Cloud server deployment method, device and system and computer equipment

Abstract

The invention relates to a cloud server deployment method and device, computer equipment, a storage medium and a computer program product. The embodiment of the invention can be applied to various scenes such as cloud technology, artificial intelligence, intelligent traffic, auxiliary driving and the like. The method comprises the following steps: in response to a cloud server creation request, determining request resource group information indicated by the cloud server creation request; determining grid resource group information adapted to the request resource group information, the grid resource group information indicating grid resources pre-divided from the first physical server, and the grid resources indicating available resources on the first physical server; and creating the cloud server on the first physical server indicated by the grid resource group information by using the available resources indicated by the grid resource group information. When the method is used for deploying the cloud server, various types of resources in the physical server can be effectively utilized.

Description

Technical Field

[0001] The present disclosure relates to the field of cloud computing technology, and particularly to a method, device, system, and computer device for deploying cloud servers. Background Art

[0002] With the continuous development of cloud computing technology, cloud server deployment is a server deployment method based on cloud computing technology. Through virtualization technology, physical server resources are divided into virtual machines, and the virtual machines are used as cloud servers. When the virtual machines are productized as cloud servers, it is required that only the same type of cloud servers can be deployed on one physical server.

[0003] However, when a certain type of resource on the physical server is allocated completely, due to the different resource usage requirements of each type of resource, other resources of the physical server will not be fully utilized, resulting in waste of resources on the physical server. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, system, and computer device for deploying cloud servers that can effectively utilize various types of resources in a physical server during cloud server deployment.

[0005] In a first aspect, the present disclosure provides a method for deploying a cloud server. The method includes:

[0006] Responding to a cloud server creation request, determining request resource group information indicated by the cloud server creation request, where the request resource group information indicates resources required for creating the cloud server;

[0007] Determining grid resource group information adapted to the request resource group information, where the grid resource group information indicates grid resources pre-partitioned from a first physical server, the grid resources indicate available resources on the first physical server, and the remaining resource quantity of resources of a specified resource type on the first physical server is lower than a resource quantity threshold, and the specified resource type is different from the available resource type of the available resources;

[0008] On the first physical server indicated by the grid resource group information, creating the cloud server by using the available resources indicated by the grid resource group information.

[0009] In a second aspect, the present disclosure further provides a device for deploying a cloud server. The device includes:

[0010] A resource determination module, configured to respond to a cloud server creation request and determine request resource group information indicated by the cloud server creation request, where the request resource group information indicates resources required for creating the cloud server;

[0011] A resource adaptation module determines grid resource group information that adapts to the requested resource group information. The grid resource group information indicates grid resources pre-partitioned on a first physical server. The grid resources indicate available resources on the first physical server. The remaining resource quantity of resources of a specified resource type on the first physical server is lower than a resource quantity threshold, and the specified resource type is different from the available resource type of the available resources.

[0012] A cloud server creation module is configured to create the cloud server on the first physical server indicated by the grid resource group information by using the available resources indicated by the grid resource group information.

[0013] In a third aspect, the present disclosure further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of any of the above method embodiments are implemented.

[0014] In a fourth aspect, the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0015] In a fifth aspect, the present disclosure further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0016] The cloud server deployment method, apparatus, system, computer device, storage medium, and computer program product described above, in response to a cloud server creation request, determine the requested resource group information indicated by the cloud server creation request, accurately determine the resources required to create the cloud server, and utilize the requested resource group information to accurately match a physical server that meets the requirements, thereby improving the success rate of cloud server creation. Grid resource group information adapted to the requested resource group information is determined, where the grid resource group information indicates at least one grid resource pre-allocated from a physical server, and the grid resource indicates available resources on a physical server where the remaining resource quantity of a resource of a specified resource type is less than a resource quantity threshold. Therefore, physical servers where the remaining resource quantity of a resource of a specified resource type is less than a resource quantity threshold can be fully utilized. Grid resource group information adapted to the requested resource group information is accurately found among such physical servers, and a physical server capable of creating a cloud server is determined. The cloud server is then created on the physical server indicated by the grid resource group information using the available resources indicated by the grid resource group information. Since the physical server indicated by the grid resource group information is a physical server whose remaining resource quantity of the specified resource type is lower than the resource quantity threshold, when the remaining quantity of a certain specified resource of the physical server is small, other resources in the physical server can also be effectively utilized, thereby improving the utilization rate of resources in the physical server. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 This is an application environment diagram of a cloud server deployment method in one embodiment;

[0019] Figure 2 A schematic diagram of a process for deploying a cloud server in one embodiment;

[0020] Figure 3 A schematic diagram of a process for creating a cloud server on a second physical server in one embodiment;

[0021] Figure 4 A schematic diagram of the architecture between a cloud server management system and a GPU physical server in one embodiment;

[0022] Figure 5 is a schematic diagram of a resource abstraction layer in one embodiment;

[0023] Figure 6 Schematic diagram of grouped monitoring resource changes in one embodiment;

[0024] Figure 7 Schematic diagram of the structure among the scheduler, resource monitor and migrator in another embodiment;

[0025] Figure 8 Flow schematic diagram of the creation process of a cloud server in another embodiment;

[0026] Figure 9 Flow schematic diagram of the deletion process of a cloud server in another embodiment;

[0027] Figure 10 Flow schematic diagram of the migration process of a cloud server in another embodiment;

[0028] Figure 11 Schematic diagram of the resource operation of the scheduler in another embodiment;

[0029] Figure 12 Flow schematic diagram of the process of specifically deploying a cloud server in another embodiment;

[0030] Figure 13 Structural schematic block diagram of a cloud server deployment device in one embodiment;

[0031] Figure 14 Internal structural schematic diagram of a computer device in one embodiment. Detailed implementation manners

[0032] In order to make the objectives, technical solutions and advantages of the present disclosure clearer and more understandable, the present disclosure will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0033] It should be noted that the terms "first", "second", etc. in the description and claims of this article and the above accompanying drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or equipment that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or equipment.

[0034] Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and networks within a wide area network or local area network to achieve data computing, storage, processing, and sharing.

[0035] Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model. It can form a resource pool, be used on demand, and is flexible and convenient. Cloud computing technology will become an important support. The back-end services of the technical network system require a large amount of computing and storage resources, such as video websites, picture websites, and more portal websites. With the high development and application of the Internet industry, in the future, each item may have its own identification mark and needs to be transmitted to the back-end system for logical processing. Data at different levels will be processed separately, and various industry data requires a powerful system back-end support, which can only be achieved through cloud computing.

[0036] Public Cloud usually refers to the cloud that can be used provided by a third-party provider for users. Public Cloud can generally be used through the Internet and may be free or inexpensive. The core attribute of Public Cloud is shared resource services. There are many instances of this cloud that can provide services in today's entire open public network.

[0037] The embodiments of the present disclosure provide a method for deploying a cloud server, which can be applied to, for example Figure 1In the application environment shown. A cloud server deployment method provided by an embodiment of the present disclosure can be applied in a public cloud environment. Among them, the terminal 102 communicates with the cloud server management system 104 through the network. The cloud server management system 104 can be connected to multiple physical servers. When the terminal 102 needs to use a virtual machine, the terminal 102 can send a cloud server creation request to the cloud server management system 104. In response to the cloud server creation request, the cloud server management system 104 determines the requested resource group information indicated by the cloud server creation request. The requested resource group information indicates the resources required to create the cloud server. The cloud server management system 104 determines the grid resource group information that adapts to the requested resource group information. The grid resource group information indicates the grid resources pre-partitioned from the first physical server among the physical servers 106. The grid resources indicate the available resources on the first physical server. The remaining resource quantity of the resources of the specified resource type on the first physical server is lower than the resource quantity threshold. The available resource type different from the available resources of the specified resource type. The cloud server management system 104 creates a cloud server on the first physical server indicated by the grid resource group information, using the available resources indicated by the grid resource group information. Among them, the cloud server management system 104 can be implemented by an independent server or a server cluster composed of multiple servers. The physical server 106 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be a server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the cloud server management system can be directly or indirectly connected through wired or wireless communication methods, and this application does not make any restrictions here.

[0038] In one embodiment, as Figure 2 shown, a cloud server deployment method is provided. Taking the cloud server management system 104 applied in Figure 1 as an example, the method includes the following steps:

[0039] S202, in response to the cloud server creation request, determine the requested resource group information indicated by the cloud server creation request, where the requested resource group information indicates the resources required to create the cloud server.

[0040] Among them, the cloud server creation request can be a request issued by a terminal. When the terminal needs to use a cloud server and there is no cloud server corresponding to the terminal in the physical server, the terminal can send a cloud server creation request to the cloud server management system. The requested resource group information may include: various resource types required for creating a cloud server and the amount of resources of each resource type. The requested resource group information may also include: some other data required for creating a cloud server, such as some parameter information of the cloud server network to be configured, some parameter information of the operating system of the cloud server to be configured, information on the type of the cloud server to be configured, the type of data stored in the cloud server to be configured, and so on. The cloud server management system can be a software system for managing and monitoring cloud servers, which provides a centralized way to manage and operate cloud servers, including aspects such as the deployment, configuration, monitoring, migration, deletion, maintenance, and security of cloud servers. In some embodiments of the present disclosure, the application environment where the cloud server is located can be a public cloud environment.

[0041] Specifically, when the terminal needs to use a cloud server and there is no cloud server that has been created by the terminal in the physical server, a cloud server needs to be created in the physical server first. At this time, the terminal sends the cloud server creation request to the cloud server management system. In response to the cloud server creation request, the cloud server management system can determine the requested resource group information indicated by the cloud server creation request.

[0042] S204, determine grid resource group information that adapts to the requested resource group information. The grid resource group information indicates grid resources pre-partitioned on a first physical server. The grid resources indicate the available resources on the first physical server. The remaining resource quantity of the resources of the specified resource type on the first physical server is lower than the resource quantity threshold, and the specified resource type is different from the available resource type of the available resources.

[0043] Among them, the specified resource type can be set according to actual needs. For example, it can be one of resources such as CPU (Central Processing Unit) resources, hard disk resources, GPU (graphics processing unit) resources, memory resources, and FPGA resources. When the remaining resource quantity of a certain specified resource type in a physical server, such as GPU resources, is relatively small and lower than the resource quantity threshold, and there are still remaining resources of other types of resources in this physical server, such as CPU resources, memory resources, and hard disk resources, at this time, in order to ensure the full utilization of resources, this physical server can be determined as the first physical server. One or more of the remaining resources (one or more of CPU resources, memory resources, and hard disk resources) can be pre-divided into grid resources. Then, grid resource group information describing the grid resources is generated. Usually, the available resources indicated by the grid resource group information can only be used to create one cloud server, so as to ensure the full utilization of various resources in the first physical server. The pre-divided grid resources can be determined according to the type of cloud server in the actual application process. For example, when the number of type A cloud servers used is relatively large, the grid resources can be divided according to the resources required to create type A cloud servers, so as to ensure the full utilization of the remaining resources in the physical server.

[0044] Specifically, since the requested resource group information indicates the resources required to create a cloud server. The grid resource group information also indicates the available resources. Therefore, the grid resource group information adapted to the requested resource group information can be determined among all the grid resource group information. More specifically, the resources required to create a cloud server and the available resources indicated by the grid resource group information can be used for matching, so as to determine the grid resource group information adapted to the requested resource group information.

[0045] In some exemplary embodiments, the exhaustion of GPU resources in a GPU physical server is used as an example for illustration. When the remaining resource quantity of GPU resources in GPU physical server A and GPU physical server B is lower than the resource quantity threshold. At this time, there are still remaining CPU resources, memory resources, and hard disk resources in GPU physical server A and GPU physical server B. If there are 20 cores remaining in the CPU resources of GPU physical server A, 16G remaining in the memory resources, and 50GB remaining in the hard disk resources. In GPU physical server B, there are 10 cores remaining in the CPU resources, 16G remaining in the memory resources, and 50GB remaining in the hard disk resources. At this time, if creating a cloud server of type A1 requires using 8 cores of CPU, 8G of memory, and 20G of hard disk resources, then two grid resources corresponding to the cloud server of type A1 can be partitioned in GPU physical server A, and the grid resource group information describing the two grid resources can be a1 and a2 respectively. One grid resource corresponding to the cloud server of type A1 can be partitioned in GPU physical server B, and the grid resource group information describing this grid resource can be a3. At this time, if the resources indicated by the requested resource group information are to use 8 cores of CPU, 8G of memory, and 20G of hard disk resources (a cloud server of type A1 needs to be created), then the grid resource group information adapted to the requested resource group information can be a1, a2, and a3. At this time, either GPU physical server A or GPU physical server B can be selected to create the cloud server.

[0046] If creating a cloud server of type A2 requires using 10 cores of CPU, 16GB of memory, and 40G of hard disk resources, and one grid resource corresponding to the cloud server of type A2 has been partitioned in GPU physical server B, and the grid resource group information describing this grid resource can be a4. If there are still two grid resources corresponding to the cloud server of type A1 in GPU physical server A, and the grid resource group information is still a1 and a3, if the resources indicated by the requested resource group information are to use 10 cores of CPU, 16G of memory, and 40G of hard disk resources (a cloud server of type A2 needs to be created), then the grid resource group information adapted to this requested resource group information can be a3. At this time, the cloud server can only be created in GPU physical server B.

[0047] S206. On the first physical server indicated by the grid resource group information, use the available resources indicated by the grid resource group information to create the cloud server.

[0048] Specifically, after finding the adapted grid resource group information, the first physical server indicated by this grid resource group information can be found. Then, use the available resources indicated by the grid resource group information to create the cloud server on the found first physical server.

[0049] In some exemplary embodiments, continuing with the above example for illustration, after finding the adapted grid resource group information a1 and a2, the first physical server indicated by the grid resource group information a1 and a2 can be determined. The first physical server can be GPU physical server A. In GPU physical server A, cloud servers can be created using the available resources indicated by the grid resource group information.

[0050] In some specific embodiments, during the process of creating a cloud server, a VM Allocation object can be generated according to the resource group information requested, which indicates the resources required for creating the cloud server. VM Allocation refers to the process of allocating the resources of the first physical server to virtual machines, including allocating CPU resources, memory resources, hard disk resources, network resources, etc. to virtual machines to ensure that the virtual machines can run normally and meet the user's requirements. After generating the VM Allocation object, the resources allocated to the virtual machine (i.e., the resources required for creating the cloud server indicated by the requested resource group information) can be used to generate a virtual machine in the first physical server. This virtual machine is usually used as a cloud server. If the generation of the virtual machine is successful, it can be determined that the creation of the cloud server is successful.

[0051] In the above cloud server deployment method, in response to a cloud server creation request, the requested resource group information indicated by the cloud server creation request is determined, which can accurately determine the resources required for creating the cloud server. Using the requested resource group information, a physical server that meets the requirements can be accurately matched, thereby improving the success rate of creating a cloud server. The grid resource group information that adapts to the requested resource group information is determined. The grid resource group information indicates at least one grid resource pre-partitioned from a physical server. The grid resource indicates the available resources on a physical server where the remaining resource quantity of a specified resource type is lower than the resource quantity threshold. Therefore, physical servers where the remaining resource quantity of a specified resource type is lower than the resource quantity threshold can be fully utilized. Among these physical servers, the grid resource group information that adapts to the requested resource group information can be accurately found, and the physical server capable of creating a cloud server can be determined. On the physical server indicated by the grid resource group information, the cloud server is created using the available resources indicated by the grid resource group information. Since the physical server indicated by the grid resource group information is a physical server where the remaining resource quantity of a specified resource type is lower than the resource quantity threshold, when the remaining quantity of a certain specified resource in the physical server is less, other resources in the physical server can still be effectively utilized, improving the utilization rate of resources in the physical server.

[0052] In one embodiment, the method further includes:

[0053] In response to the cloud server being successfully created on the first physical server, a first binding relationship is created between the grid resource group information and the cloud server, the first binding relationship indicating that the available resources indicated by the grid resource group information have been used to create the cloud server.

[0054] Specifically, in response to the successful creation of a cloud server on the first physical server, since the cloud server is generated using the available resources indicated by the grid resource group information, a first binding relationship can be established between the grid resource group information and the cloud server. This first binding relationship can determine that the available resources indicated by the grid resource group information have been used to create the cloud server. When other subsequent requests for resource group information are adapted, the grid resource group information need not be adapted. This ensures that the available resources indicated by the grid resource group information will not be reused. When multiple cloud server creation requests are received, since the available resources indicated by the grid resource group information will only be used once, the success rate of cloud server creation can be guaranteed.

[0055] In some exemplary embodiments, for example, a cloud server creation request A is first received, and the requested resource group information A indicated by the cloud server creation request A is determined. When multiple grid resource group information exists, such as S1 grid resource group information, S2 grid resource group information, and S3 grid resource group information, the S2 grid resource group information is adapted to the requested resource group information A. Therefore, a first binding relationship is established between the S2 grid resource group information and the successfully created cloud server. When a cloud server creation request B is received, the requested resource group information B indicated by the cloud server creation request B is determined. Because the S2 grid resource group information has a first binding relationship, when adapting using the requested resource group information B, it is not adapted to the S2 grid resource group information. The requested resource group information B is adapted only in the S1 grid resource group information and the S3 grid resource group information to ensure that the available resources indicated by the grid resource group information are only used once.

[0056] In response to a cloud server deletion request, when the binding relationship of the cloud server is a first binding relationship, the first binding relationship is released, and the cloud server on the first physical server is deleted.

[0057] The cloud server deletion request may be a cloud server deletion request for a successfully created cloud server after the cloud server has been successfully created on a physical server. In this embodiment, the cloud server deletion request may be a cloud server deletion request for a successfully created cloud server on the first physical server. In some application scenarios, when a cloud server needs to be destroyed, the terminal may issue a cloud server deletion request.

[0058] Specifically, in response to a cloud server deletion request, it is possible to determine the cloud server that has been successfully created in the first physical server indicated by the cloud server deletion request. When the binding relationship of the cloud server is a first binding relationship, it can be determined that the successfully created cloud server was created using the available resources indicated by the grid resource group information, since the available resources indicated by the grid resource group information cannot create multiple cloud servers at the same time. Therefore, the grid resource group information will have a first binding relationship, so it is necessary to first release the first binding relationship and release the available resources indicated by the grid resource group information. After the first binding relationship is released, the cloud server on the first physical server can be deleted. After the first binding relationship is released, since the first binding relationship does not exist, the available resources indicated by the grid resource group information can subsequently be used to create cloud services.

[0059] In some specific embodiments, since a VM Allocation object is generated when a cloud server is created, a first binding relationship may be created between the VM Allocation object and the grid resource group information. When a cloud server is deleted, the VM Allocation object corresponding to the cloud server is deleted. If a first binding relationship exists between the VM Allocation and the grid resource group information, the first binding relationship must be removed.

[0060] In this embodiment, by establishing a first binding relationship between a cloud server and grid resource group information, upon receiving different cloud server creation requests, if the grid resource group information matches the requested resource group information indicated by the different cloud server creation requests, it is possible to ensure that the available resources indicated by the grid resource group information will not be reused, and only one cloud server can be created, thereby ensuring the success rate of cloud server creation. Furthermore, in response to a cloud server deletion request, if the cloud server's binding relationship is the first binding relationship, the first binding relationship can be first released, releasing the available resources indicated by the grid resource group information, thereby ensuring that they can be reused in subsequent responses to other cloud server creation requests.

[0061] In one embodiment, the method further comprises:

[0062] When the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be no less than the resource quantity threshold, and the available resources indicated by the grid resource group information on the first physical server have been used to create the cloud server, migrating the cloud server to a physical server adapted to the resources required by the cloud server;

[0063] After the cloud server is migrated, the grid resource group information of the first physical server is deleted.

[0064] Among them, cloud server migration refers to the process of migrating one or more cloud servers from one physical server to another physical server.

[0065] Specifically, after one or more cloud servers created on the first physical server are destroyed or deleted, releasing the available resources on the first physical server, usually the remaining resource quantity of the resources of the specified resource type on the first physical server increases. When the remaining resource quantity of the specified resource type on the first physical server is updated from being lower than the resource quantity threshold to not being lower than the resource quantity threshold, and the available resources indicated by the grid resource group information on the first physical server have been used to create cloud servers. First, it is possible to determine the resources required by the cloud servers successfully created on the first physical server. Then, the cloud servers created on the first physical server are migrated to the physical server adapted to the resources required by the cloud servers. After the cloud servers created with the available resources indicated by the grid resource group information on the first physical server are migrated, the grid resource group information of the first physical server is deleted. It should be noted that after the grid resource group information of the first physical server is deleted, there are no pre-divided grid resources on the first physical server.

[0066] In some exemplary embodiments, it is exemplified by specifying that the resource type is GPU resource. When the GPU resource on the first physical server changes from the remaining 1 core to the remaining 10 cores, and the GPU resource threshold is 2 cores, it can be determined that the remaining resource quantity of the specified resource type on the first physical server is updated from being lower than the resource quantity threshold to not being lower than the resource quantity threshold. And if the quantities of other types of resources in the first physical service are: 1 core CPU, 8G memory, and 60G hard disk capacity at this time. If the cloud server created using the available resources indicated by the grid resource group information in the first physical server is not deleted, and if the lowest standard cloud server to be created subsequently requires 2 core CPU resources, 4G memory, and 10G hard disk capacity, but the remaining 1 core CPU resource is insufficient, it will cause the GPU resource to be unable to be fully utilized although it has recovered. If the available GPU resources, memory resources, and hard disk resources in the first physical server are used to divide the grid resources again, during the process of deleting the cloud server, due to the different available resources indicated by the grid resource information used by various cloud servers, it will cause chaotic resource allocation and the various types of resources in the physical server cannot be reasonably allocated. And the resources in the monitored physical server also need to monitor the remaining quantities of different types of resources, further causing chaotic implementation logic and reducing the creation success rate of the cloud server. Therefore, when the available resources indicated by the grid resource group information on the first physical server have been used to create a cloud server. If the resources required for the created cloud server are 10 core CPU, 8G memory, and 20G hard disk, it is necessary to find other physical servers that are different from the first physical server and are adapted to the resources required by the cloud server. Migrate the cloud server to the other adapted physical servers. Delete the grid resource group information in the first physical server. At this time, since the migration of the cloud server has been completed, the available resources in the first physical service include: 10 core GPU, 11 core CPU, 16G memory, and 80G hard disk capacity. Therefore, it can meet the creation of the lowest standard cloud server. In addition, if there is a need to create a high-performance cloud server, such as requiring 10 core GPU, 10 core CPU, 16G memory, and 50G hard disk capacity, it can also make full use of the available resources on the first physical server to create a high-performance cloud server.

[0067] In this embodiment, when the remaining resource quantity of the resource of the specified resource type on the first physical server is updated to not be lower than the resource quantity threshold, and the available resources indicated by the grid resource group information on the first physical server have been used to create a cloud server, the cloud server can be migrated first. Thus, the resources in the first physical server are released, and other types of cloud servers can be created according to the available resources on the current first physical server, which can make full use of the available resources on the first physical server, and the logical implementation is simple, which can ensure the reasonable use and allocation of resources.

[0068] In one embodiment, the method further includes:

[0069] When the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be not lower than the resource quantity threshold, and there is grid resource group information indicating available resources that are not used on the first physical server, delete the grid resource group information indicating available resources that are not used.

[0070] Specifically, when the remaining resource quantity of the specified resource type on the first physical server is updated from being lower than the resource quantity threshold to being not lower than the resource quantity threshold, and there is grid resource group information indicating available resources that are not used on the first physical server. There are the following two situations for the grid resource group information indicating available resources that are not used. 1. All the available resources indicated by the grid resource group information of the first physical server are not used. At this time, all the grid resource group information of the first physical server can be deleted. 2. Some of the available resources indicated by the grid resource group information of the first physical server have been used, and some of the available resources indicated by the grid resource group information have not been used. At this time, the grid resource group information indicating available resources that are not used can be directly deleted. For the available resources indicated by some of the grid resource group information that have been used, the cloud services created by the used available resources can be migrated in the manner of the above embodiment. After migration, delete some of the grid resource group information that has been used.

[0071] In this embodiment, when the remaining resource quantity of the specified resource type on the first physical server is updated from being lower than the resource quantity threshold to being not lower than the resource quantity threshold, by deleting the grid resource group information indicating available resources that are not used, the available resources indicated by the grid resource group information can be released, so that the resources on the first physical server can be allocated more reasonably, the resource utilization rate can be improved, and the resource waste rate can be reduced.

[0072] In one embodiment, as Figure 3 shown, the method further includes:

[0073] S302. When there is no grid resource group information that adapts to the requested resource group information, determine the host resource group information that adapts to the requested resource group information, where the host resource group information indicates the host resources determined from the second physical server, and the remaining resource quantity of the resources of the specified resource type on the second physical server is not lower than the resource quantity threshold.

[0074] Among them, the host resources can usually be all the available resources in the second physical server. If there are both host resources and grid resources in the second physical server, the host resources can be the available resources other than the grid resources. The host resource group information is usually the information describing the host resources, such as the resource type of the available resources in the second physical server, the resource quantity of the available resources, etc.

[0075] Specifically, when there is no grid resource group information of the first physical server that matches the requested resource group information, it can be determined that the grid resources divided from the remaining available resources in the first physical server do not meet the requirements for creating the cloud server. At this time, it is necessary to match in the host resource group information of the second physical server to determine the host resource group information that matches the requested resource group information.

[0076] S304, Deduct the resources required to create the cloud server indicated by the requested resource group information from the available resources indicated by the host resource group information.

[0077] Specifically, after determining the host resource group information, since the available resources indicated by the host resource group need to be used. And the available resources indicated by the host resource group are usually all the available resources in the second physical server. Therefore, in order to ensure that the available resources indicated by the host resource group are not reused and to ensure the success rate of creating other cloud servers, the resources required to create the cloud server indicated by the requested resource group information can be deducted.

[0078] S306, Create the cloud server on the second physical server indicated by the host resource group information by using the available resources indicated by the host resource group information.

[0079] Specifically, after determining the matching host resource group information, the second physical server indicated by the host resource group information can be found. Then, use the available resources indicated by the host resource group information to create a cloud server on the found second physical server.

[0080] In some specific embodiments, during the process of creating a cloud server, the resources required for creating the cloud server can be indicated according to the requested resource group information to generate a VM Allocation object. VM Allocation refers to the process of allocating the resources of the second physical server to a virtual machine. This includes allocating CPU resources, memory resources, hard disk resources, network resources, etc. to the virtual machine to ensure that the virtual machine can run normally and meet the user's requirements. After generating the VM Allocation object, the resources allocated to the virtual machine (i.e., the resources required for creating the cloud server indicated by the requested resource group information) can be used to generate a virtual machine in the second physical server, and at the same time, the resources allocated to the virtual machine are deducted from the available resources indicated by the host resource group information. This virtual machine is usually used as a cloud server. If the generation of the virtual machine is successful, it can be determined that the creation of the cloud server is successful.

[0081] In some exemplary embodiments, when the resources required for creating the cloud server indicated by the requested resource group information are 10-core CPU, 16G memory, 10-core GPU, and 100GB hard disk capacity, and the available resources indicated by the grid resource group information of the first physical server do not meet the above requirements. Therefore, no further matching will be performed from the grid resource group information of the first physical server. Matching can be performed from the host resource group of the second physical server (a physical server whose remaining resource quantity of all types of resources is not lower than the resource quantity threshold). The host resource group information that matches the requested resource group information is determined. If the available resources indicated by the host resource group information are 50-core CPU, 64G memory, 30-core GPU, and 500G hard disk capacity, then a part of the available resources indicated by the host resource group information corresponding to the resources required for creating the cloud server indicated by the requested resource group information can be used to create a cloud server in the second physical server indicated by the host resource group information. At the same time, the available resources indicated by the host resource group information are deducted. After deduction, the available resources indicated by the host resource group information are 40-core CPU, 48G memory, 20-core GPU, and 400G hard disk capacity.

[0082] In this embodiment, when there is no grid resource group information that matches the requested resource group information, the host resource group information that matches the requested resource group information can be determined according to the host resource group information of the second physical server. And the resources required for creating the cloud server indicated by the requested resource group information are deducted from the host resource group information, which can deduct the corresponding available resources after creating the cloud server, ensuring that resources are not reused. And when creating other cloud servers subsequently, due to the deduction of the corresponding available resources, it can also ensure that resources are not mismatched, ensuring the success rate of creating cloud servers.

[0083] In one embodiment, the method further includes:

[0084] In response to successfully creating the cloud server on the second physical server, create a second binding relationship between the host resource group information and the cloud server, where the second binding relationship indicates that the available resources indicated by the host resource group information have been used to create the cloud server.

[0085] Specifically, when the cloud server is successfully created on the second physical server, a second binding relationship can be created between the host resource group information and the successfully created cloud server. Since the available resources indicated by the host resource group information can generally be used to create multiple cloud servers, the second binding relationship can be used subsequently to determine which available resources in which second physical server are used to create the cloud server, and it can also be determined in which second physical server the cloud server is created. In this way, when matching cloud servers, for example, when it is necessary to locate in which physical server a certain cloud service is created, the second binding relationship can be directly used for determination.

[0086] In this embodiment, by using the second binding relationship, it is possible to accurately match which physical server the cloud server is created in, facilitating the search and management of cloud servers.

[0087] In one embodiment, the method further includes:

[0088] In response to a cloud server deletion request, when the binding relationship of the cloud server is the second binding relationship,

[0089] Add the deducted available resources to the available resources indicated by the host resource group information;

[0090] Release the second binding relationship and delete the cloud server on the second physical server.

[0091] Wherein, in this embodiment, the cloud server deletion request may be a cloud server deletion request for a cloud server successfully created on the second physical server.

[0092] Specifically, in response to the cloud server deletion request, it can be determined the cloud server successfully created in the second physical server indicated by the cloud server deletion request. When the binding relationship of the cloud server is the second binding relationship, the corresponding host resource group information can be found according to the second binding relationship. Then, add the available resources deducted when creating the cloud server to the available resources indicated by the host resource group information. And release the second binding relationship, and at the same time delete the cloud server on the second physical server.

[0093] In some exemplary embodiments, for example, after creating a cloud server using the available resources indicated by the host resource group information, the cloud server needs to be deleted. The resources required to create the cloud server are 10-core CPU, 16G memory, 10-core GPU, and 100GB hard disk capacity. The available resources indicated by the host resource group information after deduction are 40-core CPU, 48G memory, 20-core GPU, and 400G hard disk capacity. Therefore, the deducted 10-core CPU, 16G memory, 10-core GPU, and 100GB hard disk capacity can be added to the available resources indicated by the host resource group information. The final available resources indicated by the host resource group information are 50-core CPU, 64G memory, 30-core GPU, and 500G hard disk capacity. It can be understood that the above is only for illustration.

[0094] In a specific embodiment, since a VM Allocation object needs to be generated when a cloud server is created, a second binding relationship can be created between the VM Allocation object and the host resource group information. When a cloud server needs to be deleted, the VM Allocation object corresponding to the cloud server needs to be deleted. If a second binding relationship exists between the VM Allocation and the host resource group information, the second binding relationship needs to be released first, and then the resources deducted from the available resources indicated by the host resource group information when generating the VM Allocation object are returned to the available resources indicated by the host resource group information.

[0095] In this embodiment, when the cloud server is deleted and the binding relationship of the cloud service is the second binding relationship, the deducted available resources are added to the available resources indicated by the host resource group information, and then the second binding relationship is released, and the available resources indicated by the deducted host resource group information are released. When the cloud server is subsequently created, the released available resources can also be used, thereby improving resource utilization.

[0096] In one embodiment, the method further comprises:

[0097] When a physical server is detected to be online, host resource group information is created according to resource information of the physical server;

[0098] When the remaining resource quantity of the resource of the specified resource type in the physical server is lower than the resource quantity threshold, dividing the available resources of the preset resource quantity from the available resources indicated by the host resource group information;

[0099] Grid resource group information is generated for the preset amount of available resources.

[0100] Among them, the online of a physical server refers to connecting the physical server to the cloud server management system and making it available. At this time, the physical server can receive and process various requests sent by the cloud server management system. The resource information may include: hard disk resources, GPU resources, CPU resources, FPGA resources, and so on.

[0101] Specifically, after the physical server is online, the Compute Agent component in the physical server can collect resource information such as the physical server's IP (Internet Protocol), CPU model, number of CPUs, memory size, Disk (hard disk) type, number of Disks (hard disks), etc. The Compute Agent component will synchronize the collected information to the cloud server management system. The Compute Agent component can be a component used to execute various tasks in the computer. When the cloud server management system detects that the physical server is online and can receive the resource information sent by the Compute Agent component, the cloud server management system creates host resource group information based on the resource information. When the remaining resource quantity of the resource of a specified resource type in the physical server is lower than the resource quantity threshold, the cloud server management system can allocate a preset resource quantity of available resources from the available resources indicated by the host resource group information. Then, for the description information of the preset resource quantity of available resources, the cloud server management system generates grid resource group information. The description information usually may include: the resource type describing the available resources, and the resource quantity information of the available resources of each resource type.

[0102] In some exemplary embodiments, as Figure 4 shown, when it is detected that a GPU physical server is online, the cloud server management system can obtain the resource information of each GPU physical server (GPU Physical Server 1, GPU Physical Server 2, GPU Physical Server 3) and generate host resource group information. The cloud server management system can monitor the remaining resource quantity of the GPU resources in each GPU physical server. When the remaining resource quantity of the GPU resources in GPU Physical Server 2 and GPU Physical Server 3 is lower than the resource quantity threshold, a preset resource quantity of available resources can be allocated from the available resources indicated by the host resource group information, and grid resource group information is generated for the preset quantity of available resources.

[0103] In some specific embodiments, after the resource information of a GPU physical server is transmitted to the cloud service management system, the cloud service management system can generate host resource group information, and the content of the host resource group information is as follows:

[0104] Host resource group information = (Resource details)

[0105] Resource details = {CPU: 100 cores, 100 cores; MEMORY: 500GB, 500GB; GPU cards: 4, 4}.

[0106] In addition, when the GPU resources in the GPU physical server are lower than the resource quantity threshold, grid resource group information can be created.

[0107] Grid resource group information = (Resource details)

[0108] Resource details = {CPU: 16 cores; MEMORY: 32G}

[0109] MEMORY refers to memory.

[0110] After creating the grid resource group information, the content of the host resource group information becomes:

[0111] Host resource group information = (Resource details)

[0112] Resource details = {CPU: 100 cores, 84 cores; MEMORY: 500GB, 468GB; GPU cards: 4, 4}.

[0113] When the grid resource group information is deleted, the resources occupied by the grid resource group information will be returned to the host resource group information. It should be noted that the above is only an example with the resource details indicated in the host resource group information. In the actual application process, other types of information can also be indicated in the host resource group information.

[0114] In addition, generating the creation of grid resource group information can include carrying the physical server IP. When it is detected that the GPU resources in the GPU physical server are lower than the resource quantity threshold, the host resource group information of this physical server can be found according to this physical server IP. If the host resource group information meets the resource requirements for creating the grid resource group information, corresponding resource deduction operations will be performed on the host resource group information, and the corresponding grid resource group information will be generated. If the host resource group information is not found or does not meet the resource requirements for creating the grid resource group information, it means that the creation of the grid resource group information fails. For example, the resource requirements for creating the grid resource group information are 10-core CPU and 64G memory, but the available resources in the host resource group information are 5-core CPU and 32G memory. It can be determined that the host resource group information does not meet the resource requirements for creating the grid resource group information, and the creation of the grid resource group information fails.

[0115] In this embodiment, according to the relationship between the remaining resource quantity of the resources of the specified resource type in the physical server and the resource quantity threshold, available resources with a preset resource quantity can be flexibly divided from the available resources indicated by the host resource group information, so as to generate grid resource group information. Subsequently, the available resources indicated by the grid resource group information can be utilized to create cloud servers, enabling the flexible utilization of resources in the physical server and improving the resource utilization rate.

[0116] In one embodiment, as Figure 5 shown, the grid resource group information and the host resource group information can be stored at a specific location in the cloud server management system or an external database. For example, it is stored in the resource abstraction layer constructed by the cloud server management system. Since various resource information in the physical server generates the grid resource group information and the host resource group information, when processing resources, it is only necessary to process the grid resource group information and the host resource group information in the resource abstraction layer, and there is no need to operate on the actual resources.

[0117] In one embodiment, the method further includes:

[0118] In response to the offline trigger event of the physical server, when a cloud server is created in the physical server, migrate the cloud server to a physical server adapted to the resources required by the cloud server;

[0119] After the cloud server is migrated, delete the host resource group information and the grid resource group information of the physical server that needs to be offline, and take the physical server offline.

[0120] Among them, the offline trigger event can be issued by the physical server when it needs to be shut down or when the physical server fails.

[0121] Specifically, in response to the offline trigger time of the physical server, it is determined that the physical server needs to be offline. When a cloud server is created in the physical server, in order to ensure the user experience during the use of the cloud server and the stability of the user's use of the cloud server. The cloud server can be migrated to a physical server adapted to the resources required by the cloud server. After the cloud server is migrated, delete the host resource group information and the grid resource group information of the physical server that needs to be offline, and take the physical server offline. Subsequently, when determining the grid resource group information adapted to the requested resource group information or the host resource group information adapted to the requested resource group information, since the host resource group information and the grid resource group information of the physical server have been deleted and the physical server has been taken offline, the host resource group information and the grid resource group information of the physical server will not be determined as the host resource group information or the grid resource group information adapted to the requested resource group information, so a cloud server will not be created in the physical server.

[0122] In response to the offline trigger event of the physical server, when no cloud server is created in the physical server, delete the host resource group information and grid resource group information of the physical server to be taken offline, and take the physical server offline.

[0123] Specifically, in response to the offline trigger event of the physical server, determine that the physical server needs to be taken offline. When no cloud server is created in the physical server, the host resource group information and grid resource group information of the physical server to be taken offline can be directly deleted. After the deletion is completed, the physical server is directly taken offline.

[0124] In this embodiment, in response to the offline trigger event of the physical server, when a cloud server is created in the physical server, the cloud server needs to be migrated to ensure the stability during the operation of the cloud service and also not affect the user experience of using the cloud server. When no cloud server is created in the physical server, the physical server can be directly taken offline. In addition, when the physical server is taken offline, the host resource group information and grid resource group information need to be deleted, so that the subsequent request for resource group information will not determine that the adapted grid resource group information or host resource group information is the host resource group information and grid resource group information of the offline physical server, reducing the probability of failure in creating a cloud server.

[0125] In one embodiment, the migrating the cloud server to a physical server adapted to the resources required by the cloud server includes:

[0126] Determine the grid resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server;

[0127] Migrate the cloud server to the physical server indicated by the grid resource group information adapted to the cloud server resources, and the physical server indicated by the grid resource group information adapted to the cloud server resources is different from the current physical server where the cloud server is located.

[0128] Among them, the cloud server resources may include: CPU resources, memory resources, hard disk resources, GPU resources, etc.

[0129] Specifically, it is possible to first determine the cloud server resources occupied by the cloud server to be migrated. To ensure the full utilization of the resources in the first physical server, when the cloud server is migrated, first

[0130] Determine grid resource group information adapted to the cloud server resources based on the cloud server resources, and determine the physical servers indicated by the grid resource group information. Then, migrate the cloud server to be migrated to the determined physical servers. Usually, the physical servers indicated by the grid resource group information adapted to the cloud server resources are different from the physical servers where the current cloud server is located, that is, the physical servers where the cloud server is located before migration and the physical servers where the cloud server is located after migration are different. After the cloud service migration is completed, a first binding relationship is established between the grid resource group information and the cloud server.

[0131] In this embodiment, when a cloud server needs to be migrated, determine grid resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server. First, match in the grid resource group information, and the physical servers with grid resource group information usually have the remaining resource quantity of resources of a specified resource type lower than the resource quantity threshold. Therefore, during the migration of the cloud server, other resources in the physical server can also be effectively utilized, improving the utilization rate of resources in the physical server.

[0132] In one embodiment, the method further includes:

[0133] When there is no grid resource group information adapted to the cloud server resources, determine host resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server;

[0134] Migrate the cloud server to the physical server indicated by the host resource group information adapted to the cloud server resources, and the physical server indicated by the host resource group information adapted to the cloud server resources is different from the physical server where the cloud server is located.

[0135] Specifically, when after searching all the grid resource group information, there is no grid resource group information adapted to the cloud server resources, it can be determined that the cloud server cannot be migrated to the first physical server. Therefore, host resource group information adapted to the cloud server resources can be determined according to the cloud server resources occupied by the cloud server. Then, migrate the cloud server to the physical server indicated by the host resource group information adapted to the cloud server resources. Usually, the physical server indicated by the host resource group information adapted to the cloud server resources is different from the physical server where the current cloud server is located, that is, the physical servers where the cloud server is located before migration and the physical servers where the cloud server is located after migration are different.

[0136] In some specific embodiments, when it is determined that a cloud server needs to be migrated, usually the cloud server management system will receive a cloud server migration request. The cloud server management system will asynchronize the cloud server migration request into a Task task and store it in the database. There is a built-in asynchronous thread in the cloud server management system. The cloud server management system can use the asynchronous thread to continuously read the Task task from the database and then perform the cloud server migration operation. In this embodiment, by asynchronizing the request, the waiting time during the cloud server migration process can be reduced. During the waiting for migration, other tasks can also be executed, improving the execution efficiency of other tasks in the cloud server management system.

[0137] In addition, it should be noted that the process of migrating the cloud server to the physical server indicated by the host resource group information adapted to the cloud server resources is roughly the same as the process of creating a cloud server on the second physical server. After the cloud service migration is completed, a second binding relationship can be established between the host resource group information and the cloud server.

[0138] In this embodiment, when there is no grid resource group information adapted to the cloud server resources, the host resource group information adapted to the cloud server resources can be determined, and the cloud server can be migrated to the physical server indicated by the host resource group information adapted to the cloud server resources, which can ensure the success of the cloud server migration and ensure that the cloud server will not suddenly disappear due to the offline of the physical server during the operation, thereby improving the stability during the cloud service operation.

[0139] In one embodiment, the method further includes:

[0140] Grouping the online physical servers according to the resource types and resource amounts of the available resources indicated by the host resource group information of the online physical servers to obtain multiple groups of physical servers;

[0141] Monitoring the remaining resource quantities of the resources of the specified resource type for each group of physical servers, where the specified resource types monitored in each group of physical servers are different.

[0142] Specifically, for the physical servers that have been put on line, in some embodiments of the present disclosure, the physical servers that have been put on line are usually physical servers that have established a connection with the cloud server management system. They can be grouped according to the resource type and resource quantity of the available resources indicated by the host resource group information of the physical servers that have been put on line. For example, group the physical servers with more GPU resources according to GPU resources. Or group the physical servers with more CPU resources according to CPU resources. After grouping, multiple groups of physical servers are obtained. Then, the remaining resource quantity of the resources of the specified resource type in each group of physical servers can be monitored, where the specified resource types monitored in each group of physical servers are different. For example, divide the physical servers into three groups A, B, and C. The physical servers in group A can monitor the remaining resource quantity of CPU resources. The physical servers in group B can monitor the remaining resource quantity of GPU resources. The physical servers in group C can detect the remaining resource quantity of hard disk resources.

[0143] In some specific embodiments, as Figure 6 shown, grouping can be performed according to the resource characteristics of the host resource group information. For example, if a physical server has GPU card resources in its host resource group information and the quantity of GPU card resources is large, then this physical server will be added to the GPU Namespace, and the change of GPU resources will be monitored in the GPU Namespace. Another physical server has Disk resources in its host resource group information and the Disk resources are large, then this physical server will be added to the Disk Namespace, and the change of hard disk resources will be monitored in the Disk Namespace. Generally, the host resource group information of a physical server will only belong to one Namespace, and there is a Monitor instance in each Namespace. The Monitor instance is used to monitor the specified resources on the host resource group information, that is, the Monitor instance in each Namespace only monitors the resource change of one specified type of resource. Generally, the resource change can be the remaining resource quantity. During the working process of the Monitor instance, it will scan the specified resources on the host resource group information in each Namespace at every preset time period, such as 10s, 1 minute, 5 minutes, etc.

[0144] In this embodiment, by grouping according to the resource types and resource amounts of the available resources indicated by the host resource group information of the physical servers that have been put on the line, and monitoring the remaining resource amounts of the resources of the specified resource type of each group of physical servers, it is possible to ensure that only the remaining resource amounts of the resources of the specified resource type of each group of physical servers are monitored, and the specified resources monitored by the physical servers between different groups are different, thereby ensuring that there will be no logical confusion when monitoring resources and resource allocation, and improving the utilization rate of resources.

[0145] In one embodiment, the requested resource group information further indicates the cloud server type of the cloud server to be created. The determination of the grid resource group information adapted to the resource group information includes:

[0146] Determine the grid resource group information adapted to the cloud server type according to the cloud server type;

[0147] From the grid resource group information adapted to the cloud server type, determine the grid resource group information adapted to the resources required to create the cloud server;

[0148] Determine the grid resource group information adapted to the requested resource group information according to the grid resource group information adapted to the resources required to create the cloud server.

[0149] Among them, the cloud service type can be the model of the cloud server, such as model A, model B, etc.

[0150] Specifically, the grid resource group information also indicates the cloud server type. In generating the grid resource group information, the cloud server type corresponding to the available resources indicated by the grid resource group information can be specified in the grid resource group information. During the process of creating a cloud server, first determine the cloud server type of the cloud server to be created indicated by the requested resource group information. Then, according to the cloud server type, determine the grid resource group information adapted to the cloud server type. Furthermore, from the grid resource group information adapted to the cloud server type, determine again the grid resource group information adapted to the resources required to create the cloud server, so as to determine the final grid resource group information adapted to the requested resource group information.

[0151] For example, the available resources indicated by the grid resource group information are 10-core CPUs, 16G of memory, 60G of hard disk, and 5-core GPUs, and the cloud service type is S1 type. If the cloud server type of the cloud server to be created indicated by the request resource group information is S2 type, then in this case, since the cloud server types do not match, it can be directly determined that the grid resource group information does not match the request resource group information. If the cloud server type of the cloud server to be created indicated by the request resource group information is S1 type, then at this time, it can be determined that the cloud server types match. Subsequently, according to the resources required by the cloud server in the request resource group information, a match is made with the available resources indicated by this grid resource group information. If the resources required by the cloud server in the request resource group information are 10-core CPUs, 16G of memory, 30G of hard disk, and 3-core GPUs, then it can be determined that this grid resource group information matches the request resource group information.

[0152] In this embodiment, when creating the grid resource group information, the cloud server type can be created in the grid resource group information. Subsequently, during the process of determining the grid resource group information that matches the request resource group information, a preliminary screening is first performed using the cloud server type, which can quickly delete the grid resource group information that does not meet the requirements and improve the efficiency of cloud server creation.

[0153] In one embodiment, the determining of the host resource group information that matches the request resource group information further includes:

[0154] Determine the physical server type that matches the cloud server type according to the cloud server type.

[0155] Determine the host resource group information that matches the physical server type according to the physical server type.

[0156] From the host resource group information that matches the cloud server type, determine the host resource group information that matches the resources required to create the cloud server;

[0157] Determine the host resource group information that matches the request resource group information according to the host resource group information that matches the resources required to create the cloud server.

[0158] Specifically, usually, when creating host resource group information, since the host resource group information is created based on the available resources in the physical server, the type of the physical server can also be indicated in the host resource group information. During the process of creating cloud services, not all different types of cloud servers can be adapted to all physical servers. For example, cloud servers of type A require GPU resources, but in some physical servers, such as physical servers of type X, there are no GPU resources. At this time, cloud servers of type A cannot be created in physical servers of type X. Therefore, in order to improve the creation speed of cloud servers, according to the types of different cloud servers, the types of physical servers that are adapted to this type can be determined. Thus, a mapping table containing the mapping relationship between the cloud server type and the type of physical server is generated. Furthermore, according to the cloud server type of the cloud server to be created indicated by the requested resource group information, the type of physical server that is adapted to the cloud server type is matched from the mapping table. Thus, filtering is performed according to the physical server type, and according to the physical server type, the host resource group information that is adapted to the physical server type is determined. Furthermore, from the host resource group information that is adapted to the physical server type, the host resource group information that is adapted to the resources required to create the cloud server is determined again, so as to determine the final host resource group information that is adapted to the requested resource group information.

[0159] In this embodiment, by determining the type of physical server that is adapted to the cloud server type according to the cloud server type, and then using the physical server type for filtering, the host resource group information that does not meet the requirements can be quickly deleted, improving the creation efficiency of cloud servers.

[0160] In one embodiment, as Figure 7 shown, the embodiments of the present disclosure also provide another cloud server deployment method, which is executed by a cloud server management system. Further, it can be executed by a scheduler, a resource monitor, and a migrator in the cloud server management system. It can be applied to scenarios where users need to deploy or manage cloud servers. The method includes: a cloud server creation process, a cloud server deletion process, and a cloud server migration process (including the offline of physical servers, the deletion of grid resource group information, etc.).

[0161] As Figure 8 shown, the cloud server creation process includes:

[0162] S402, when the scheduler detects that a physical server is online, the scheduler creates host resource group information in the resource abstraction layer of the scheduler according to the resource information of the physical server.

[0163] S404, the resource monitor groups the physical servers that have been launched according to the resource types and resource quantities of the available resources indicated by the host resource group information of the launched physical servers, and obtains multiple groups of physical servers.

[0164] S406, the resource monitor monitors the remaining resource quantity of the resources of the specified resource type for each group of physical servers, where the specified resource types monitored in each group of physical servers are different.

[0165] S408, when the resource monitor detects that the remaining resource quantity of the resources of the specified resource type in a physical server is lower than the resource quantity threshold, it sends a request for creating grid resource group information to the scheduler.

[0166] S410, the scheduler divides the available resources of a preset resource quantity from the available resources indicated by the host resource group information, and generates grid resource group information for this physical server at the resource abstraction layer for the available resources of the preset resource quantity.

[0167] S412, in response to a cloud server creation request, the scheduler determines the requested resource group information indicated by the cloud server creation request, and the requested resource group information indicates the resources required to create a cloud server and the cloud server type of the cloud server to be created.

[0168] S414, the scheduler determines the grid resource group information suitable for the cloud server type according to the cloud server type. From the grid resource group information suitable for the cloud server type, the scheduler determines the grid resource group information suitable for the resources required to create a cloud server.

[0169] S416, the scheduler determines the grid resource group information suitable for the requested resource group information according to the grid resource group information suitable for the resources required to create a cloud server. The scheduler creates a cloud server on the first physical server indicated by the grid resource group information, using the available resources indicated by the grid resource group information.

[0170] S418, in response to successfully creating a cloud server on the first physical server, the scheduler creates a first binding relationship between the grid resource group information and the cloud server, and the first binding relationship indicates that the available resources indicated by the grid resource group information have been used to create the cloud server.

[0171] S420, when there is no grid resource group information suitable for the requested resource group information, the scheduler determines the physical server type suitable for the cloud server type according to the cloud server type.

[0172] S422, the scheduler determines the host resource group information suitable for the physical server type according to the physical server type. From the host resource group information suitable for the cloud server type, the scheduler determines the host resource group information suitable for the resources required to create a cloud server.

[0173] S424, the scheduler determines the host resource group information that matches the requested resource group information according to the host resource group information for the resources required to create a cloud server as adapted. The scheduler deducts, from the available resources indicated by the host resource group information, the resources required to create the cloud server indicated by the requested resource group information;

[0174] S426, the scheduler creates a cloud server on the second physical server indicated by the host resource group information, using the available resources indicated by the host resource group information.

[0175] S428, in response to successfully creating a cloud server on the second physical server, the scheduler creates a second binding relationship between the host resource group information and the cloud server, where the second binding relationship indicates that the available resources indicated by the host resource group information have been used to create the cloud server.

[0176] As Figure 9 shown, the deletion process of a cloud server includes:

[0177] S502, in response to a cloud server deletion request, when the binding relationship of the cloud server is the first binding relationship, the scheduler releases the first binding relationship and deletes the cloud server on the first physical server.

[0178] S504, in response to a cloud server deletion request, when the binding relationship of the cloud server is the second binding relationship, the scheduler adds back the deducted available resources to the available resources indicated by the host resource group information;

[0179] S506, the scheduler releases the second binding relationship and deletes the cloud server on the second physical server.

[0180] As Figure 10 shown, the migration process of a cloud server includes:

[0181] S602, in response to a physical server offline trigger event, when there is a cloud server created on the physical server, the migrator issues a migration instruction to the scheduler.

[0182] Alternatively, S604, when the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be not less than the resource quantity threshold, and the available resources indicated by the grid resource group information on the first physical server have been used to create a cloud server, the migrator issues a migration instruction to the scheduler.

[0183] Further, when the resource monitor detects that the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be not lower than the resource quantity threshold, and the resource monitor determines through the scheduler that the available resources indicated by the grid resource group information on the first physical server have been used to create a cloud server. The resource monitor needs to send a cloud server migration request to the migrator. The migrator will asynchronously convert the request into a Task task, store it in the internal storage unit of the migrator module, and at the same time return a Task Id generated for the Task task to the resource monitor. The resource monitor needs to call the task query interface provided by the migrator to query the task execution result through the Task Id. The built-in asynchronous thread of the migrator will continuously read the Task task from the internal storage unit, issue a migration instruction to the scheduler. Then call the scheduler to initiate a cloud server migration operation.

[0184] S606, the scheduler determines the grid resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server.

[0185] S608, the scheduler migrates the cloud server to the physical server indicated by the grid resource group information adapted to the cloud server resources, and the physical server indicated by the grid resource group information adapted to the cloud server resources is different from the physical server where the current cloud server is located.

[0186] Or, S610, when there is no grid resource group information adapted to the cloud server resources, the scheduler determines the host resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server.

[0187] S612, the scheduler migrates the cloud server to the physical server indicated by the host resource group information adapted to the cloud server resources, and the physical server indicated by the host resource group information adapted to the cloud server resources is different from the physical server where the cloud server is located.

[0188] S614, after the cloud server is migrated, the scheduler deletes the grid resource group information of the first physical server.

[0189] S616, after the cloud server is migrated, the scheduler deletes the host resource group information and grid resource group information of the physical server to be taken offline, and takes offline the physical server.

[0190] S618, in response to the offline trigger event of the physical server, when no cloud server is created in the physical server, the scheduler deletes the host resource group information and grid resource group information of the physical server to be taken offline, and takes offline the physical server.

[0191] As Figure 11As shown in the figure, the scheduler in the cloud server management system serves as the total entry point for resource operations of physical servers. All resource-related operations (such as physical server online, physical server offline, cloud server creation, cloud server migration, cloud server deletion, creation of host resource group information, creation of grid resource group information, deletion of host resource group information and grid resource group information, etc.) must pass through the scheduler, and then the scheduler operates on the resource abstraction layer. This can ensure the consistency of resource data, avoid problems of disordered resource data, and thus effectively improve the utilization rate of resources.

[0192] As Figure 12 shown in the figure, a specific embodiment is used to illustrate the process of deploying a cloud server. In response to a cloud service creation request, the cloud server creation request is parsed to obtain request resource group information, and the request resource group information is parsed to determine the resources required to create the cloud server indicated by the request resource group information and the cloud server type of the cloud server to be created. First, the grid resource group information is searched according to the cloud server type. If the search is successful, the grid resource group information is continued to be searched according to the resources required to create the cloud server. If the search is successful, it is first determined whether there is no first binding relationship in the found grid resource group information. If not, the request resource group information is converted into a VM Allocation object. A first binding relationship is established between the found grid resource group information and the VM Allocation object, and then a virtual machine is generated using the available resources indicated by the grid resource group information, and the cloud server is deployed using the virtual machine.

[0193] When the search fails or there is a first binding relationship in the found grid resource group information, the physical server type suitable for the cloud server type can be determined according to the cloud server type. Then, the host resource group information is searched according to the physical server type. If the search is successful, the host resource group information is continued to be searched according to the resources required to create the cloud server. If the search is successful, the available resources indicated by the host resource group are deducted, and the request resource group information is converted into a VM Allocation object. A second binding relationship is established between the found host resource group information and the VM Allocation object, and then a virtual machine is generated using the available resources indicated by the host resource group information, and the cloud server is deployed using the virtual machine.

[0194] This application also provides some application scenarios, which apply the above cloud server deployment method. Specifically, the cloud server deployment method can also be applied to scenarios where cloud servers are used for data storage and data analysis in intelligent transportation and assisted driving, or scenarios where cloud servers are used to perform statistics and control on various function operations in the fields of intelligent transportation and assisted driving. It can also be applied to scenarios in the field of artificial intelligence where computing resources in cloud servers are used to train neural network models and apply neural network models, etc.

[0195] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0196] Based on the same inventive concept, an embodiment of the present disclosure further provides a cloud server deployment device for implementing the above-mentioned cloud server deployment method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the cloud server deployment device provided below can refer to the limitations on the cloud server deployment method in the above text, and will not be repeated here.

[0197] In one embodiment, as Figure 13 shown, a cloud server deployment device 1300 is provided, including: a resource determination module 1302, a resource adaptation module 1304, and a cloud server creation module 1306, where:

[0198] The resource determination module 1302 is configured to determine the requested resource group information indicated by the cloud server creation request in response to the cloud server creation request, and the requested resource group information indicates the resources required to create the cloud server.

[0199] The resource adaptation module 1304 determines the grid resource group information adapted to the requested resource group information. The grid resource group information indicates the grid resources pre-partitioned on the first physical server, and the grid resources indicate the available resources on the first physical server. The remaining resource quantity of the resources of the specified resource type on the first physical server is lower than the resource quantity threshold, and the specified resource type is different from the available resource type of the available resources.

[0200] The cloud server creation module 1306 is configured to create the cloud server on the first physical server indicated by the grid resource group information by using the available resources indicated by the grid resource group information.

[0201] In one embodiment of the device, the device also includes: a binding relationship creation module, which is used to create a first binding relationship between the grid resource group information and the cloud server in response to the successful creation of the cloud server on the first physical server, and the first binding relationship indicates that the available resources indicated by the grid resource group information have been used to create the cloud server.

[0202] The binding relationship releasing module is used to respond to a cloud server deletion request, and when the binding relationship of the cloud server is the first binding relationship, release the first binding relationship and delete the cloud server on the first physical server.

[0203] In one embodiment of the device, the device also includes: a migration module, which is used to migrate the cloud server to a physical server that is adapted to the resources required by the cloud server when the remaining resource quantity of resources of the specified resource type on the first physical server is updated to no less than the resource quantity threshold and the available resources indicated by the grid resource group information on the first physical server have been used to create the cloud server.

[0204] An information deletion module is configured to delete the grid resource group information of the first physical server after the cloud server is migrated.

[0205] In one embodiment of the device, the information deletion module is further used to delete the indicated unused grid resource group information of available resources when the remaining resource quantity of the specified resource type on the first physical server is updated to no less than the resource quantity threshold, and the first physical server has the indicated unused grid resource group information of available resources.

[0206] In one embodiment of the device, the resource adaptation module 1304 is also used to determine the host resource group information that is adapted to the requested resource group information when there is no grid resource group information that is adapted to the requested resource group information, the host resource group information indicates the host resources determined from the second physical server, and the remaining resource quantity of the resources of the specified resource type on the second physical server is not less than the resource quantity threshold.

[0207] The device further includes: an information deduction module, configured to deduct the resources required for creating the cloud server indicated by the request resource group information from the available resources indicated by the host resource group information.

[0208] The cloud server creation module 1306 is further configured to create the cloud server on the second physical server indicated by the host resource group information using the available resources indicated by the host resource group information.

[0209] In one embodiment of the device, the binding relationship creation module is further used to create a second binding relationship between the host resource group information and the cloud server in response to successful creation of the cloud server on the second physical server, wherein the second binding relationship indicates that the available resources indicated by the host resource group information have been used to create the cloud server.

[0210] In one embodiment of the device, the binding relationship release module is also used to respond to a cloud server deletion request. When the binding relationship of the cloud server is a second binding relationship, the deducted available resources are added to the available resources indicated by the host resource group information; the second binding relationship is released, and the cloud server on the second physical server is deleted.

[0211] In one embodiment of the device, the device also includes: a resource group information creation module, which is used to create host resource group information based on the resource information of the physical server when it is detected that the physical server is online; when the remaining resource quantity of resources of the specified resource type in the physical server is lower than the resource quantity threshold, divide the available resources of a preset resource quantity from the available resources indicated by the host resource group information; and generate grid resource group information for the available resources of the preset resource quantity.

[0212] In one embodiment of the device, the device also includes: an offline processing module, which is used to respond to the offline trigger event of the physical server, and when a cloud server is created in the physical server, migrate the cloud server to a physical server that is adapted to the resources required by the cloud server; after the cloud server is migrated, delete the host resource group information and grid resource group information of the physical server that needs to be offline, and take the physical server offline.

[0213] In one embodiment of the device, the offline processing module is further used to respond to the offline trigger event of the physical server, when no cloud server is created in the physical server, delete the host resource group information and grid resource group information of the physical server that needs to be offline, and take the physical server offline.

[0214] In one embodiment of the device, the migration module is further used to determine grid resource group information adapted to the cloud server resources based on the cloud server resources occupied by the cloud server; and migrate the cloud server to a physical server indicated by the grid resource group information adapted to the cloud server resources, wherein the physical server indicated by the grid resource group information adapted to the cloud server resources is different from the physical server where the cloud server is currently located.

[0215] In one embodiment of the device, the migration module is further configured to, when there is no grid resource group information adapted to the cloud server resources, determine host resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server; and migrate the cloud server to the physical server indicated by the host resource group information adapted to the cloud server resources, where the physical server indicated by the host resource group information adapted to the cloud server resources is different from the physical server where the cloud server is located.

[0216] In one embodiment of the device, the device further includes: a grouping module, configured to group the online physical servers according to the resource type and resource amount of the available resources indicated by the host resource group information of the online physical servers, so as to obtain multiple groups of physical servers.

[0217] A monitoring module, configured to monitor the remaining resource quantity of the resources of the specified resource type of each group of physical servers, where the specified resource types monitored in each group of physical servers are different.

[0218] In one embodiment of the device, the requested resource group information further indicates the cloud server type of the cloud server to be created; the resource adaptation module 1304 is further configured to determine grid resource group information adapted to the cloud server type according to the cloud server type; determine grid resource group information adapted to the resources required for creating the cloud server from the grid resource group information adapted to the cloud server type; and determine grid resource group information adapted to the requested resource group information according to the grid resource group information adapted to the resources required for creating the cloud server.

[0219] In one embodiment of the device, the resource adaptation module 1304 is further configured to determine a physical server type adapted to the cloud server type according to the cloud server type; determine host resource group information adapted to the physical server type according to the physical server type; determine host resource group information adapted to the resources required for creating the cloud server from the host resource group information adapted to the cloud server type; and determine host resource group information adapted to the requested resource group information according to the host resource group information adapted to the resources required for creating the cloud server.

[0220] Each module in the above cloud server deployment device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or be independent of it, or can be stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0221] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structural diagram may be as shown in Figure 14 . The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store request resource group information and the remaining resource quantities of resources of a specified resource type. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a cloud server deployment method.

[0222] Those skilled in the art can understand that Figure 14 the structure shown in is only a block diagram of some structures related to the solution of the present disclosure, and does not constitute a limitation on the computer device to which the solution of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0223] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, it implements the steps in any of the above method embodiments.

[0224] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps in any of the above method embodiments.

[0225] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, it implements the steps in any of the above method embodiments.

[0226] It should be noted that the request resource group information, various resource information in the physical server, the first binding relationship, the second binding relationship, etc. involved in the present disclosure are all information and data authorized by the user or fully authorized by all parties. And the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards.

[0227] Those of ordinary skill in the art can understand that all or part of the processes in the above-described embodiment methods can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-described method embodiments. Among them, any reference to a memory, database, or other medium used in the various embodiments provided by the present disclosure can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAMs), magnetoresistive random access memories (MRAMs), ferroelectric random access memories (FRAMs), phase change memories (PCMs), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the various embodiments provided by the present disclosure can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the various embodiments provided by the present disclosure can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0228] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0229] The above-described embodiments only represent several implementation manners of the present disclosure. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present disclosure. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present disclosure, several modifications and improvements can still be made, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the appended claims.

Claims

1. A cloud server deployment method, characterized in that: The method comprises: In response to a cloud server creation request, determining requested resource group information indicated by the cloud server creation request, the requested resource group information indicating resources required to create the cloud server; determining grid resource group information adapted to the requested resource group information, the grid resource group information indicating grid resources pre-allocated from a first physical server, the grid resources indicating available resources on the first physical server, a remaining resource quantity of resources of a specified resource type on the first physical server being lower than a resource quantity threshold, the specified resource type being different from an available resource type of the available resources; The cloud server is created on the first physical server indicated by the grid resource group information using the available resources indicated by the grid resource group information.

2. The method according to claim 1, characterized in that The method further comprises: In response to successful creation of the cloud server on the first physical server, creating a first binding relationship between the grid resource group information and the cloud server, the first binding relationship indicating that the available resources indicated by the grid resource group information have been used to create the cloud server; In response to a cloud server deletion request, when the binding relationship of the cloud server is a first binding relationship, the first binding relationship is released, and the cloud server on the first physical server is deleted.

3. The method according to claim 1, characterized in that The method further comprises: When the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be no less than the resource quantity threshold, and the available resources indicated by the grid resource group information on the first physical server have been used to create the cloud server, migrating the cloud server to a physical server adapted to the resources required by the cloud server; After the cloud server is migrated, the grid resource group information of the first physical server is deleted.

4. The method according to claim 1, wherein The method further comprises: When the remaining resource quantity of the resources of the specified resource type on the first physical server is updated to be no less than the resource quantity threshold, and the first physical server has the indicated unused grid resource group information of available resources, the indicated unused grid resource group information of available resources is deleted.

5. The method according to claim 1, wherein The method further comprises: When there is no grid resource group information adapted to the requested resource group information, determining host resource group information adapted to the requested resource group information, the host resource group information indicating host resources determined from the second physical server, and a remaining resource quantity of resources of a specified resource type on the second physical server is not less than the resource quantity threshold; Deducting the resources required for creating the cloud server indicated by the request resource group information from the available resources indicated by the host resource group information; The cloud server is created on the second physical server indicated by the host resource group information using the available resources indicated by the host resource group information.

6. The method according to claim 5, characterized in that The method further comprises: In response to the successful creation of the cloud server on the second physical server, a second binding relationship is created between the host resource group information and the cloud server, wherein the second binding relationship indicates that the available resources indicated by the host resource group information have been used to create the cloud server.

7. The method according to claim 6, characterized in that The method further comprises: In response to a cloud server deletion request, when the binding relationship of the cloud server is a second binding relationship, Adding the deducted available resources to the available resources indicated by the host resource group information; Release the second binding relationship and delete the cloud server on the second physical server.

8. The method according to claim 1, characterized in that The method further comprises: When a physical server is detected to be online, host resource group information is created according to resource information of the physical server; When the remaining resource quantity of the resource of the specified resource type in the physical server is lower than the resource quantity threshold, dividing the available resources of the preset resource quantity from the available resources indicated by the host resource group information; Grid resource group information is generated for the preset amount of available resources.

9. The method according to any one of claims 5 to 8, characterized in that The method further comprises: In response to a physical server offline triggering event, when a cloud server is created in the physical server, migrating the cloud server to a physical server adapted to resources required by the cloud server; After the cloud server is migrated, the host resource group information and grid resource group information of the physical server that needs to be taken offline are deleted, and the physical server is taken offline.

10. The method according to any one of claims 5 to 8, characterized in that The method further comprises: In response to the offline triggering event of the physical server, when no cloud server is created in the physical server, the host resource group information and grid resource group information of the physical server that needs to be offline are deleted, and the physical server is offline.

11. The method according to claim 3 or 9, characterized in that Migrating the cloud server to a physical server adapted to the resources required by the cloud server includes: Determining grid resource group information adapted to the cloud server resources based on the cloud server resources occupied by the cloud server; The cloud server is migrated to a physical server indicated by grid resource group information adapted to cloud server resources, where the physical server indicated by the grid resource group information adapted to cloud server resources is different from the physical server where the cloud server is currently located.

12. The method according to claim 11, characterized in that The method further comprises: When there is no grid resource group information adapted to the cloud server resources, determining host resource group information adapted to the cloud server resources according to the cloud server resources occupied by the cloud server; The cloud server is migrated to a physical server indicated by host resource group information adapted to cloud server resources, where the physical server indicated by the host resource group information adapted to cloud server resources is different from the physical server where the cloud server is located.

13. The method according to any one of claims 5 to 8, characterized in that The method further comprises: For the online physical servers, group them according to the resource types and resource amounts of the available resources indicated by the host resource group information of the online physical servers to obtain multiple groups of physical servers; The remaining resource quantity of the resources of the specified resource type of each group of physical servers is monitored, wherein the specified resource type monitored in each group of physical servers is different.

14. The method according to any one of claims 1 to 8, characterized in that The requested resource group information further indicates the cloud server type of the cloud server that needs to be created; and determining grid resource group information that is adapted to the requested resource group information includes: Determining grid resource group information adapted to the cloud server type according to the cloud server type; Determining, from the grid resource group information adapted to the cloud server type, grid resource group information adapted to resources required to create the cloud server; Grid resource group information adapted to the requested resource group information is determined based on grid resource group information adapted to resources required for creating the cloud server.

15. A cloud server deployment device, characterized in that: The device comprises: a resource determination module, configured to determine, in response to a cloud server creation request, requested resource group information indicated by the cloud server creation request, wherein the requested resource group information indicates resources required to create the cloud server; a resource adaptation module, determining grid resource group information adapted to the requested resource group information, the grid resource group information indicating grid resources pre-allocated from a first physical server, the grid resources indicating available resources on the first physical server, a remaining resource quantity of resources of a specified resource type on the first physical server being lower than a resource quantity threshold, the specified resource type being different from an available resource type of the available resources; The cloud server creation module is configured to create the cloud server on the first physical server indicated by the grid resource group information using the available resources indicated by the grid resource group information.

16. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 14 are implemented.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 14 are implemented.

18. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 14 are implemented.

Images