Cloud computing technology-based instance management method and related apparatus

Abstract

Provided in the present application is a cloud computing technology-based instance management method. The method is applied to a cloud management platform. The cloud management platform is used to manage at least one computing node, at least one instance running on the first computing node. The method comprises: acquiring a first instance creation request input by a tenant, the first instance creation request comprising deployment information, specification information and priority information of a first instance; and, on the basis of the first instance creation request, creating the first instance on the first computing node, the priority of the first instance being different from the priority of the at least one instance, and the first instance running on a resource in the first computing node that satisfies a first priority. During the running process of the first instance, when the priority of the instance is dynamically adjusted, the cloud management platform directly adjusts a corresponding resource policy on the computing node, thereby, while ensuring high availability and high reliability of instances for tenants, improving the capability of dynamically adjusting sensitive services and improving the utilization rate of resources on computing nodes.

Description

A cloud computing-based instance management method and related apparatus

[0001] This application claims priority to Chinese Patent Application No. 202411999706.3, filed with the China National Intellectual Property Administration on December 31, 2024, entitled "An Instance Management Method and Related Device Based on Cloud Computing Technology", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This invention relates to the field of cloud computing technology, and in particular to an instance management method and related apparatus based on cloud computing technology. Background Technology

[0003] With the development of cloud computing technology, traditional applications deploy services using virtual machines, while cloud-native applications utilize agile containers, running them on virtual machine nodes to ensure security and isolation. In traditional "stable-state" application scenarios, a single application is typically deployed on a single virtual machine, resulting in relatively stable processor and memory usage, as well as relatively stable QoS (Quality of Service) requirements. In cloud-native application scenarios, cloud resources are typically managed and provisioned using virtual machine-based containers. A single virtual machine node may run multiple containers, and the types and number of containers can change frequently. In this case, the priority of virtual machines changes due to the frequent changes in the containers they deploy.

[0004] Currently, in cloud infrastructure, virtual machines running in the same host group generally have the same priority. Different priority services need to be isolated through independent host group resource pools or different hosts. This approach cannot support differentiated QoS requirements on a single host, nor can it flexibly support scenarios involving changes in virtual machine priority. With the development of container-based and other agile services, a dynamic instance scheduling solution is needed that can both support differentiated priority guarantees and adapt to changes in priority. Summary of the Invention

[0005] This application provides an instance management method and related apparatus based on cloud computing technology. By configuring the priority of instances to be deployed by tenants through a cloud management platform, instances of different priorities are set up on computing nodes that meet the tenants' priority requirements. This matches the needs of tenants' agile business changes while improving the resource utilization of computing nodes. The technical solution is as follows:

[0006] Firstly, this application provides an instance management method based on cloud computing technology. This method is applied to a cloud management platform for managing infrastructure comprising at least one region, where each region includes at least one availability zone, and each availability zone further includes at least one compute node. At least one instance runs on a first compute node in a first availability zone, and each instance can run different services from different tenants. The method specifically includes the following steps: obtaining a first instance creation request input by a tenant. This first instance creation request includes deployment information, specification information, and priority information for the first instance. The deployment information indicates that the first instance will be deployed to a first availability zone, the specification information indicates the instance specifications, and the priority information indicates that the first instance has a first priority. Further, based on the first instance creation request, a first instance is created on the first compute node in the first availability zone. The priority of the first instance differs from the priority of the at least one instance, and the first instance runs on resources on the first compute node that satisfy the first priority.

[0007] In the solution provided in this application, by combining the instance priority information defined by the tenant with their defined deployment and instance specification information, corresponding instances are deployed in the infrastructure. This provides resource-efficient instances that meet the tenant's priority requirements, and the compute nodes running these instances support the simultaneous operation of instances with different priority requirements. Therefore, when deploying applications or dynamically adjusting instance priorities, tenants do not need to perform further operations such as instance migration. This ensures high availability of tenant instances and the applications running on them while enabling dynamic adjustment of tenant instance priorities, thereby improving resource utilization on compute nodes.

[0008] In conjunction with the first aspect, in one possible implementation of the first aspect, before creating the first instance based on the instance creation request, the method may further include the following steps: obtaining resource load information of multiple computing nodes in the infrastructure, wherein the resource load information is used to indicate the amount of resources used by instances of different priorities deployed on each of the multiple computing nodes.

[0009] In the solution provided in this application, by pre-defining the orchestration of resources with different priorities on the infrastructure, the cloud management platform can more accurately know the usage and remaining amount of resources with different priorities in each computing node in the infrastructure, forming an accurate resource profile of multiple computing nodes. This allows the cloud management platform to accurately identify and call resources that meet the tenant's priority information, instance specification information, and deployment information when the tenant requests to create, deploy, or adjust instances in the infrastructure.

[0010] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform, based on the resource load information of multiple computing nodes in the infrastructure obtained by the cloud management platform, determines that there are resources that meet the first priority in the first computing node according to the deployment information, specification information, priority information and resource load information of the first instance, and creates the first instance on the first computing node.

[0011] In the solution provided in this application, the cloud management platform can obtain accurate resource profiles of each compute node before creating an instance. For example, it can periodically obtain resource profiles of multiple compute nodes in the infrastructure before the tenant enters an instance creation request, or request resource profiles of multiple compute nodes from the infrastructure after obtaining the request. This allows for a more accurate understanding of the amount of resources used and the remaining resources of instances with different priorities deployed on each compute node. The platform then combines the tenant's request for priority information, instance specifications, and deployment information to create the first instance, ensuring that the tenant's instance can be successfully deployed and improving the instance deployment success rate.

[0012] In conjunction with the first aspect, in one possible implementation, the resources provided by each computing node among multiple computing nodes include dedicated resources, shared resources, and excess resources. Specifically, when the first priority is high, the resources satisfying the first priority are dedicated resources, used to instruct an instance to use a first quantity of resources exclusively; or, when the first priority is medium, the resources satisfying the first priority are shared resources, used to instruct multiple instances to share a second quantity of resources; or, when the first priority is low, the resources satisfying the first priority are excess resources, used to instruct multiple instances to share a third quantity of resources, where the third quantity is less than the second quantity. That is, instances with different priorities use different resources in the infrastructure, including the same instance or different instances. After the tenant defines the instance priority, the cloud management platform will determine the computing nodes in the infrastructure that can satisfy the corresponding priority information for resource matching based on the priority information.

[0013] In the solution provided in this application, the cloud management platform can provide tenants with resources that meet various priority requirements. For example, one approach is to establish different resource pools, which can be established on a single compute node or shared across multiple compute nodes. These resource pools can be divided into dedicated resources, shared resources, and premium resources based on priority, each corresponding to a different priority level. High-priority instances can use a certain number of resources in the high-priority resource pool independently; medium-priority instances can share a certain number of resources in the medium-priority resource pool with other instances; and low-priority instances can share a certain number of resources in the low-priority resource pool with more instances. Specifically, the number of resources in the low-priority resource pool is roughly equivalent to the number of resources in the medium-priority resource pool, but the number of instances in the low-priority resource pool is greater than the number of instances in the medium-priority resource pool; or, the number of instances in the low-priority resource pool is roughly equivalent to the number of instances in the medium-priority resource pool, but the number of resources in the low-priority resource pool is less than the number of resources in the medium-priority resource pool. Therefore, by providing different resources to meet the priority instance requirements of tenants, tenants can set different priority instances according to business needs, reduce business costs, and dynamically adjust instance priorities to achieve rapid adjustment of agile business and reduce the negative impact on the business itself, such as latency jitter caused by instance redeployment and further degradation of service quality.

[0014] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform obtains the priority change information of the first instance, the priority change information of the first instance is used to indicate that the priority of the first instance is adjusted from the first priority to the second priority, and according to the priority change information of the first instance, the platform schedules resources on the first computing node that meet the second priority for the use of the first instance.

[0015] In the solution provided in this application, tenants can dynamically adjust the priority of an instance by inputting the instance's priority change information. Alternatively, the instance itself can send the instance's priority change information to the cloud management platform based on the actual operation of the business. The cloud management platform then adjusts the resources available to the instance on the original node to meet the new priority, ensuring that resources are still provided on the same compute node after the instance's priority is dynamically adjusted, thereby improving the reliability and stability of the instance.

[0016] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform obtains a second instance creation request input by the tenant. The second instance creation request includes deployment information and priority information of the second instance. The deployment information of the second instance is used to indicate that the second instance is deployed on the first instance, and the priority information of the second instance is used to indicate that the priority of the second instance is the third priority. Based on the second instance creation request, the second instance is created in the first instance.

[0017] In the solution provided in this application, the cloud management platform can create a second instance on the first instance based on the second instance creation request input by the tenant, and supports the configuration of priority information for the second instance. In this way, it provides the ability to configure priority for sensitive instance scenarios such as deploying containers and serverless functions on virtual machines, thereby further improving resource utilization.

[0018] In conjunction with the first aspect, in one possible implementation of the first aspect, when the first priority and the third priority are different, the cloud management platform schedules resources on the first computing node that meet the third priority to provide to the first instance according to the third priority.

[0019] In the solution provided in this application, the second instance running on the first instance is a sensitive instance, which has a higher frequency of priority changes and adjustments. The cloud management platform schedules resources on the computing nodes that meet the third priority of the second instance for use by the first instance, matching the priority requirements of the sensitive instance and realizing rapid adjustment of instance resources.

[0020] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform obtains the priority change information of the second instance, which is used to indicate that the priority of the second instance is adjusted from the third priority to the fourth priority, and according to the priority change information of the second instance, schedules resources on the first computing node that meet the fourth priority to provide to the first instance.

[0021] In the solution provided in this application, the tenant can dynamically adjust the priority of the instance by inputting the priority change information of the second instance. Alternatively, the second instance itself can send the priority change information of the second instance to the cloud management platform based on the actual operation of the business. The cloud management platform can then adjust the resources of the instance in the original node to meet the new priority based on the priority change information, so that the first instance can use them. This achieves dynamic adjustment of the corresponding resources on the original computing node to meet the dynamic changes in the priority of the second instance, thereby improving the reliability and stability of the sensitive instance.

[0022] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform obtains resource hotspot information of the first computing node. This resource hotspot information is used to indicate that the QoS indicator on the first computing node is lower than a preset threshold, and to adjust the amount of resources on the first computing node that meet medium and / or low priority.

[0023] In the solution provided in this application, when a QoS indicator degradation event occurs in a compute node, causing the QoS indicators of each instance on the compute node to decrease and fall below a preset threshold, it may affect the applications and services deployed by the tenant on the compute node. After obtaining the relevant hotspot information, the cloud management platform adjusts the amount of resources on the compute node that meet medium and / or low priorities, prioritizing the resources required by high-priority instances.

[0024] In conjunction with the first aspect, in one possible implementation of the first aspect, the cloud management platform obtains resource hotspot information of the first computing node, wherein the resource hotspot information is used to indicate that the quality of service (QoS) index on the first computing node is lower than a preset threshold, the cloud management platform determines a third instance in the first computing node based on the resource hotspot information, and migrates the third instance to the second computing node of the infrastructure, wherein the priority of the third instance is medium priority or low priority.

[0025] In the solution provided in this application, when a QoS indicator degradation event occurs in a compute node, causing the QoS indicators of each instance on the compute node to decrease and fall below a preset threshold, it may affect the applications and services deployed by the tenant on that compute node. After obtaining the relevant hotspot information, the cloud management platform can mitigate the QoS indicator degradation on the current compute node by migrating medium-priority or low-priority instances to other compute nodes, while ensuring the stability of high-priority instances.

[0026] The second aspect or any implementation thereof is a device implementation corresponding to the first aspect or any implementation thereof. Specifically, this application provides a cloud management platform for managing infrastructure, wherein each region of at least one area of ​​the infrastructure is provided with at least one availability zone, and the at least one availability zone is provided with a first availability zone including at least one computing node, wherein at least one instance runs on the first computing node in the at least one computing node. The cloud management platform specifically includes: an acquisition module and a creation module, wherein the acquisition module is used to acquire a first instance creation request input by a tenant, the first instance creation request including deployment information, specification information and priority information of the first instance, wherein the deployment information of the first instance is used to indicate that the first instance is deployed to the first availability zone, the specification information of the first instance is used to indicate the instance specification of the first instance, and the priority information of the first instance is used to indicate that the priority of the first instance is the first priority; and the creation module is used to create the first instance on the first computing node according to the first instance creation request, wherein the priority of the first instance is different from the priority of at least one instance on the first computing node, and the first instance runs on resources in the first computing node that satisfy the first priority.

[0027] In conjunction with the second aspect, in one possible implementation of the second aspect, before creating the first instance based on the instance creation request, the acquisition module is also used to acquire resource load information of multiple computing nodes in the infrastructure. The resource load information is used to indicate the amount of resources used by instances of different priorities deployed on each of the multiple computing nodes.

[0028] In conjunction with the second aspect, in one possible implementation of the second aspect, the cloud management platform further includes a determination module, which is used to determine, based on the deployment information, specification information, priority information and resource load information of the first instance, that there are resources in the first computing node that meet the first priority, while the creation module is specifically used to create the first instance on the first computing node.

[0029] In conjunction with the second aspect, in one possible implementation of the second aspect, the resources provided by each computing node among multiple computing nodes include dedicated resources, shared resources, and super-shared resources. Specifically, when the first priority is high priority, the resources satisfying the first priority are dedicated resources, used to indicate that an instance uses a first number of resources alone; or, when the first priority is medium priority, the resources satisfying the first priority are shared resources, used to indicate that multiple instances share a second number of resources; or, when the first priority is low priority, the resources satisfying the first priority are super-shared resources, used to indicate that multiple instances share a third number of resources, where the third number is less than the second number.

[0030] In conjunction with the second aspect, in one possible implementation of the second aspect, the acquisition module is further used to acquire priority change information of the first instance, wherein the priority change information of the first instance is used to indicate that the priority of the first instance is adjusted from the first priority to the second priority, and the cloud management platform also includes a scheduling module, which is used to schedule resources on the first computing node that meet the second priority for use by the first instance according to the priority change information of the first instance.

[0031] In conjunction with the second aspect, in one possible implementation of the second aspect, the acquisition module is further configured to acquire a second instance creation request input by the tenant, wherein the second instance creation request includes deployment information and priority information of the second instance, the deployment information of the second instance is used to indicate that the second instance is deployed on the first instance, the priority information of the second instance is used to indicate that the priority of the second instance is the third priority, and the creation module is further configured to create the second instance in the first instance according to the second instance creation request.

[0032] In conjunction with the second aspect, in one possible implementation of the second aspect, when the first priority and the third priority are different, the scheduling module is also used to schedule resources on the first computing node that meet the third priority according to the third priority and provide them to the first instance.

[0033] In conjunction with the second aspect, in one possible implementation of the second aspect, the acquisition module is further used to acquire the priority change information of the second instance, wherein the priority change information of the second instance is used to indicate that the priority of the second instance is adjusted from the third priority to the fourth priority, and the scheduling module is further used to schedule resources on the first computing node that meet the fourth priority to provide to the first instance according to the priority change information of the second instance.

[0034] In conjunction with the second aspect, in one possible implementation of the second aspect, the acquisition module is further used to acquire resource hotspot information of the first computing node. The resource hotspot information is used to indicate that the QoS index on the first computing node is lower than a preset threshold. The cloud management platform also includes an adjustment module, which is used to adjust the amount of resources on the first computing node that meet medium and / or low priority.

[0035] In conjunction with the second aspect, in one possible implementation of the second aspect, the acquisition module is further used to acquire resource hotspot information of the first computing node. The resource hotspot information is used to indicate that the QoS indicator on the first computing node is lower than a preset threshold. The determination module is further used to determine a third instance in the first computing node based on the resource hotspot information. The priority of the third instance is medium priority or low priority. The cloud management platform also includes a migration module, which is used to migrate the third instance to the second computing node of the infrastructure.

[0036] Thirdly, this application provides a computing device cluster, including at least one computing device, each computing device including a processor and a memory; the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster performs the method described in the first aspect and any implementation thereof.

[0037] Fourthly, this application provides a computer program product containing instructions that, when executed by a cluster of computer devices, cause the cluster of computer devices to perform the method described in the first aspect and any implementation thereof.

[0038] Fifthly, this application provides a computer-readable storage medium including computer program instructions, which, when executed by a cluster of computing devices, enable the cluster of computing devices to perform the method described in the first aspect and any implementation thereof. Attached Figure Description

[0039] Figure 1 is a schematic diagram of a public cloud application scenario provided in this application;

[0040] Figure 2 is a schematic diagram of an instance management method based on cloud computing technology provided in this application;

[0041] Figure 3 is a flowchart illustrating an instance management method based on cloud computing technology provided in this application;

[0042] Figure 4 is another flowchart illustrating the instance management method based on cloud computing technology provided in this application;

[0043] Figure 5 is another flowchart illustrating the instance management method based on cloud computing technology provided in this application;

[0044] Figure 6 is another architectural diagram of the instance management method based on cloud computing technology provided in this application;

[0045] Figure 7 is a resource diagram of an instance management method based on cloud computing technology provided in this application;

[0046] Figure 8 is a schematic diagram of a cloud management platform provided in this application;

[0047] Figure 9 is a schematic diagram of a computing device provided in this application;

[0048] Figure 10 is a schematic diagram of a computing device cluster provided in this application;

[0049] Figure 11 is a schematic diagram of another structure of the computing device cluster provided in this application;

[0050] Figure 12 is a schematic diagram of another structure of the computing device cluster provided in this application. Detailed Implementation

[0051] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0052] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0053] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0054] First, the terms appearing in this article will be explained as follows:

[0055] Public cloud services, or Infrastructure as a Service (IaaS), refer to providing computing, networking, and storage resources from a public cloud service provider's infrastructure as a service. In this service model, users do not need to build their own physical data center; instead, they use the infrastructure that provides computing, storage, and networking services by renting it. Public cloud services from different tenants are isolated from each other.

[0056] Tenant / User: The user who rents the infrastructure. Tenants can register an account with the public cloud service provider through a browser or other client. The public cloud service provider will record the accounts of different tenants and realize the isolation of public cloud services for different tenants based on the accounts, as described in this application.

[0057] Cloud management platform: A platform provided by public cloud service providers for interacting with users. Users can register an account on the cloud management platform and rent public cloud services using the account, thus becoming a tenant of the public cloud service. The cloud management platform is also used to manage infrastructure and to isolate the computing, network, and / or storage resources rented by different tenants based on their accounts.

[0058] Region: Public cloud service providers set up public cloud data centers in different geographical regions. Public cloud devices in public cloud data centers in different regions need to communicate with each other through a remote connection gateway.

[0059] An Availability Zone (AZ) is a collection of one or more cloud data centers with independent water and electricity supply. Within an Availability Zone, computing, network, and storage resources are logically divided into multiple clusters. Multiple Availability Zones within a single zone are connected via high-speed fiber optic cables to meet users' needs for building highly available systems across AZs.

[0060] An instance refers to a cloud data center host deployed in an availability zone within a region, used to run public cloud services. This instance provides computing, networking, or storage resources and includes, but is not limited to, virtual machines, containers, serverless functions, databases, bare metal servers, object storage buckets, ELB elastic load balancers, cloud gateways, and cloud caches. It is worth noting that instances typically run tenanted applications or services, and tenants may further deploy other instances on them; this application does not limit this.

[0061] Infrastructure: The facilities that support cloud computing services, including at least one data center, each data center including multiple servers, on which computing instances such as virtual machines or containers run to enable elastic cloud computing services. For example, in the case where the infrastructure includes multiple data centers, the multiple data centers can be distributed in different geographical regions and remotely connected to each other through a backbone network.

[0062] To facilitate the explanation of the instance management method based on cloud computing technology in this application, the following explains the relationships between the terms in this specification. In this specification, "user" and "tenant" refer to users who use resources in the infrastructure to deploy instances and run related services. Tenant A is equivalent to the first tenant, and tenant B is equivalent to the second tenant.

[0063] Please refer to Figure 1 below, which is a schematic diagram of a public cloud application scenario provided in this application. As shown in Figure 1, the cloud management platform 20 is used to manage infrastructure 1. Infrastructure 1 includes availability zones set up in multiple regions. Exemplary regions include region 10, region 11, and region 12. Each region contains multiple availability zones. For example, the availability zones in region 10 include availability zones 101, 102, and 103; the availability zones in region 11 include availability zones 111, 112, and 113; and the availability zones in region 12 include availability zones 121, 122, and 123. Each availability zone contains multiple computing nodes. For example, availability zone 101 includes computing nodes 1011 and 1012. Based on this architecture, a public cloud system providing cloud services is formed.

[0064] Please refer to Figure 1. In Figure 1, cloud management platform 20 is used to manage infrastructure 1. Tenant A connects to the Internet through client 21 and logs in to cloud management platform 20 with an account pre-registered with tenant A and bound to it. Tenant B connects to the Internet through client 22 and logs in to cloud management platform 20 with an account pre-registered with tenant B and bound to it. Cloud management platform 20 provides a configuration interface. Tenant A accesses the configuration interface through client 21 and enters configuration information 1 on the configuration interface. Tenant B accesses the configuration interface through client 22 and enters configuration information 2 on the configuration interface. Cloud management platform 20 obtains configuration information 1 from the configuration interface and configures and / or manages tenant A's cloud services according to configuration information 1. Cloud management platform 20 obtains configuration information 2 from the configuration interface and configures and / or manages tenant B's cloud services according to configuration information 2.

[0065] For example, client 21 and client 22 can be terminal devices such as mobile phones with Internet access, personal computers, personal digital assistants, thin clients, vehicle computing nodes, or other terminal devices with Internet access capabilities.

[0066] Please refer to Figure 2 below. Figure 2 is a schematic diagram of an instance management method based on cloud computing technology provided in this application embodiment. Specifically, Figure 2 is the specific architecture of the system shown in Figure 1 under the application scenario of this method. As shown in Figure 2, the infrastructure 1 includes multiple computing nodes, such as computing node 1011... The computing node 1011 includes a hardware layer and a software layer. Instance 10111 is set on the software layer of computing node 1011, and multiple instances can run simultaneously on computing node 1011, such as instance 10112... The hardware layer of computing node 1011 includes memory 10114, processor 10115, network card 10116, and hard disk 10117. Instance 10111 is connected to the Internet through network card 10116, and the instance management platform 20 manages instances 10111 and instance 10112 running on computing node 1011 through network card 10116. Typically, instances within the same compute node run with the same priority. For example, instances 10112 and 10113 both have low priorities. Services with different priorities are isolated through independent compute node resource pools or different compute nodes.

[0067] As agile services continue to develop, tenants have further increased their demands for instance priority, especially the ability to dynamically adjust instance priority. As mentioned above, the capabilities provided by the current infrastructure and cloud management platform cannot keep up with the rapid development of agile services. In use cases that require flexible support for instance priority changes, the ability to guarantee differentiated priorities and the ability to dynamically schedule instances with adaptive priority adjustment are significantly insufficient.

[0068] To address the aforementioned issues, this application provides an instance management method and cloud management platform based on cloud computing technology. The cloud management platform configures instances with different priorities based on the tenant's desired deployment priorities. This allows for the allocation of instances with varying priorities across compute nodes that meet the tenant's priority requirements, accommodating the tenant's rapidly changing business needs. Furthermore, it adjusts resources in the infrastructure promptly to meet the corresponding instance priorities based on the actual priority requirements of the applications running on the instances or user-inputted priority adjustments. Additionally, if resource hotspots are detected on compute nodes, the platform adjusts resource allocation under different priorities or relocates relevant instances to other compute nodes, thereby improving the resource utilization of compute nodes while ensuring the QoS indicators of tenant instances and their operation.

[0069] Based on the application scenarios and architectures described in Figures 1 and 2, the following provides a detailed introduction to the instance management method based on cloud computing technology provided in this application.

[0070] Please refer to Figure 3, which is a flowchart of an instance management method based on cloud computing technology provided in this application. As shown in Figure 3, the method includes, but is not limited to, the following steps:

[0071] S301, obtain the first instance creation request input by the tenant. The first instance creation request includes the deployment information, specification information and priority information of the first instance.

[0072] Tenant A requests the cloud management platform 20 to create a corresponding instance through the interface provided by the cloud management platform 20 or by inputting a first instance creation request through the client 21. The first instance creation request includes the deployment information, specification information, and priority information of the first instance.

[0073] Specifically, this will be further explained with reference to computing node 1011 and instance 10111 in Figure 2:

[0074] In one example of this application, the cloud management platform 20 provides a cloud service interface to tenant A through client 21. Tenant A can select or enter a creation request for instance 10111 through this cloud service interface, such as entering deployment information for instance 10111 as availability zone 101 or directly as compute node 1011, or selecting the corresponding identifier for availability zone 101 or compute node 1011. Further, tenant A enters the instance specifications of instance 10111, such as a central processing unit (CPU) specification including 2 cores, a memory specification including 4GB, and / or a hard disk specification including 70GB. Furthermore, tenant A enters the priority information of instance 10111, such as high priority, medium priority, or low priority; this application does not limit this.

[0075] In another example, tenant A can use the resource orchestration template provided by cloud management platform 20 to call Infrastructure as a Code (IaaC) to perform orchestration management of instances, such as instance creation, instance modification, and instance deletion. In this application, tenant A can input a first instance creation request through the resource orchestration template and call the resource orchestration engine of cloud management platform 20 to create the first instance in infrastructure 1.

[0076] In another example, the cloud management platform 20 provides a cloud service interface to tenant A through client 21. Tenant A can use this interface to directly state their actual needs for the target business. The artificial intelligence (AI) model provided by the cloud management platform 20, based on tenant A's input of the target business needs, such as latency requirements, QoS metrics, and business type, recommends corresponding instances and instance information to tenant A. For example, the deployment information for instance 10111 may be Availability Zone 101 or directly compute node 1011; the instance specifications of instance 10111 may include 2 CPU cores, 4GB of memory, and 70GB of hard disk space; and the priority information of instance 10111 may be high priority. After tenant A confirms that the instance information is correct, this instance information becomes an instance creation request, and the cloud management platform 20 proceeds to the next step based on this first instance creation request.

[0077] S302, Based on the first instance creation request, create a first instance on the first compute node, wherein the priority of the first instance is different from the priority of at least one instance, and the first instance runs on a resource in the first compute node that satisfies the first priority.

[0078] Further explanation will be provided in conjunction with computing node 1011 and instance 10111 in Figure 2. For ease of explanation, for example, the first computing node is 1011 and the first instance is instance 10111.

[0079] Based on the first instance creation request input by tenant A, cloud management platform 20 creates instance 10111 with a CPU specification of 2 cores, a memory specification of 4GB and / or a hard disk specification of 70GB in compute node 1011 in availability zone 101 of infrastructure 1. Instance 10111 has a high priority, while instances 10112 and 10113 in compute node 1011 have low priorities. The priority of instance 10111 is different from that of instances 10112 and 10113.

[0080] Furthermore, referring to Figure 4, which is another flowchart illustrating the instance management method based on cloud computing technology provided in this application, the specific details are as follows:

[0081] The steps in S401 are as described in S301, and will not be repeated here.

[0082] S402. Based on the creation request and resource load information of the first instance, determine that there are resources in the first computing node that meet the first priority.

[0083] Before creating instance 10111 on compute node 1011, cloud management platform 20 determines, based on the creation request and resource load information of the first instance, that there are resources in compute node 1011 that meet the priority of instance 10111.

[0084] Prior to this, the cloud management platform 20 first obtains resource load information of multiple computing nodes in infrastructure 1. This resource load information is used to represent the usage and remaining amount of resources that meet different priorities on each computing node in infrastructure 1. For details, please refer to Figure 7. Figure 7 is a resource diagram of an instance management method based on cloud computing technology provided in this application. As shown in Figure 7:

[0085] For example, the resources on computing node 1011 can take the CPU core as an example. These resources can also be storage resources on the computing node, such as memory and hard disk, or network resources, such as network bandwidth and the number of connections. This application does not limit these resources.

[0086] On compute node 1011, high-priority, medium-priority, and low-priority instances can be deployed. High-priority instances use dedicated CPU cores, meaning each instance uses a first set of CPU cores to run its application or service. This first set of CPU cores can be considered a high-priority resource pool, providing services to high-priority instances. Medium-priority instances use shared CPU and core resources, meaning multiple medium-priority instances share a second set of CPU cores to run their application or service. This second set of CPU cores can be considered a medium-priority resource pool, providing services to medium-priority instances. Low-priority instances use shared CPU and core resources, meaning multiple low-priority instances share a third set of CPU cores to run their application or service. This third set of CPU cores can be considered a low-priority resource pool, providing services to low-priority instances. The third set of CPU cores in the low-priority resource pool can be less than or equal to the second set of CPU cores, but the number of low-priority instances must be greater than the number of medium-priority instances.

[0087] For example, the resource pools corresponding to the different priorities mentioned above can achieve the binding relationship between resources and instances by binding cores, binding memory, etc., so that when the corresponding applications and services are running on the instance, the corresponding resources provide resources to the instance.

[0088] In some embodiments, the cloud management platform 20 can obtain the status information of each computing node in multiple regions using push and pull modes. For example, for static data such as QoS indicator information, query requests can be sent to each computing node in each region to proactively obtain resource information and / or QoS indicator information for each region and each computing node. For dynamically changing data such as resource capacity information and specification information, each region can proactively push its own cloud resource capacity information and / or specification information to the cloud management platform 20 when cloud resources change, forming a resource profile.

[0089] For example, compute node 1011 or infrastructure 1 is equipped with a cloud resource query plugin and a QoS indicator query plugin. The cloud resource query plugin is used to receive capacity and / or specification information of cloud resources from each compute node. The cloud resource query plugin can also determine the utilization rate and hot resources of cloud resources in each region and each compute node based on the capacity and / or specification information of cloud resources that meet different priorities. The QoS indicator query plugin is used by each compute node to receive query requests sent by the cloud management platform 20 to query the QoS indicator information of the region.

[0090] In another example, the cloud management platform 20 periodically obtains the resource usage and remaining resources on each compute node in infrastructure 1 based on instances of different priorities, and generates a resource profile. This process can occur after receiving the first instance creation request from tenant A, before receiving the first instance creation request from tenant A, or before creating the first instance on the first compute node. For example, after receiving the first instance creation request from tenant A, the platform obtains a resource profile of the availability zone or compute node corresponding to the deployment information of the first instance.

[0091] Based on this, the cloud management platform 20 determines, according to the creation request and resource load information of instance 10111, that there are resources in compute node 1011 that meet high priority requirements. Specifically:

[0092] Based on the creation request of instance 10111, cloud management platform 20 determines that there are resources in the high-priority resource pool of compute node 1011 in availability zone 101 of infrastructure 1 that meet the specifications of 2 CPU cores, 4GB memory and / or 70GB hard disk.

[0093] S403. Create an instance on the first compute node.

[0094] The cloud management platform 20 sends an instance creation instruction to the compute node 1011. This instruction instructs the compute node 1011 to create an instance 10111 with 2 CPU cores, 4GB of memory, and / or 70GB of hard disk space, using resources from the high-priority resource pool. The compute node 1011 also deploys instances 10111, 10112, and 10113, where instance 10111 has a high priority, while instances 10112 and 10113 have low priorities. Therefore, the resources used by instance 10111 are different from those used by instances 10112 and 10113.

[0095] Based on this, when the priority information of instance 10111 changes, or when the priority of instance 10111 needs to be dynamically adjusted due to business needs, only the corresponding resources need to be adjusted in compute node 1011, without having to fetch the corresponding data across compute nodes, reducing the fluctuation of cloud services, improving the reliability of instances, and further improving resource utilization.

[0096] S404. Obtain the priority change information of the first instance entered by the tenant.

[0097] The cloud management platform 20 obtains the priority change information of instance 10111 input by tenant A. This priority change information indicates that tenant A has changed the priority of instance 10111 from high priority to medium priority.

[0098] In another example of this application, when instance 10111 is running an application deployed on that instance by tenant A, due to business changes, the application sends priority change information to instance 10111, instructing instance 10111 to change its priority from high priority to medium priority, so as to meet the actual needs of the application and business.

[0099] S405. Based on the priority change information of the first instance, schedule resources on the computing node that meet the second priority for use by the instance.

[0100] Based on the priority change information of instance 10111 obtained, the cloud management platform 20 schedules resources on computing node 1011 that meet the medium priority requirements for use by instance 10111.

[0101] For example, instance 10111 is scheduled from high priority to medium priority, and its corresponding resources are also scheduled from exclusive resources in the high priority resource pool to shared resources in the medium priority resource pool. Taking CPU core resources as an example, the cloud management platform 20 adjusts the CPU binding relationship of instance 10111 from being individually bound to 2 CPU cores to being jointly bound to 2 CPU cores with at least one other instance, or binding to a certain range of CPU cores. According to the actual business of different instances, the CPU core quota is dynamically scheduled to meet the specification requirements of 2 CPU cores, and the CPU core resources that were originally individually bound are released, or the released CPU core resources are allocated to the medium priority resource pool to improve the utilization of idle resources.

[0102] In another example of this application, when instance 10111 is scheduled from medium priority to high priority, its corresponding resources are also scheduled from shared resources in the medium priority resource pool to dedicated resources in the high priority resource pool. Before scheduling by the cloud management platform 20, the resource load information in compute node 1011 is obtained. Based on the resource load information, it is determined that the resources in the high priority resource pool of compute node 1011 meet the resource requirements of instance 10111. For example, if the resource is a 2-CPU core resource, and there are 2 idle CPU core resources in the high priority resource pool, then the 2 idle CPU core resources are scheduled and bound to instance 10111, and its CPU core binding relationship in the medium priority resource pool is released.

[0103] If there are no two idle CPU cores in the high-priority resource pool, the resource load information of other compute nodes on availability zone 101 is obtained to determine the compute node that meets the priority requirement. For example, if there are three idle CPU cores in the high-priority resource pool of compute node 1012, instance 10111 is hot-migrated to compute node 1012, and compute node 1012 provides the corresponding high-priority resources.

[0104] In another example of this application, when instance 10111 is deleted, the cloud management platform 20 releases the resources bound to instance 10111.

[0105] Therefore, when the priority of an instance is dynamically adjusted or changed, the cloud management platform can provide highly reliable priority changes. Whether it is dynamic scheduling based on different priorities within the same compute node or hot migration and adjustment of resources between different compute nodes, it can ensure high availability and high reliability of the instance, as well as a relatively high level of resource utilization.

[0106] S406. Obtain the second instance creation request input by the tenant.

[0107] Specifically, further explanation is given in conjunction with Figure 6, which is another architectural diagram of the instance management method based on cloud computing technology provided in this application. For ease of explanation, for example, the first computing node is computing node 1012, the first instance is instance 10121, and the second instance is 10123, which does not impose any additional limitations on this application.

[0108] The cloud management platform 20 obtains the creation request for instance 10123 input by tenant A. The creation request for instance 10123 includes deployment information and priority information. The deployment information of instance 10123 indicates instance 10121, and the priority information of instance 10123 indicates that the priority of instance 10123 is high priority.

[0109] It is worth noting that instance 10121 may have been created before the creation request for instance 10123 input by tenant A was obtained, or it may have been created together with the creation request for instance 10123. Tenant A may input the creation request for instance 10121 at the same time as inputting the creation request for instance 10123, or may not input it and let the cloud management platform 20 automatically create instance 10121 according to the creation request for instance 10123. This application does not limit this.

[0110] S407. Create a second instance on the first instance on the first compute node.

[0111] The cloud management platform 20 creates instance 10123 on instance 10121. For example, instance 10121 can be a virtual machine instance, and instance 10123 can be a container instance. Container instances are agile instances, and deployment and change operations are more frequent. In order to meet the dynamic adjustment needs of agile services, this application supports the deployment of different types of instances with different priorities in the same computing node, and provides resources that can meet the corresponding priorities.

[0112] S408. Determine whether the first priority and the third priority are the same.

[0113] The cloud management platform 20 determines whether the priorities of instance 10123 and instance 10121 are the same based on their priorities.

[0114] S409. When the first priority and the third priority are different, the resources on the first computing node that meet the third priority are scheduled to be provided to the first instance.

[0115] When the priorities of instance 10121 and instance 10123 are different, such as instance 10123 having a high priority and instance 10121 having a medium priority, the cloud management platform 20 schedules resources on compute node 1012 that meet the high priority requirement to provide to instance 10121. For example, if instance 10121 has a medium priority, it is bound to CPU core resources in the medium priority resource pool on compute node 1012. When instance 10123 is created, if instance 10123 has a high priority, then CPU core resources in the high priority resource pool on compute node 1012 are scheduled and bound to instance 10123 for its exclusive use, thereby prioritizing the priority of agile services.

[0116] S410. Obtain the priority change information of the second instance input by the tenant.

[0117] The cloud management platform 20 obtains the priority change information of instance 10123 input by tenant A. This priority change information indicates that tenant A has changed the priority of instance 10123 from high priority to medium priority.

[0118] In another example of this application, when instance 10123 is running an application deployed on that instance by tenant A, due to business changes, the application sends priority change information to instance 10123, instructing instance 10123 to change its priority from high priority to medium priority to meet the actual needs of the application and business. Alternatively, the cloud management platform 20 monitors each computing node in infrastructure 1, and switches the corresponding resource orchestration strategy when a priority change occurs.

[0119] S411. Schedule resources on the first compute node that meet the fourth priority to provide to the first instance.

[0120] Based on the priority change information of instance 10123 obtained, the cloud management platform 20 schedules resources on computing node 101 that meet the medium priority for use by instance 10121.

[0121] For example, if instance 10123 is scheduled from high priority to medium priority and deployed on instance 10121, then the resources corresponding to instance 10121 will also be scheduled from exclusive resources in the high-priority resource pool to shared resources in the medium-priority resource pool. Taking CPU core resources as an example, the cloud management platform 20 adjusts the CPU binding relationship of instance 10121 from being individually bound to 2 CPU cores to being jointly bound to 2 CPU cores with at least one other instance, or binding to a certain range of CPU cores. Based on the actual business of different instances, the platform dynamically schedules the CPU core quota to meet the specification requirements of 2 CPU cores, and releases the originally individually bound CPU core resources, or allocates the released CPU core resources to the medium-priority resource pool to improve the utilization of idle resources.

[0122] In another example of this application, when instance 10123 is scheduled from medium priority to high priority and deployed on instance 10121, the resources corresponding to instance 10121 are also scheduled from shared resources in the medium priority resource pool to dedicated resources in the high priority resource pool. Before scheduling by the cloud management platform 20, the resource load information in compute node 1012 is obtained. Based on the resource load information, it is determined that the resources in the high priority resource pool of compute node 1012 meet the resource requirements of instance 10121. For example, if the resource is a 2-CPU core resource, and there are 2 idle CPU core resources in the high priority resource pool, then the 2 idle CPU core resources are scheduled and bound to instance 10121, and their CPU core binding relationship in the medium priority resource pool is released.

[0123] If there are no two idle CPU cores in the high-priority resource pool, the resource load information of other compute nodes on availability zone 101 is obtained to determine the compute node that meets the priority requirement. For example, if there are three idle CPU cores in the high-priority resource pool of compute node 1013, instances 10121 and 10123 are hot-migrated to compute node 1013, and compute node 1013 provides the corresponding high-priority resources.

[0124] Therefore, when the priority of agile instances deployed on an instance is dynamically adjusted or changed, the cloud management platform can provide highly reliable priority changes. Whether it is dynamic scheduling based on different priorities within the same compute node or hot migration adjustment of resources between different compute nodes, it can ensure high availability and high reliability of the instance, as well as a relatively high level of resource utilization.

[0125] Furthermore, please refer to Figure 5. As shown in Figure 5, this is another flowchart illustrating the instance management method based on cloud computing technology provided in this application. Specifically:

[0126] S501. Obtain the resource hotspot information of the first computing node.

[0127] Referring to Figure 5, for ease of explanation, the first computation node is computation node 1011. Specifically:

[0128] The cloud management platform 20 obtains resource hotspot information of computing node 1011, wherein the resource hotspot information is used to indicate that the quality of service (QoS) index on computing node 1011 is lower than a preset threshold.

[0129] In some embodiments, the cloud management platform 20 can use push and pull modes to obtain resource hotspot information of computing nodes in multiple regions. For example, for static data such as QoS indicator information, query requests can be sent to each computing node in each region to actively obtain QoS indicator information for each region and each computing node.

[0130] In another example of this application, a QoS indicator query plugin is provided in compute node 1011 or infrastructure 1. The QoS indicator query plugin is used by each compute node to receive query requests sent by the cloud management platform 20 to query the QoS indicator information of a region or compute node.

[0131] When resource contention occurs on compute node 1011 at a certain priority level, causing the QoS index to fall below the preset threshold, the cloud management platform 20 will use the resource hotspot information to determine whether the resource contention is occurring in the medium-priority resource pool or the low-priority resource pool of compute node 1011.

[0132] For example, when resource contention occurs, the cloud management platform 20 dynamically adjusts the resource allocation quotas in the medium-priority and low-priority resource pools based on the overall resource utilization and QoS indicators of the current medium-priority and low-priority resource pools in computing node 1011. If resource contention occurs in the low-priority resource pool, but the QoS indicators of the resources in the medium-priority resource pool are good, then some resources in the medium-priority resource pool are transferred to the low-priority resource pool.

[0133] In another example of this application, when resource contention occurs, the cloud management platform 20 dynamically allocates resource quotas in the medium-priority and low-priority resource pools based on the overall resource utilization and QoS indicators of the current high-priority, medium-priority, and low-priority resource pools in the computing node 1011. If resource contention occurs in both the medium-priority and low-priority resource pools, but there are idle resources in the high-priority resource pool, the idle resources in the high-priority resource pool are adjusted to the medium-priority and low-priority resource pools according to a certain strategy.

[0134] Therefore, when resource hotspot information occurs, the resources on the computing node are dynamically adjusted to further ensure high reliability and high quality of service and reduce the impact on user services.

[0135] S502. Send migration instructions to the first computing node.

[0136] In another example of this application, after resource contention occurs, the cloud management platform 20 sends a migration instruction to the compute node 1011. This instruction instructs the migration of a specific hot instance, such as instance 10113, to another compute node. Specifically, the cloud management platform 20 determines that the resources on compute node 1012 can meet the needs of instance 10113 based on the amount of resources required by the hot instance and the resource amounts obtained from other compute nodes under different priorities in availability zone 101.

[0137] S503. Migrate the third instance to the second computing node of the infrastructure.

[0138] The cloud management platform migrated instance 10113 to compute node 1012. Instance 10113 is a hot instance, which is an instance with greater resource demand or more deteriorated QoS indicators compared to other instances. By hot migrating it to other compute nodes, the QoS indicators of all instances on the original compute nodes are maintained, hotspots are eliminated, and instance 10113 can also obtain sufficient resources on the new compute nodes.

[0139] This application also provides a cloud management platform, as shown in Figure 8, which is a structural schematic diagram of a cloud management platform provided in this application. Details are as follows:

[0140] The acquisition module 801 is used to acquire the first instance creation request input by the tenant. The first instance creation request includes the deployment information, specification information and priority information of the first instance. The deployment information of the first instance is used to indicate that the first instance is deployed to the first availability zone, the specification information of the first instance is used to indicate the instance specification of the first instance, and the priority information of the first instance is used to indicate that the priority of the first instance is the first priority.

[0141] Creation module 802 is used to create a first instance on a first compute node according to a first instance creation request, wherein the priority of the first instance is different from the priority of at least one instance, and the first instance runs on a resource in the first compute node that satisfies the first priority.

[0142] The determination module 803 is used to determine, based on the deployment information, specification information, priority information and resource load information of the first instance, that there are resources in the first computing node that meet the first priority.

[0143] The scheduling module 804 is used to schedule resources on the first computing node that meet the second priority according to the priority change information of the first instance for the first instance to use.

[0144] Adjustment module 805 is used to adjust the amount of resources on the first computing node that meet medium and / or low priority requirements.

[0145] Migration module 806 is used to migrate a third instance to a second compute node of the infrastructure.

[0146] The acquisition module 801, creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806 can all be implemented in software or hardware. For example, the implementation of the acquisition module 801 will be described below. Similarly, the implementation of the creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806 can refer to the implementation of the acquisition module 801.

[0147] As an example of a software functional unit, module 801 may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Further, the aforementioned computing instance may be one or more. For example, module 801 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region or in different regions. Further, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone or in different availability zones, each availability zone including one data center or multiple geographically proximate data centers. Typically, a region may include multiple availability zones.

[0148] As an example of a hardware functional unit, the acquisition module 801 may include at least one computing device, such as a server. Alternatively, the acquisition module 801 may also be a device implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be implemented using a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), generic array logic (GAL), or any combination thereof.

[0149] It should be noted that, in other embodiments, the acquisition module 801 can be used to execute any step in the instance management method based on cloud computing technology, the creation module 802 can be used to execute any step in the instance management method based on cloud computing technology, the determination module 803 can be used to execute any step in the instance management method based on cloud computing technology, the scheduling module 804 can be used to execute any step in the instance management method based on cloud computing technology, the adjustment module 805 can be used to execute any step in the instance management method based on cloud computing technology, and the migration module 806 can be used to execute any step in the instance management method based on cloud computing technology. The steps implemented by the acquisition module 801, creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806 can be specified as needed. By implementing different steps in the instance management method based on cloud computing technology through the acquisition module 801, creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806, all functions of the cloud management platform can be realized.

[0150] The cloud management platform and apparatus of the embodiments of this application have been described in detail above. In order to facilitate better implementation of the above solutions of the embodiments of this application, relevant equipment for cooperating in the implementation of the above solutions is also provided below.

[0151] This application provides a computing device. Please refer to Figure 9 below. Figure 9 is a schematic diagram of the structure of a computing device according to a cloud computing-based instance management method provided in this application. The computing device 800 includes: a bus 807, a processor 808, a memory 810, and a communication interface 809. The processor 808, the memory 810, and the communication interface 809 communicate with each other via the bus 807. The computing device 800 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in the computing device 800.

[0152] Bus 807 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, only one line is used in Figure 9, but this does not imply that there is only one bus or one type of bus. Bus 807 can include pathways for transmitting information between various components of computing device 800 (e.g., memory 810, processor 808, communication interface 809).

[0153] The processor 808 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0154] The memory 810 may include volatile memory, such as random access memory (RAM). The processor 808 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0155] The memory 810 stores executable program code, which the processor 808 executes to implement the functions of the acquisition module 801, creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806, thereby realizing an instance management method based on cloud computing technology. In other words, the memory 810 stores instructions from the cloud management platform for executing the instance management method based on cloud computing technology.

[0156] The communication interface 809 uses transceiver modules, such as, but not limited to, network interface cards and transceivers, to enable communication between the computing device 800 and other devices or communication networks.

[0157] This application also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0158] Please refer to Figure 10 below. Figure 10 is a schematic diagram of a computing device cluster according to an embodiment of the instance management method based on cloud computing technology of this application. As shown in Figure 10, the computing device cluster includes at least one computing device 800. The memory 810 of one or more computing devices 800 in the computing device cluster may store the same cloud management platform for executing the instance management method based on cloud computing technology.

[0159] In some possible implementations, one or more computing devices 800 in the computing device cluster can also be used to execute some of the instructions of the cloud management platform for performing instance management methods based on cloud computing technology. In other words, a combination of one or more computing devices 800 can jointly execute the instructions of the cloud management platform for performing instance management methods based on cloud computing technology.

[0160] It should be noted that the memory 810 in different computing devices 800 within the computing device cluster can store different instructions for executing some functions of the cloud management platform. That is, the instructions stored in the memory 810 of different computing devices 800 can implement the functions of one or more modules among the acquisition module 801, creation module 802, determination module 803, scheduling module 804, adjustment module 805, and migration module 806.

[0161] Please refer to Figure 11 below. Figure 11 is a schematic diagram of another structure of a computing device cluster based on the instance management method of cloud computing technology according to an embodiment of this application. As shown in Figure 11, two computing devices 800A and 800B are connected through a communication interface 809. The memory in computing device 800A stores instructions for executing the acquisition module 801, the creation module 802, and the determination module 803. The memory in computing device 800B stores instructions for executing the functions of the scheduling module 804, the adjustment module 805, and the migration module 806. In other words, the memory 810 of computing devices 800A and 800B jointly stores the instructions used by the cloud management platform to execute the instance management method based on cloud computing technology.

[0162] The connection method between the computing device clusters shown in Figure 11 can be based on the fact that the instance management method based on cloud computing technology provided in this application requires a large amount of data calculation for the acquisition module 801, creation module 802, and determination module 803. Considering the computational load of the scheduling module 804, adjustment module 805, and migration module 806, in order to avoid overloading the computing device 800A, the functions implemented by the scheduling module 804, adjustment module 805, and migration module 806 are delegated to the computing device 800B for execution.

[0163] It should be understood that the functions of computing device 800A shown in Figure 11 can also be performed by multiple computing devices 800. Similarly, the functions of computing device 800B can also be performed by multiple computing devices 800.

[0164] Please refer to Figure 12 below. Figure 12 is a schematic diagram of another structure of a computing device cluster in the instance management method based on cloud computing technology according to an embodiment of this application. In some possible implementations, one or more computing devices in the computing device cluster can be connected via a network. The network can be a wide area network (WAN) or a local area network (LAN), etc. Figure 12 shows one possible implementation, where two computing devices 800C and 800D are connected via a network. Specifically, they are connected to the network through the communication interface in each computing device. In this type of possible implementation, the memory 810 in computing device 800C stores instructions for executing the acquisition module 801, the creation module 802, and the determination module 803. Simultaneously, the memory 810 in computing device 800D stores instructions for executing the functions of the scheduling module 804, the adjustment module 805, and the migration module 806.

[0165] The connection method between the computing device clusters shown in Figure 12 can be considered as follows: the instance management method based on cloud computing technology provided in this application requires a large amount of data transmission and needs to be connected through the network. The execution of these functions is relatively independent. In order to achieve the best storage and computing performance, the data computing functions implemented by the scheduling module 804, the adjustment module 805, and the migration module 806 are considered to be executed by the computing device 800D.

[0166] It should be understood that the functions of computing device 800C shown in Figure 12 can also be performed by multiple computing devices 800. Similarly, the functions of computing device 800D can also be performed by multiple computing devices 800.

[0167] In some possible implementations, the memory 810 of one or more computing devices 800 in the computing device cluster may also store partial instructions for executing instance management methods based on cloud computing technology. In other words, a combination of one or more computing devices 800 can jointly execute instructions for executing instance management methods based on cloud computing technology.

[0168] This application also provides a computer program product containing instructions. The computer program product may be a software or program product containing instructions, capable of running on a computing device or stored on any usable medium. When the computer program product is run on at least one computer device, it causes the at least one computer device to execute the above-described instance management method applied to a cloud management platform for performing cloud computing technology.

[0169] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the above-described instance management method applied to a cloud management platform for performing cloud computing technology.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

[0171] Those skilled in the art will clearly understand that the specific working process of the system, device or unit described above can be referred to the corresponding process in the aforementioned cloud management platform embodiments, and will not be repeated here.

[0172] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An instance management method based on cloud computing technology, characterized in that, The method is applied to a cloud management platform for managing infrastructure, the infrastructure including at least one region, each of the at least one region including at least one availability zone, a first availability zone including at least one compute node, and the first compute node having at least one instance running on it, the method comprising: Obtain the first instance creation request input by the tenant. The first instance creation request includes the deployment information, specification information and priority information of the first instance. The deployment information of the first instance is used to indicate that the first instance is deployed to the first availability zone. The specification information of the first instance is used to indicate the instance specification of the first instance. The priority information of the first instance is used to indicate that the priority of the first instance is the first priority. Based on the first instance creation request, the first instance is created on the first compute node, wherein the priority of the first instance is different from the priority of the at least one instance, and the first instance runs on resources in the first compute node that satisfy the first priority.

2. The method according to claim 1, characterized in that, Before creating the first instance based on the instance creation request, the method further includes: Obtain resource load information of multiple computing nodes in the infrastructure, wherein the resource load information is used to indicate the amount of resources used by instances of different priorities deployed on each of the multiple computing nodes.

3. The method according to claim 2, characterized in that, The step of creating the first instance according to the instance creation request includes: Based on the deployment information of the first instance, the specification information of the first instance, the priority information of the first instance, and the resource load information, it is determined that there are resources in the first computing node that meet the first priority. Create the first instance on the first compute node.

4. The method according to any one of claims 1 to 3, characterized in that, The resources provided by each of the multiple computing nodes include dedicated resources, shared resources, and excess resources. When the first priority is high priority, the resources that satisfy the first priority are dedicated resources, and the dedicated resources are used to indicate that the instance uses a first number of resources exclusively; or, When the first priority is medium priority, the resources that satisfy the first priority are shared resources, and the shared resources are used to indicate that multiple instances share and use a second number of resources; or, When the first priority is low priority, the resources that satisfy the first priority are super-shared resources. The super-shared resources are used to indicate that the multiple instances share a third number of resources, wherein the third number is less than the second number.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Obtain the priority change information of the first instance, which is used to indicate that the priority of the first instance is adjusted from the first priority to the second priority; Based on the priority change information of the first instance, resources on the first computing node that meet the second priority are scheduled for use by the first instance.

6. The method according to any one of claims 1 and 5, characterized in that, The method further includes: Obtain the second instance creation request input by the tenant. The second instance creation request includes deployment information and priority information of the second instance. The deployment information of the second instance is used to indicate that the second instance is deployed on the first instance. The priority information of the second instance is used to indicate that the priority of the second instance is the third priority. Based on the second instance creation request, create the second instance within the first instance.

7. The method according to claim 6, characterized in that, When the first priority and the third priority are different, the method further includes: Based on the third priority, resources on the first computing node that meet the third priority are scheduled and provided to the first instance.

8. The method according to claim 6 or 7, characterized in that, The method further includes: Obtain the priority change information of the second instance, which is used to indicate that the priority of the second instance is adjusted from the third priority to the fourth priority; Based on the priority change information of the second instance, resources on the first computing node that meet the fourth priority are scheduled and provided to the first instance.

9. The method according to any one of claims 1 to 8, characterized in that, The method further includes: Obtain resource hotspot information of the first computing node, wherein the resource hotspot information is used to indicate that the Quality of Service (QoS) index on the first computing node is lower than a preset threshold; Adjust the amount of resources on the first computing node that meet the medium priority and / or low priority requirements.

10. The method according to any one of claims 1 to 8, characterized in that, The method further includes: Obtain resource hotspot information of the first computing node, wherein the resource hotspot information is used to indicate that the Quality of Service (QoS) index on the first computing node is lower than a preset threshold; Based on the resource hotspot information, a third instance in the first computing node is determined, and the priority of the third instance is medium priority or low priority. Migrate the third instance to the second computing node of the infrastructure.

11. A cloud management platform for managing infrastructure, the infrastructure including at least one region, each of the at least one region including at least one availability zone, a first availability zone including at least one compute node, and the first compute node having at least one instance running on it, the cloud management platform comprising: The acquisition module is used to acquire a first instance creation request input by a tenant. The first instance creation request includes deployment information, specification information and priority information of the first instance. The deployment information of the first instance is used to indicate that the first instance is deployed to the first availability zone. The specification information of the first instance is used to indicate the instance specification of the first instance. The priority information of the first instance is used to indicate that the priority of the first instance is the first priority. A creation module is configured to create the first instance on the first computing node according to the first instance creation request, wherein the priority of the first instance is different from the priority of the at least one instance, and the first instance runs on resources in the first computing node that satisfy the first priority.

12. The cloud management platform according to claim 11, characterized in that, Before creating the first instance based on the instance creation request, the acquisition module is further configured to: Obtain resource load information of multiple computing nodes in the infrastructure, wherein the resource load information is used to indicate the amount of resources used by instances of different priorities deployed on each of the multiple computing nodes.

13. The cloud management platform according to claim 12, characterized in that, The cloud management platform also includes: The determining module is configured to determine, based on the deployment information, specification information, priority information, and resource load information of the first instance, whether there are resources in the first computing node that satisfy the first priority; then, The creation module is used to create the first instance on the first computing node.

14. The cloud management platform according to any one of claims 11 to 13, characterized in that, The resources provided by each of the multiple computing nodes include dedicated resources, shared resources, and excess resources. When the first priority is high priority, the resources that satisfy the first priority are dedicated resources, and the dedicated resources are used to indicate that the instance uses a first number of resources exclusively; or, When the first priority is medium priority, the resources that satisfy the first priority are shared resources, and the shared resources are used to indicate that multiple instances share and use a second number of resources; or, When the first priority is low priority, the resources that satisfy the first priority are super-shared resources. The super-shared resources are used to indicate that the multiple instances share a third number of resources, wherein the third number is less than the second number.

15. The cloud management platform according to any one of claims 11 to 14, characterized in that, The acquisition module is further configured to acquire priority change information of the first instance, wherein the priority change information of the first instance is used to indicate that the priority of the first instance is adjusted from the first priority to the second priority; therefore, the cloud management platform further includes: The scheduling module is used to schedule resources on the first computing node that meet the second priority for use by the first instance based on the priority change information of the first instance.

16. The cloud management platform according to any one of claims 11 and 15, characterized in that, The acquisition module is further configured to acquire a second instance creation request input by the tenant. The second instance creation request includes deployment information and priority information for the second instance. The deployment information indicates that the second instance should be deployed onto the first instance, and the priority information indicates that the second instance has a priority of third priority. The creation module is also used to create the second instance in the first instance based on the second instance creation request.

17. The cloud management platform according to claim 16, characterized in that, When the first priority and the third priority are different, the scheduling module is further configured to schedule resources on the first computing node that meet the third priority to provide to the first instance according to the third priority.

18. The cloud management platform according to claim 16 or 17, characterized in that, The acquisition module is further configured to acquire priority change information of the second instance, wherein the priority change information of the second instance is used to indicate that the priority of the second instance is adjusted from the third priority to the fourth priority; then, The scheduling module is also used to schedule resources on the first computing node that meet the fourth priority to be provided to the first instance based on the priority change information of the second instance.

19. The cloud management platform according to any one of claims 11 to 18, characterized in that, The acquisition module is further configured to acquire resource hotspot information of the first computing node, the resource hotspot information being used to indicate that the Quality of Service (QoS) index on the first computing node is lower than a preset threshold; therefore, the cloud management platform further includes: An adjustment module is used to adjust the amount of resources on the first computing node that meet the medium priority and / or low priority.

20. The cloud management platform according to any one of claims 11 to 18, characterized in that, The acquisition module is also used to acquire resource hotspot information of the first computing node, and the resource hotspot information is used to indicate that the quality of service (QoS) index on the first computing node is lower than a preset threshold. The determining module is further configured to determine a third instance in the first computing node based on the resource hotspot information, wherein the priority of the third instance is medium priority or low priority. Therefore, the cloud management platform also includes: A migration module is used to migrate the third instance to a second computing node of the infrastructure.

21. A computing device cluster, characterized in that, It includes at least one computing device, each computing device including a processor and memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method as described in any one of claims 1 to 10.

22. A computer program product containing instructions, characterized in that, When the instructions are executed by a cluster of computer devices, the cluster of computer devices causes the cluster of computer devices to perform the method as described in any one of claims 1 to 10.

23. A computer-readable storage medium, characterized in that, Includes computer program instructions, which, when executed by a cluster of computing devices, perform the method as described in any one of claims 1 to 10.

Images