g., YAML file) in which the plurality of preset tags are represented by corresponding fields.

In step S2022, the resource margin of the preset edge cluster corresponding to the preset label is obtained, and one preset edge cluster whose resource margin meets the operation requirement of the target container group is determined as the target cluster.

On the premise of meeting the operation requirement of the target container group, the embodiment can determine the preset edge cluster with larger resource allowance as the target cluster.

In some optional embodiments, the deployment method of the preset edge cluster includes: and deploying a preset edge cluster in an edge machine room.

Based on the improved scheme, the embodiment can fully meet the selection requirement of customers on the edge clusters, so that the PODs to be created can be borne through the edge clusters meeting the requirements. By means of arranging an edge cluster in an edge machine room, stability and reliability of the edge cluster can be fully guaranteed.

Step S203, selecting a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, where the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters. Please refer to step S103 in the embodiment shown in fig. 1 in detail, which is not described herein.

Step S204, the node fields of the target container group are configured according to the node names of the target nodes so as to schedule the target container group to the target nodes. Please refer to step S104 in the embodiment shown in fig. 1 in detail, which is not described herein.

In this embodiment, a cluster deployment method for edge computing is provided, which may be used for tenant clusters, and fig. 3 is a flowchart of the cluster deployment method for edge computing according to an embodiment of the disclosure, as shown in fig. 3, where the flowchart includes the following steps:

Step S301, in response to a container group creation request, creates a target container group in the tenant cluster. Please refer to step S201 in the embodiment shown in fig. 2 in detail, which is not described herein.

Step S302, a target cluster for carrying the target container group is selected from a plurality of preset edge clusters, and the plurality of preset edge clusters are respectively connected with the tenant clusters in a communication manner. Please refer to step S202 in the embodiment shown in fig. 2, which is not described herein.

Step S303, selecting a target node corresponding to a target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters. Please refer to step S203 in the embodiment shown in fig. 2 in detail, which is not described herein.

Step S304, the node fields of the target container group are configured according to the node names of the target nodes, so that the target container group is scheduled to the target nodes. Please refer to step S204 in the embodiment shown in fig. 2 in detail, which is not described herein.

Step S305, notifying the target cluster to create a real container group corresponding to the target container group.

The target cluster of the embodiment of the present disclosure may be created together when creating the real container group if there is other information such as configmap (configuration map information), secret (key information), etc. Container configuration information, resource configuration information, network configuration information, life cycle management information and the like can be specifically deployed when the real container group is created; it should be understood that, the specific creation process for the real container group may be selected from the related container group creation schemes, which is not described in detail in this embodiment.

Step S306, receiving the deployment information of the real container group sent by the target cluster, and synchronizing the deployment information to the target container group.

After the target cluster successfully creates the real container group, the deployment information of the real container group is sent to the tenant cluster, and the tenant cluster user synchronizes the related information to the target container group.

Based on the scheme, the embodiment can better construct the mapping of the target container group to the real container group, and improves the reliability of cluster deployment.

With reference to fig. 5, virtual nodes corresponding to the preset edge clusters one by one are created through virtual node agents (virtual-kubelet) in the Kubernetes scene, so that the embodiment of the disclosure can provide a technical scheme of Kubernetes deployment based on the edge SERVERLESS (without a server) of virtual-kubelet. It can be seen that, in this embodiment, virtual Node is implemented based on Virtual-kubelet, and runs in the tenant k8s cluster in the form of POD, and if the tenant creates a POD in the tenant k8s cluster, the Virtual Node is used to synchronize the POD into the corresponding edge cluster. Specifically, the tenant cluster (Kubernetes) in this embodiment includes an application programming interface server (kube-apiserver), a scheduler (edge-cluster-schedule), and a virtual node proxy (virtual-kubelet), where the user sends a request for creating a virtual POD to kube-apiserver, after receiving the request, kube-apiserver may feed back a message that the virtual POD of the user is successfully created (when the virtual POD created is not actually available), and if the edge-cluster-schedule observes a Pending (Pending) POD, the edge-cluster-schedule views the resource remaining of the preset edge cluster from a large number of preset edge clusters in the resource pool, selects an appropriate edge cluster for the POD, and returns a message of the appropriate edge cluster to the edge-cluster-schedule; the edge-cluster-schedule performs POD scheduling, including: setting a node name of a target node for the POD, for example, scheduling only the POD configured with a spec.schedulerName field with a value of edge-server, and selecting a proper edge cluster for the POD according to the resource condition of the edge cluster and the corresponding affinity of the POD; virtual-kubelet observes POD changes, including: and if the value of the node field of the POD is found to be the node name of the target node, notifying the proper edge cluster to create the real POD, and receiving a message that the real POD fed back by the edge cluster is successfully created, wherein the POD can be normally used at the moment to provide relevant services for users.

In the Kubernetes scene, virtual Node is a Node which is automatically registered and created with kube-apiserver by the process of Virtual-kubelet; specifically, the Virtual Node related information to be registered is determined through a preset configuration file (e.g., YAML file) preset by a user. Node Name is information related to the edge k8s cluster, for example, in fig. 4, virtual Node Name corresponding to Shanghai telecom cluster (shanghaict) is shanghaict, and Virtual Node Name corresponding to Hangzhou telecom cluster (hangzhouct) is hangzhouct; and so on, creating Virtual Node names according to the edge cluster names.

In this embodiment, since the real container group and the real nodes are disposed in the edge cluster, the maintenance of the real container group and the real nodes is provided by the cloud computing manufacturer, so that the operation and maintenance free function of the nodes is realized for the user, and the purposes of Node Less (Node without traditional meaning) and the like are realized for the user, thereby effectively reducing the attention of the user to the IAAS (Infrastructure AS A SERVICE, services as Infrastructure) layer, reducing the occupation of the user time, and improving the use experience and satisfaction degree of the user to the cloud service related products.

The embodiment also provides a cluster deployment device for edge computing, which is used for implementing the foregoing embodiments and preferred embodiments, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

The present embodiment provides a cluster deployment apparatus for edge computing, as shown in fig. 6, including:

a creation module 601, configured to create a target container group in the tenant cluster in response to a container group creation request.

The screening module 602 is configured to screen a target cluster for carrying the target container group from a plurality of preset edge clusters, where the plurality of preset edge clusters are respectively connected to the tenant cluster in a communication manner.

The selecting module 603 is configured to select a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, where the plurality of virtual nodes are in one-to-one correspondence with a plurality of preset edge clusters.

The updating module 604 is configured to configure the node fields of the target container group according to the node name of the target node, so as to schedule the target container group onto the target node.

In some optional embodiments, a preset label is set in the container group creation request, where the preset label is used to characterize the configuration requirement of the target container group. The screening module 602 is specifically configured to screen the target cluster from the plurality of preset edge clusters based on the preset label and the resource margins of the plurality of preset edge clusters.

In some optional embodiments, the preset tag includes at least one of a regional tag, a machine room tag, an operator tag, and a city tag; the screening module 602 includes:

And the screening unit is used for screening out preset edge clusters which accord with the preset labels from the plurality of preset edge clusters.

The determining unit is used for obtaining the resource allowance of the preset edge cluster which accords with the preset label and determining one preset edge cluster, of which the resource allowance meets the operation requirement of the target container group, as the target cluster.

In some optional embodiments, the deployment method of the preset edge cluster includes: and deploying a preset edge cluster in an edge machine room.

In some alternative embodiments, the cluster name of the target cluster is the same as the node name of the target node; the selection module 603 includes:

and the acquisition unit is used for acquiring the cluster name of the target cluster.

And the selection unit is used for selecting a virtual node with the same node name as the cluster name from the plurality of virtual nodes in the tenant cluster as a target node.

In some optional embodiments, the cluster deployment apparatus for edge computing further includes a virtual node generation module; the virtual node generation module is used for creating a plurality of virtual nodes corresponding to a plurality of preset edge clusters in the tenant clusters, and node names of the plurality of virtual nodes are the same as cluster names of the plurality of preset edge clusters in a one-to-one correspondence.

In some alternative embodiments, the updating module 604 is specifically configured to set the value of the node field of the target container group to the node name of the target node.

In some optional embodiments, the cluster deployment apparatus for edge computing further includes a notification module and a synchronization module.

And the notification module is used for notifying the target cluster to create a real container group corresponding to the target container group.

The synchronization module is used for receiving the deployment information of the real container group sent by the target cluster and synchronizing the deployment information to the target container group.

As shown in fig. 4 and 5, embodiments of the present disclosure may also provide a cluster deployment system for edge computing, the system including a tenant cluster and a target cluster, the tenant cluster including an application programming interface server (kube-apiserver) and a scheduler (edge-cluster-schedule) in a Kubernetes scenario.

An application programming interface server for creating a target container group in the tenant cluster in response to the container group creation request.

The system comprises a dispatcher, a target container group, a target node, a node field and a node field, wherein the dispatcher is used for screening a target cluster for bearing the target container group from a plurality of preset edge clusters, selecting a target node corresponding to the target cluster from a plurality of virtual nodes in a tenant cluster, and configuring the node field of the target container group according to the node name of the target node so as to dispatch the target container group to the target node.

The plurality of preset edge clusters are respectively in communication connection with the tenant clusters, and the plurality of virtual nodes are in one-to-one correspondence with the plurality of preset edge clusters.

In some optional embodiments, a preset label is set in the container group creation request, where the preset label is used to characterize the configuration requirement of the target container group. The scheduler is used for screening target clusters from the plurality of preset edge clusters based on the preset labels and the resource allowance of the plurality of preset edge clusters.

In some optional embodiments, the preset tag includes at least one of a regional tag, a machine room tag, an operator tag, and a city tag. The scheduler is used for screening out preset edge clusters which meet preset labels from a plurality of preset edge clusters, acquiring resource allowance of the preset edge clusters which meet the preset labels, and determining one preset edge cluster, of which the resource allowance meets the operation requirement of the target container group, as a target cluster.

In some optional embodiments, the deployment method of the preset edge cluster includes: and deploying a preset edge cluster in an edge machine room.

In some alternative embodiments, the cluster name of the target cluster is the same as the node name of the target node; the scheduler is used for acquiring the cluster name of the target cluster and selecting a virtual node with the same node name as the cluster name from a plurality of virtual nodes in the tenant cluster as the target node.

In some optional embodiments, the cluster deployment system for edge computing further includes a virtual node agent (virtual-kubelet) configured to create a plurality of virtual nodes in the tenant cluster in one-to-one correspondence with a plurality of preset edge clusters, the node names of the plurality of virtual nodes being identical in one-to-one correspondence with the cluster names of the plurality of preset edge clusters.

In some alternative embodiments, the scheduler is specifically configured to set the value of the node field of the target container group to the node name of the target node.

In some alternative embodiments, the virtual node agent is configured to notify the target cluster to create a real container group corresponding to the target container group, and to receive deployment information of the real container group sent by the target cluster, and to synchronize the deployment information onto the target container group.

Further functional descriptions of the above respective modules and respective units are the same as those of the above corresponding embodiments, and are not repeated here.

The cluster deployment device for edge computing in this embodiment is presented in the form of a functional unit, where a unit refers to an ASIC (Application SPECIFIC INTEGRATED Circuit) Circuit, a processor and a memory that execute one or more software or firmware, and/or other devices that can provide the above functions.

The embodiment of the disclosure also provides a computer device, which is provided with the cluster deployment device for edge computing shown in the figure 6.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a computer device according to an alternative embodiment of the disclosure, as shown in fig. 7, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components are communicatively coupled to each other using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions executing within the computer device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In some alternative embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories and multiple memories. Also, multiple computer devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 10 is illustrated in fig. 7.

The processor 10 may be a central processor, a network processor, or a combination thereof. The processor 10 may further include a hardware chip, among others. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general-purpose array logic, or any combination thereof.

Wherein the memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the methods shown in implementing the above embodiments.

The memory 20 may include a storage program area that may store an operating system, at least one application program required for functions, and a storage data area; the storage data area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, memory 20 may optionally include memory located remotely from processor 10, which may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid state disk; the memory 20 may also comprise a combination of the above types of memories.

The computer device also includes a communication interface 30 for the computer device to communicate with other devices or communication networks.

The presently disclosed embodiments also provide a computer readable storage medium, and the methods described above according to the presently disclosed embodiments may be implemented in hardware, firmware, or as recordable storage medium, or as computer code downloaded over a network that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and is to be stored in a local storage medium, such that the methods described herein may be stored on such software processes on a storage medium using a general purpose computer, special purpose processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, a solid state disk or the like; further, the storage medium may also comprise a combination of memories of the kind described above. It will be appreciated that a computer, processor, microprocessor controller or programmable hardware includes a storage element that can store or receive software or computer code that, when accessed and executed by the computer, processor or hardware, implements the methods illustrated by the above embodiments.

Although embodiments of the present disclosure have been described in connection with the accompanying drawings, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the disclosure, and such modifications and variations are within the scope defined by the appended claims.

Claims (12)

1. A cluster deployment method for edge computing, the method comprising:

In response to the container group creation request, creating a target container group in the tenant cluster;

Screening target clusters for bearing the target container group from a plurality of preset edge clusters, wherein the preset edge clusters are respectively connected with the tenant clusters in a communication way;

Selecting a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, wherein the plurality of virtual nodes are in one-to-one correspondence with the plurality of preset edge clusters;

And configuring node fields of the target container group according to the node name of the target node so as to schedule the target container group to the target node.

2. The method according to claim 1, wherein a preset tag is provided in the container group creation request, the preset tag being used for characterizing configuration requirements of the target container group; the screening the target cluster for bearing the target container group from a plurality of preset edge clusters comprises the following steps:

And screening target clusters from the plurality of preset edge clusters according to the preset labels and the resource allowance of the plurality of preset edge clusters.

3. The method of claim 2, wherein the preset tag comprises at least one of a regional tag, a machine room tag, an operator tag, and a city tag; the screening the target cluster from the plurality of preset edge clusters based on the preset label and the resource margins of the plurality of preset edge clusters includes:

screening out preset edge clusters conforming to the preset labels from the plurality of preset edge clusters;

and acquiring the resource allowance of the preset edge cluster which accords with the preset label, and determining one preset edge cluster of which the resource allowance meets the operation requirement of the target container group as the target cluster.

4.A method according to any one of claim 1 to 3, wherein,

The deployment mode of the preset edge cluster comprises the following steps: and deploying a preset edge cluster in an edge machine room.

5. The method of claim 1, wherein the cluster name of the target cluster is the same as the node name of the target node; the selecting a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster includes:

obtaining a cluster name of the target cluster;

and selecting a virtual node with the same node name as the cluster name from a plurality of virtual nodes in the tenant cluster as the target node.

6. The method of claim 5, wherein prior to said responding to the container group creation request, the method further comprises:

creating a plurality of virtual nodes in the tenant clusters, wherein the virtual nodes are in one-to-one correspondence with the preset edge clusters, and node names of the virtual nodes are the same as cluster names of the preset edge clusters in one-to-one correspondence.

7. The method according to any one of claims 1 to 3 or 5 or 6, wherein configuring the node fields of the target container group according to the node name of the target node comprises:

and setting the value of the node field of the target container group as the node name of the target node.

8. The method according to any one of claims 1 to 3 or 5 or 6, wherein after the configuring of the node fields of the target container group according to the node name of the target node, the method further comprises:

Notifying the target cluster to create a real container group corresponding to the target container group;

and receiving the deployment information of the real container group sent by the target cluster, and synchronizing the deployment information to the target container group.

9. A cluster deployment apparatus for edge computing, the apparatus comprising:

A creation module for creating a target container group in the tenant cluster in response to a container group creation request;

the screening module is used for screening target clusters for bearing the target container group from a plurality of preset edge clusters, and the preset edge clusters are respectively connected with the tenant clusters in a communication mode;

A selection module, configured to select a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, where the plurality of virtual nodes are in one-to-one correspondence with the plurality of preset edge clusters;

And the updating module is used for configuring the node field of the target container group according to the node name of the target node so as to schedule the target container group to the target node.

10. A cluster deployment system for edge computing, the system comprising a tenant cluster and a target cluster, the tenant cluster comprising an application programming interface server and a scheduler;

The application programming interface server is used for responding to a container group creation request and creating a target container group in the tenant cluster;

The scheduler is configured to screen the target cluster for carrying the target container group from a plurality of preset edge clusters, select a target node corresponding to the target cluster from a plurality of virtual nodes in the tenant cluster, and configure a node field of the target container group according to a node name of the target node, so as to schedule the target container group onto the target node;

the plurality of preset edge clusters are respectively in communication connection with the tenant clusters, and the plurality of virtual nodes are in one-to-one correspondence with the plurality of preset edge clusters.

11. A computer device, comprising:

A memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the cluster deployment method for edge computing of any of claims 1 to 8.

12. A computer-readable storage medium having stored thereon computer instructions for causing a computer to perform the cluster deployment method for edge computing of any of claims 1 to 8.