Intelligent load balancing scheduling method based on cross-cloud federated cluster

Abstract

The embodiment of the invention provides an intelligent load balancing scheduling method based on a cross-cloud federated cluster. The technical problem of load scheduling in a multi-cluster environment can be solved. The method comprises the following steps: registering an application created on the cloud to an automatic receiving and managing queue of a resource scheduling module, wherein the resource scheduling module further comprises a plurality of clusters; obtaining a resource state according to a preset time interval; receiving an input service request; in response to the service request, executing a service arrangement operation to obtain a service arrangement result; and executing an application arrangement template analysis operation to obtain an application arrangement template analysis result, and then, calling a resource state, executing a computing power resource adaptation operation and a scheduling strategy configuration operation according to a service arrangement result and an application arrangement template analysis result to obtain a computing power resource adaptation result and a scheduling strategy configuration result, and issuing a scheduling instruction to a target cluster according to the computing power resource adaptation result and the scheduling strategy configuration result. Therefore, a cross-cluster distribution strategy is provided, the multi-cluster scheduling capability is improved, and load scheduling in a multi-cluster environment is realized.

Description

Technical Field

[0001] This application relates to the technical field of information technology infrastructure, and particularly to an intelligent load balancing scheduling method based on a cross-cloud federated cluster. Background Art

[0002] With the continuous development of cloud computing, the information technology (IT) infrastructure of enterprises has gradually become more complex. Enterprises not only need to deploy in multiple regions, but also need to utilize multiple cloud service providers or hybrid cloud environments to meet the diverse needs of their businesses. With the wide application of multi-cluster architectures, how to efficiently manage and coordinate resources and services among these clusters has become one of the main challenges faced by enterprises. Taking computing power scheduling as an example, enterprises need to flexibly schedule computing resources between different regions and cloud service providers according to the requirements of applications.

[0003] Currently, many enterprises use upper-layer management platforms to manage different clusters separately for resource allocation and scheduling. That is, for the management and application creation of multi-clusters, only the upper-layer management platform can operate on different clusters individually, and there are resources going online in a single cluster.

[0004] However, this method can only manage resources for a single cluster and cannot achieve cross-cluster resource scheduling and collaboration. Since each cluster operates independently, it is also difficult to implement cross-cluster fault migration and automated recovery functions, making it more difficult to ensure the high availability and reliability of services in a multi-cluster environment. Summary of the Invention

[0005] The embodiments of this application provide an intelligent load balancing scheduling method based on a cross-cloud federated cluster, which can solve the technical problems of load scheduling in a multi-cluster environment.

[0006] To achieve the above object, the embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides an intelligent load balancing and scheduling method based on a cross-cloud federated cluster. The intelligent load balancing and scheduling method based on a cross-cloud federated cluster includes: registering an application created on the cloud into the automatic management queue of the resource scheduling module; the resource scheduling module further includes a container cluster queue; the container cluster queue includes multiple clusters; obtaining the resource status at preset time intervals; the resource status includes the resource status of the applications in the automatic management queue and the resource status of the multiple clusters; receiving an input service request; in response to the service request, performing a service orchestration operation to obtain a service orchestration result; and performing an application orchestration template parsing operation to obtain an application orchestration template parsing result; after the service orchestration operation and the application orchestration template parsing operation are completed, retrieving the resource status; based on the resource status, according to the service orchestration result and the application orchestration template parsing result, performing a computing power resource adaptation operation to obtain a computing power resource adaptation result; and performing a scheduling policy configuration operation to obtain a scheduling policy configuration result; according to the computing power resource adaptation result and the scheduling policy configuration result, sending a scheduling instruction to a target cluster; the target cluster is at least one of the multiple clusters.

[0008] Based on the above description of the intelligent load balancing and scheduling method based on a cross-cloud federated cluster provided by the embodiment of the present application, it can be seen that the intelligent load balancing and scheduling method based on a cross-cloud federated cluster includes uniformly managing the scattered container cluster computing power resources in a federated cluster manner, providing an all-round, full-perspective, and full-process intelligent scheduling tool for business operation and maintenance, operation, and managers, and realizing the rapid deployment of user services and the elastic expansion of computing power resources. In the intelligent computing power scheduling platform, the intelligent scheduling of computing power resources is completed by combining the application registration of the development cloud and the cluster resource monitoring data of the cloud herding platform. Based on the perceived cluster status data, the platform will trigger different scheduling policies to complete the scheduling between multiple clusters. By analyzing the overall process of the intelligent computing power scheduling platform, the scheduling process of computing power can be further refined. Improve the multi-cluster scheduling ability, improve the utilization rate of multi-cluster resources, share the high load situation of each cluster, overall control the cluster load capacity, and provide a cross-cluster allocation strategy. In this way, load scheduling in a multi-cluster environment is realized.

[0009] In a feasible implementation manner of the first aspect, a global control panel is set up, which simplifies the deployment and management of applications across multiple clusters and clouds, thereby reducing the operation and maintenance complexity of the multi-cloud environment.

[0010] In a feasible implementation manner of the first aspect, the intelligent load balancing and scheduling method based on a cross-cloud federated cluster further includes: receiving the dynamic resource allocation result and the application deployment result of the target cluster, and visually presenting the dynamic resource allocation result and the application deployment result.

[0011] In a feasible implementation of the first aspect, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: updating the current resource status based on the dynamic resource allocation result and the application deployment result.

[0012] In a feasible implementation of the first aspect, when performing the step of executing the scheduling policy configuration operation to obtain the scheduling policy configuration result, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: setting a scheduling policy; creating a scheduling task based on the scheduling policy; starting and executing a scheduling operation based on the scheduling task; and confirming the scheduling result based on the scheduling operation.

[0013] In a feasible implementation of the first aspect, the scheduling task includes an automatic scheduling task and a manual scheduling task. Before performing the step of starting and executing a scheduling operation based on the scheduling task, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: if the scheduling task is a manual scheduling task, manually configuring a distribution policy to specify the target cluster and the weight ratio; if the scheduling task is an automatic scheduling task, determining the target cluster and the weight ratio according to the scheduling algorithm configured by different scheduling policies; when performing the step of starting and executing a scheduling operation based on the scheduling task, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: starting and executing a scheduling operation based on the scheduling task and the distribution policy according to the target cluster and the weight ratio.

[0014] In a feasible implementation of the first aspect, before performing the step of starting and executing a scheduling operation based on the scheduling task and the distribution policy according to the target cluster and the weight ratio, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: configuring an overlay policy and a load policy; when performing the step of starting and executing a scheduling operation based on the scheduling task and the distribution policy according to the target cluster and the weight ratio, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: starting and executing a scheduling operation based on the scheduling task, the distribution policy, the overlay policy, and the load policy according to the target cluster and the weight ratio.

[0015] In a feasible implementation of the first aspect, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: updating the resource status of the cluster with the cluster identity as the identifier; triggering a first scheduling policy when the computing power load of a certain cluster exceeds the policy threshold; the first scheduling policy is a load-based scheduling policy; after the overload protection operation, filtering the cluster identities of the clusters in the first member cluster whose load exceeds the set threshold to obtain a second member cluster; selecting the cluster with the lowest load in the second member cluster as the scheduling target cluster; and sending the cluster identity in the scheduling target cluster to a first object to complete the scheduling of the scheduling target cluster.

[0016] In a second aspect, an embodiment of the present application provides an intelligent load balancing and scheduling system based on a cross-cloud federated cluster, including: a resource scheduling module, a computing power perception module, a computing power scheduling operation module, and a computing power orchestration and scheduling module; the resource scheduling module is configured to register an application created on the cloud into the automatic management queue of the resource scheduling module; the resource scheduling module further includes a container cluster queue; the container cluster queue includes multiple clusters; the computing power perception module is configured to obtain the resource status according to a preset time interval; the resource status includes the resource status of the applications in the automatic management queue and the resource status of multiple clusters; the computing power scheduling operation module is configured to receive an input service request and send the service request to the computing power orchestration and scheduling module; the computing power orchestration and scheduling module is configured to, in response to the service request, perform a service orchestration operation to obtain a service orchestration result; and, perform an application orchestration template parsing operation to obtain an application orchestration template parsing result; when the service orchestration operation and the application orchestration template parsing operation are completed, the computing power orchestration and scheduling module is further configured to retrieve the resource status of the computing power perception module; the computing power orchestration and scheduling module is further configured to, based on the resource status, according to the service orchestration result and the application orchestration template parsing result, perform a computing power resource adaptation operation to obtain a computing power resource adaptation result; and, perform a scheduling policy configuration operation to obtain a scheduling policy configuration result; the computing power orchestration and scheduling module is further configured to, according to the computing power resource adaptation result and the scheduling policy configuration result, issue a scheduling instruction to at least one of the multiple clusters.

[0017] Unify the management of dispersed container cluster computing power resources in a federated cluster manner, and provide an all-round, full-perspective, and full-process intelligent scheduling tool for business operation and maintenance, operation, and managers, so as to achieve the rapid deployment of user services and the elastic expansion of computing power resources. In the intelligent computing power scheduling platform, the intelligent scheduling of computing power resources is completed by combining the application registration of the development cloud and the cluster resource monitoring data of the cloud pasture platform. Based on the perceived cluster status data, the platform will trigger different scheduling policies to complete the scheduling between multiple clusters. By analyzing the overall process of the intelligent computing power scheduling platform, the scheduling process of computing power can be further refined. Improve the multi-cluster scheduling ability, enhance the utilization rate of multi-cluster resources, share the high load of each cluster, overall regulate the cluster load capacity, and provide a cross-cluster allocation strategy. In this way, load scheduling in a multi-cluster environment is achieved.

[0018] In a third aspect, an embodiment of the present application provides an intelligent load balancing and scheduling system based on a cross-cloud federated cluster. The intelligent load balancing and scheduling system based on a cross-cloud federated cluster includes: at least one processor; a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method provided in the first aspect.

[0019] The intelligent load balancing scheduling system based on the cross-cloud federated cluster executes the method provided in the first aspect, uniformly manages the scattered container cluster computing power resources in the form of a federated cluster, provides an intelligent scheduling tool for business operation and maintenance, operation, and management in all aspects, perspectives, and processes, and realizes the rapid deployment of user services and the elastic expansion of computing power resources. In the intelligent computing power scheduling platform, the intelligent scheduling of computing power resources is completed by combining the application registration of the development cloud and the cluster resource monitoring data of the Muyun platform. Based on the perceived cluster status data, the platform will trigger different scheduling strategies to complete the scheduling between multiple clusters. By analyzing the overall process of the intelligent computing power scheduling platform, the scheduling process of computing power can be further refined. Improve the multi-cluster scheduling ability, enhance the utilization rate of multi-cluster resources, share the high load of each cluster, overall control the cluster load ability, and provide a cross-cluster allocation strategy. In this way, load scheduling in a multi-cluster environment is realized.

[0020] In a fourth aspect, an embodiment of the present application provides a computer-readable medium, on which computer program instructions are stored, and the computer program instructions can be executed by a processor to implement the method provided in the first aspect.

[0021] The computer program instructions in the computer-readable medium implement the method provided in the first aspect, uniformly manage the scattered container cluster computing power resources in the form of a federated cluster, provide an intelligent scheduling tool for business operation and maintenance, operation, and management in all aspects, perspectives, and processes, and realize the rapid deployment of user services and the elastic expansion of computing power resources. In the intelligent computing power scheduling platform, the intelligent scheduling of computing power resources is completed by combining the application registration of the development cloud and the cluster resource monitoring data of the Muyun platform. Based on the perceived cluster status data, the platform will trigger different scheduling strategies to complete the scheduling between multiple clusters. By analyzing the overall process of the intelligent computing power scheduling platform, the scheduling process of computing power can be further refined. Improve the multi-cluster scheduling ability, enhance the utilization rate of multi-cluster resources, share the high load of each cluster, overall control the cluster load ability, and provide a cross-cluster allocation strategy. In this way, load scheduling in a multi-cluster environment is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic structural diagram of an intelligent load balancing scheduling system based on a cross-cloud federated cluster provided by an embodiment of the present application;

[0023] Figure 2 It is a schematic flowchart of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application;

[0024] Figure 3 It is an interaction schematic diagram of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application;

[0025] Figure 4 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 1 ;

[0026] Figure 5 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 2 ;

[0027] Figure 6 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 3 ;

[0028] Figure 7 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 4 ;

[0029] Figure 8 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 5 ;

[0030] Figure 9 Schematic diagram of the visual interface of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application Figure 6 ;

[0031] Figure 10 Schematic diagram of the scenario of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application;

[0032] Figure 11 Schematic diagram of the process of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application. Detailed implementation manners

[0033] Next, the technical solutions in the embodiments of the present invention will be described with reference to the accompanying drawings in the embodiments of the present invention. Among them, in the description of the embodiments of the present invention, unless otherwise specified, "a plurality" means two or more than two. "At least one (item)" or its similar expression refers to any combination of these items, including any combination of single item (item) or plural items (items). For example, at least one (item) of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, c can be single or multiple.

[0034] In addition, in order to facilitate the clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second" and the like are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art will understand that the words "first", "second" and the like do not limit the quantity and execution order, and the words "first", "second" and the like do not necessarily limit the difference. At the same time, in the embodiments of the present invention, the words "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or design. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0035] The principles and features of the present application are described below. The examples given are only used to explain the present application and are not used to limit the scope of the present application.

[0036] Container cluster management system (Kubernetes) is a group of nodes that run containerized applications. These nodes can be physical servers or virtual machines, and they work together to provide lifecycle management of containerized applications, including deployment, expansion, management, and fault self-healing. In some scenarios, the Kubernetes cluster has the following limitations: The number of nodes does not exceed 5,000. The number of Pods per node does not exceed 110. The total number of Pods does not exceed 150,000. The total number of containers does not exceed 300,000. Due to the above cluster limitations, some businesses cannot be completed in a single cluster and may need to be deployed on multiple Kubernetes clusters. In this case, multiple clusters need to be managed.

[0037] The embodiment of the present application provides an intelligent load balancing scheduling method based on a cross-cloud federated cluster, which is applicable to various multi-cluster management scenarios. For example, the scenario of dual active in the same city. Compared with the related technology that deploys services in a single large-scale cluster, the embodiment of the present application can improve the security of business processing by deploying services in multiple smaller clusters. For example, in the event of a failure in a single cluster, failover can be performed. In addition, by setting up a federated cluster, the embodiment of the present application can perform resource synchronization and cross-cluster service discovery between clusters to improve the service experience.

[0038] The embodiment of the present application provides an intelligent load balancing scheduling system based on a cross-cloud federated cluster, which can execute the intelligent load balancing scheduling method based on a cross-cloud federated cluster provided in the embodiment of the present application. Figure 1 A structural diagram of an intelligent load balancing scheduling system based on a cross-cloud federated cluster is provided in an embodiment of the present application.

[0039] like Figure 1As shown in the figure, the intelligent load balancing scheduling system 001 based on a cross-cloud federated cluster includes at least one processor 011 and a memory 012 communicatively connected to the at least one processor. Among them, the memory 012 stores instructions executable by the at least one processor 011. When the instructions are executed by the at least one processor 011, the at least one processor 011 is enabled to execute the intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application.

[0040] As Figure 3 shown, in some embodiments, an intelligent load balancing scheduling system based on a cross-cloud federated cluster provided by an embodiment of the present application includes: a resource scheduling module, a computing power perception module, a computing power scheduling operation module, and a computing power orchestration scheduling module.

[0041] The resource scheduling module is configured to register an application created on the cloud into the automatic management queue of the resource scheduling module. The resource scheduling module further includes a container cluster queue. The container cluster queue includes multiple clusters.

[0042] The computing power perception module is configured to obtain resource status according to a preset time interval. The resource status includes the resource status of applications in the automatic management queue and the resource status of multiple clusters.

[0043] The computing power scheduling operation module is configured to receive an input service request and send the service request to the computing power orchestration scheduling module.

[0044] The computing power orchestration scheduling module is configured to, in response to the service request, perform a service orchestration operation to obtain a service orchestration result. And, perform an application orchestration template parsing operation to obtain an application orchestration template parsing result.

[0045] When the service orchestration operation and the application orchestration template parsing operation are completed, the computing power orchestration scheduling module is further configured to retrieve the resource status of the computing power perception module.

[0046] The computing power orchestration scheduling module is further configured to, based on the resource status, according to the service orchestration result and the application orchestration template parsing result, perform a computing power resource adaptation operation to obtain a computing power resource adaptation result. And, perform a scheduling policy configuration operation to obtain a scheduling policy configuration result.

[0047] The computing power orchestration scheduling module is further configured to issue a scheduling instruction to at least one of the multiple clusters according to the computing power resource adaptation result and the scheduling policy configuration result.

[0048] Figure 2 is a schematic flowchart of an intelligent load balancing scheduling method based on a cross-cloud federated cluster provided by an embodiment of the present application. As Figure 2 and Figure 3As shown, in some embodiments, the intelligent load balancing scheduling method based on a cross-cloud federated cluster includes the following steps:

[0049] S1, Register the applications created on the cloud to the automatic management queue of the resource scheduling module.

[0050] In some embodiments, register the applications created on the cloud to the automatic management queue of the resource scheduling module. Exemplarily, the resource scheduling module uniformly registers and manages the container cluster, and automatically manages the applications created on the R & D cloud through reverse registration. Among them, the applications that fail to join the automatic management queue need to be manually registered to this platform. On the contrary, if the managed applications need to be released and upgraded, they need to return to the R & D cloud for application modification and then re-register for management.

[0051] The resource scheduling module further includes a container cluster queue.

[0052] The container cluster queue includes multiple clusters.

[0053] S2, Obtain the resource status according to a preset time interval.

[0054] In some embodiments, the computing power awareness module obtains the resource status according to a preset time interval.

[0055] The resource status includes the resource status of the applications in the automatic management queue and the resource status of multiple clusters.

[0056] In some embodiments, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes:

[0057] S21, Update the resource status of the cluster with the cluster identity as the identifier.

[0058] The cluster identity (Identification, ID) is used as the identifier to distinguish different clusters.

[0059] Through the resource status, that is, the resource monitoring data of the applications and clusters, it supports the completion of intelligent scheduling of computing power resources.

[0060] S3, Receive the input business request.

[0061] In some embodiments, the computing power scheduling and operation module receives the input business request and sends the business request to the computing power orchestration and scheduling module.

[0062] S4, In response to the business request, perform a business orchestration operation to obtain a business orchestration result; and perform an application orchestration template parsing operation to obtain an application orchestration template parsing result.

[0063] In some embodiments, the computing power orchestration scheduling module performs a business orchestration operation in response to the business request to obtain a business orchestration result, and performs an application orchestration template parsing operation to obtain an application orchestration template parsing result.

[0064] S5, when the business orchestration operation and the application orchestration template parsing operation are completed, the resource status is retrieved.

[0065] In some embodiments, the computing power orchestration and scheduling module calls the resource status of the computing power perception module.

[0066] S6, based on the resource status, according to the business orchestration results and the application orchestration template parsing results, execute the computing resource adaptation operation to obtain the computing resource adaptation result; and execute the scheduling policy configuration operation to obtain the scheduling policy configuration result.

[0067] In some embodiments, the computing power scheduling module performs a computing power resource adaptation operation based on the resource status, the business scheduling result and the application scheduling template parsing result to obtain a computing power resource adaptation result, and performs a scheduling policy configuration operation to obtain a scheduling policy configuration result.

[0068] In some embodiments, when executing step S6, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0069] S61, setting a scheduling strategy.

[0070] In some embodiments, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0071] S611: When the computing load of a certain cluster exceeds the policy threshold, trigger the first scheduling policy.

[0072] The first scheduling strategy is a load-based scheduling strategy. Figure 5 As shown, exemplarily, the policy thresholds include "cluster central processing unit (CPU) is greater than or equal to 15" and / or "cluster memory (MEM) is greater than or equal to 15".

[0073] S62: Create a scheduling task based on the scheduling policy.

[0074] Scheduling tasks include automatic scheduling tasks and manual scheduling tasks. Figure 6 As shown, the task content of the scheduling task includes: at least one of the task name, task type, namespace, federated application, cluster, application business domain and adjustment strategy. Exemplarily, the creation of the scheduling task can be completed by clicking the confirmation button on the screen.

[0075] likeFigure 11 As shown, exemplarily, when the first scheduling policy is triggered, when executing step S62, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0076] S621, after the overload protection operation, filter the cluster identities in the first member cluster whose load exceeds the set threshold to obtain the second member cluster.

[0077] S622, select the cluster with the lowest load in the second member cluster as the scheduling target cluster.

[0078] S623, send the cluster identity in the scheduling target cluster to the first object to complete the scheduling of the scheduling target cluster.

[0079] As Figure 11 shown, exemplarily, the first object is "propagation policy". The scheduling of the target cluster is completed through components of the multi-cloud container orchestration management platform (such as, karmada scheduler).

[0080] S63, based on the scheduling task, start and execute the scheduling operation.

[0081] As Figure 7 shown, in some embodiments, the scheduling task includes an automatic scheduling task and a manual scheduling task. Before executing step S63, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes step S621 and step S622:

[0082] S621, if the scheduling task is a manual scheduling task, manually configure the distribution policy to specify the target cluster and the weight ratio.

[0083] As Figure 4 shown, in some embodiments, the distribution policy includes the distribution time, the rollback time, and the distribution algorithm. The distribution algorithm includes minimum load first, maximum available capacity first, and / or load balancing first.

[0084] S622, if the scheduling task is an automatic scheduling task, determine the target cluster and the weight ratio according to the scheduling algorithm configured according to different scheduling policies.

[0085] After executing step S621 and step S622, when executing step S63, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0086] S631, based on the scheduling task and the distribution policy, start and execute the scheduling operation according to the target cluster and the weight ratio.

[0087] As Figure 8As shown, in some embodiments, before performing step S631, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes step S630:

[0088] S630, configure an overlay policy and a load policy.

[0089] As Figure 8 shown, exemplarily, the overlay policy includes one or more of an application name, an overlay type, a target cluster, and an operation type. It can be understood that there can be multiple application names, overlay types, target clusters, and operation types.

[0090] After performing step S630, when performing step S631, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes:

[0091] S6311, based on the scheduling task, the distribution policy, the overlay policy, and the load policy, according to the target cluster and the weight ratio, start and execute a scheduling operation.

[0092] In this way, after determining the distribution policy, it is possible to select whether to modify the overlay policy and the load policy. After completing the configuration of these three policies, a scheduling task can be started to complete the cross-cluster scheduling of the task.

[0093] S64, confirm the scheduling result based on the scheduling operation.

[0094] Exemplarily, taking the computing power scheduling policy based on load as an example, the orchestration scheduling module will determine the scheduling and deployment decision of the application based on the computing power cluster status obtained by the sensing module, combined with the overload protection policy, and uniformly orchestrate the determined cluster ID and the application programming interface (API) of the application to form a determined scheduling decision.

[0095] S7, send a scheduling instruction to the target cluster according to the computing power resource adaptation result and the scheduling policy configuration result.

[0096] The target cluster is at least one of multiple clusters.

[0097] As Figure 9 shown, in some embodiments, the computing power orchestration scheduling module sends a scheduling instruction to the target cluster according to the computing power resource adaptation result and the scheduling policy configuration result. The target cluster is at least one of multiple clusters.

[0098] As Figure 10As shown, in some embodiments, a multi-cloud container orchestration management system (such as the Karmada component) completes scheduling among multiple clusters. Exemplarily, the scheduling decision will be sent to Karmada in the form of Custom Resource Definitions (CRDs), thereby triggering the scheduling process within the multi-cluster management platform.

[0099] The following presents the scheduling process within Karmada.

[0100] Exemplarily, after the orchestration scheduling module defines the second object "deployment", it will match through the first object "propagation Policy" and generate a binding relationship, namely the third object "propagationBinding". Then, through an override policy, a fourth object "works" is generated. This fourth object "works" is actually an encapsulation of each resource object in the sub-cluster.

[0101] Among them, the working mechanisms of the first object "propagation Policy" and the third object "propagation Binding" are as follows.

[0102] After defining the first object "propagation Policy", the previously defined first resource template (K8s resource template) to be deployed will automatically match with the first object "propagation Policy". After the match, the second object "deployment" will be distributed to the target clusters. For example, the target clusters are three clusters, namely Cluster A, Cluster B, and Cluster C. In this way, the second object "deployment" has bindings with Cluster A, Cluster B, and Cluster C, and this binding relationship is the fifth object "Resource Bindding".

[0103] After the third object "propagation Binding" is generated, for example, three clusters are generated, namely cluster D, cluster E, and cluster F. After finding cluster D, cluster E, and cluster F, the Binding Controller will work. Then, a binding relationship based on the resource template and the second object "deployment" is generated, and then the fourth object "works" is generated. It can be understood that the fourth object "works" can be an objectively existing "works" resource created in a cluster (such as k8s). The fourth object "works" can be a presentation of the actual object resources across clusters in the control cluster. The fourth object "works" as a whole is an encapsulation of the resources in a certain sub-cluster.

[0104] At the same time, a status of the fourth object "works" has a feedback of the sub-cluster resources. The entire data serialization format file for work (work yaml) can be seen, and it can be seen that the entire sub-cluster is under the original file directory (manifests). It should be noted that if the central processing unit or memory resources in a cluster where a resource to be scheduled is about to be located are insufficient, the scheduling algorithm will re-match other clusters for scheduling.

[0105] Exemplarily, when a member cluster does not have enough resources to accommodate the sixth object Pod in it, karmada will re-schedule the sixth object pod of other member clusters. In this way, the scaling and scheduling of the cluster are realized.

[0106] In some embodiments, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0107] S8, receiving the dynamic resource allocation result and the application deployment result of the target cluster, and visually presenting the dynamic resource allocation result and the application deployment result.

[0108] In some embodiments, the computing power scheduling operation module receives the dynamic resource allocation result and the application deployment result of the target cluster, and visually presents the dynamic resource allocation result and the application deployment result.

[0109] In some embodiments, the intelligent load balancing scheduling method based on the cross-cloud federated cluster further includes:

[0110] S9, updating the current resource status based on the dynamic resource allocation result and the application deployment result.

[0111] In some embodiments, based on the dynamic resource allocation result and the application deployment result, the computing power perception module updates the current resource status.

[0112] Unify and manage the scattered container cluster computing power resources through the federal cluster method, and provide an all-round, full-perspective, and full-process intelligent scheduling tool for business operation and maintenance, operation, and managers to achieve the rapid deployment of user services and the elastic expansion of computing power resources. In the intelligent computing power scheduling platform, the intelligent scheduling of computing power resources is completed by combining the application registration of the development cloud and the cluster resource monitoring data of the Muyun platform. Based on the perceived cluster status data, the platform will trigger different scheduling strategies to complete the scheduling between multiple clusters. By analyzing the overall process of the intelligent computing power scheduling platform, the scheduling process of computing power can be further refined. Improve the multi-cluster scheduling ability, enhance the utilization rate of multi-cluster resources, share the high load of each cluster, overall control the cluster load capacity, and provide a cross-cluster allocation strategy. In this way, load scheduling in a multi-cluster environment is achieved.

[0113] Based on the same application concept, an intelligent load balancing scheduling system based on a cross-cloud federal cluster is also provided in an embodiment of the present application. The method corresponding to the intelligent load balancing scheduling system based on the cross-cloud federal cluster may be the intelligent load balancing scheduling method based on the cross-cloud federal cluster in the foregoing embodiment, and the principle of solving problems is similar to that of this method. The intelligent load balancing scheduling system based on the cross-cloud federal cluster provided in the embodiment of the present application includes: at least one processor; and a memory communicatively connected to the at least one processor; 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 methods and / or technical solutions of multiple embodiments of the present application described above.

[0114] Another embodiment of the present application also provides a computer-readable storage medium, on which computer program instructions are stored, and the computer program instructions can be executed by a processor to implement the methods and / or technical solutions of any one or more embodiments of the present application described above.

[0115] Specifically, one or more combinations of computer-readable media can be adopted in this embodiment. The computer-readable media can be computer-readable signal media or computer-readable storage media. The computer-readable storage media can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage media can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0116] The computer-readable signal media can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal media can also be any computer-readable media other than the computer-readable storage media, and this computer-readable media can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0117] The program code contained on the computer-readable media can be transmitted using any appropriate medium, including - but not limited to - wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0118] The computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages - such as Java, Smalltalk, C++, and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0119] The flowcharts or block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of devices, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0120] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0121] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or page components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the couplings, direct couplings, or communication connections shown or discussed among each other may be indirect couplings or communication connections through some interfaces, devices, or units, and may be in electrical, mechanical, or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] In addition, the functional units in various embodiments of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0124] The integrated unit implemented in the form of software functional units can be stored in a computer-readable storage medium. The above-mentioned software functional units are stored in a storage medium and include several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute some steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of the present application.

[0126] In addition, obviously, the term "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices stated in the apparatus claims can also be implemented by one unit or device through software or hardware. The terms such as first and second are used to represent names and do not indicate any specific order.

Claims

1. An intelligent load balancing scheduling method based on a cross-cloud federated cluster, characterized in that, Including: Register the applications created on the cloud to the automatic management queue of the resource scheduling module; the resource scheduling module further includes a container cluster queue; the container cluster queue includes multiple clusters; Obtain the resource status at preset time intervals; the resource status includes the resource status of the applications in the automatic management queue and the resource status of the multiple clusters; Receive the input business request; In response to the business request, perform a business orchestration operation to obtain a business orchestration result; And perform an application orchestration template parsing operation to obtain an application orchestration template parsing result; When the business orchestration operation and the application orchestration template parsing operation are completed, retrieve the resource status; Based on the resource status, according to the business orchestration result and the application orchestration template parsing result, perform a computing power resource adaptation operation to obtain a computing power resource adaptation result; And perform a scheduling policy configuration operation to obtain a scheduling policy configuration result; According to the computing power resource adaptation result and the scheduling policy configuration result, issue a scheduling instruction to the target cluster; The target cluster is at least one of the multiple clusters.

2. The intelligent load balancing scheduling method based on a cross-cloud federated cluster according to claim 1, wherein The intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Receive the dynamic resource allocation result and the application deployment result of the target cluster, and visually present the dynamic resource allocation result and the application deployment result.

3. The intelligent load balancing scheduling method based on a cross-cloud federated cluster according to claim 2, wherein, The intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Update the current resource status based on the dynamic resource allocation result and the application deployment result.

4. The intelligent load balancing scheduling method based on a cross-cloud federated cluster according to any one of claims 1-3, characterized in that, When performing the step of performing a scheduling policy configuration operation to obtain a scheduling policy configuration result, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Set a scheduling policy; Based on the scheduling policy, create a scheduling task; Based on the scheduling task, start and execute a scheduling operation; Based on the scheduling operation, confirm the scheduling result.

5. The intelligent load balancing scheduling method based on cross-cloud federated clusters according to claim 4, wherein, The scheduling task includes an automatic scheduling task and a manual scheduling task. Before performing the step of starting and executing a scheduling operation based on the scheduling task, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: If the scheduling task is a manual scheduling task, manually configure a distribution policy to specify the target cluster and the weight ratio; If the scheduling task is an automatic scheduling task, determine the target cluster and the weight ratio according to the scheduling algorithms configured by different scheduling policies; When performing the step of starting and executing a scheduling operation based on the scheduling task, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Based on the scheduling task and the distribution policy, start and execute a scheduling operation according to the target cluster and the weight ratio.

6. The intelligent load balancing scheduling method based on a cross-cloud federated cluster according to claim 5, wherein, Before performing the step of starting and executing a scheduling operation based on the scheduling task and the distribution policy, according to the target cluster and the weight ratio, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Configure an override policy and a load policy; When performing the step of starting and executing a scheduling operation based on the scheduling task and the distribution policy, according to the target cluster and the weight ratio, the intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Based on the scheduling task, the distribution policy, the coverage policy, and the load policy, start and execute a scheduling operation according to the target cluster and the weight ratio.

7. The intelligent load balancing and scheduling method based on cross-cloud federated clusters according to any one of claims 1-3, characterized in that, The intelligent load balancing scheduling method based on a cross-cloud federated cluster further includes: Taking the cluster identity as an identifier, update the resource status of the cluster; In the case where the computing power load of a certain cluster exceeds the policy threshold, trigger a first scheduling policy; the first scheduling policy is a load-based scheduling policy; After the overload protection operation, filter the cluster identities in the first member cluster whose load exceeds the set threshold to obtain a second member cluster; Select the cluster with the lowest load in the second member cluster as the scheduling target cluster; Send the cluster identity in the scheduling target cluster to a first object to complete the scheduling of the scheduling target cluster.

8. An intelligent load balancing and scheduling system based on a cross-cloud federated cluster, characterized in that, Including: A resource scheduling module, a computing power perception module, a computing power scheduling operation module, and a computing power orchestration scheduling module; The resource scheduling module is configured to register an application created on the cloud into the automatic management queue of the resource scheduling module; the resource scheduling module further includes a container cluster queue; the container cluster queue includes multiple clusters; The computing power perception module is configured to obtain the resource status at preset time intervals; the resource status includes the resource status of the application in the automatic management queue and the resource status of the multiple clusters; The computing power scheduling operation module is configured to receive an input service request and send the service request to the computing power orchestration scheduling module; The computing power orchestration scheduling module is configured to, in response to the service request, perform a service orchestration operation to obtain a service orchestration result; And perform an application orchestration template parsing operation to obtain an application orchestration template parsing result; In the case where the service orchestration operation and the application orchestration template parsing operation are completed, the computing power orchestration scheduling module is further configured to retrieve the resource status of the computing power perception module; The computing power orchestration scheduling module is further configured to, based on the resource status, according to the service orchestration result and the application orchestration template parsing result, perform a computing power resource adaptation operation to obtain a computing power resource adaptation result; and perform a scheduling policy configuration operation to obtain a scheduling policy configuration result; The computing power orchestration scheduling module is further configured to issue a scheduling instruction to at least one of the multiple clusters according to the computing power resource adaptation result and the scheduling policy configuration result.

9. An intelligent load balancing scheduling system based on a cross-cloud federated cluster, characterized in that, Including: At least one processor; A memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1 to 7.

10. A computer-readable medium having computer program instructions stored thereon, the computer program instructions being executable by a processor to implement the method according to any one of claims 1 to 7.

Images