CN115700481A

CN115700481A - Resource management method and related equipment of public cloud data center

Info

Publication number: CN115700481A
Application number: CN202110865855.0A
Authority: CN
Inventors: 郭雷; 柯晓棣; 胡昊然; 大卫·比加利; 冀司琪; 张建斌
Original assignee: Huawei Cloud Computing Technologies Co Ltd
Current assignee: Huawei Cloud Computing Technologies Co Ltd
Priority date: 2021-07-29
Filing date: 2021-07-29
Publication date: 2023-02-07

Abstract

The application provides a resource management method of a public cloud data center and related equipment, wherein the method comprises the following steps: the cloud management platform records resource pool information, wherein the resource pool information is used for recording resources occupied by virtual instances needing to be operated by each physical server in a hardware resource pool of a public cloud data center and instance types of the virtual instances needing to be operated by each physical server, the instance types comprise on-demand instance types and reserved instance types, and the cloud management platform centrally schedules the virtual instances of the reserved instance types to at least one physical server in the resource pool information so that the physical server related to the virtual instances of the reserved instance types before scheduling can be used by the virtual instances of the on-demand instance types. The method can reduce the resource fragments of the public cloud and improve the resource utilization rate of the public cloud under the condition of not influencing the use of the user.

Description

Resource management method and related equipment of public cloud data center

Technical Field

The application relates to the field of cloud computing, in particular to a resource management method and related equipment of a public cloud data center.

Background

With the rapid development of cloud computing Technology and various network infrastructures, traditional Internet Technology (IT) business architecture is gradually scheduled to public cloud, and more business applications are redesigned and used based on the public cloud architecture. Through the public cloud, a user can send a resource acquisition request to a cloud management platform of the public cloud, the cloud management platform performs screening analysis on resources in the hardware resource pool according to the resource request, finds out a proper physical server and performs resource deduction on the physical server, and creates a corresponding resource carrier, and the user can use different terminal devices to be connected with the resource carrier at any time and any place to perform service processing.

However, the physical server includes multiple resources, the number of each resource and the ratio between the resources are fixed, and the specification of the resource requested by the user is not fixed, for example, the ratio of the CPU to the memory of the physical server is 1. The remaining resources cannot be used for creating other resource carriers, and will be retained in the physical servers in the form of resource fragments, and for a public cloud having a large number of physical servers, the resource fragments on each physical server will cause resource waste of the public cloud, so that the resource utilization rate is low, and the operation cost of the public cloud is increased.

Disclosure of Invention

The application provides a resource management method and related equipment of a public cloud data center, which are used for solving the problem of resource waste of the public cloud due to resource fragments and reducing the operation cost of the public cloud.

In a first aspect, a resource management method for a public cloud data center is provided, where the method is applied to a cloud management platform of the public cloud data center, and the method may include the following steps: the cloud management platform records resource pool information, wherein the resource pool information is used for recording resources occupied by virtual instances needing to be operated by each physical server in a hardware resource pool of a public cloud data center and instance types of the virtual instances needing to be operated by each physical server, the instance types comprise on-demand instance types and reserved instance types, and the cloud management platform centrally schedules the virtual instances of the reserved instance types to at least one physical server in the resource pool information so that the physical server related to the virtual instances of the reserved instance types before scheduling can be used by the virtual instances of the on-demand instance types.

In a specific implementation, the on-demand instance type of the virtual instance refers to a virtual instance that can be used by a user after payment, and the reserved instance type refers to a virtual instance that cannot be used by the user within a reserved time after payment and can be used after the reserved time is over. For example, if the user pays for the size X virtual machine on 6/month and 1/day and chooses to reserve the time for 3 months, the user can use the size X virtual machine after 9/month and 1/day. It should be understood that there is a certain deviation between the prices of the reserved instances and the on-demand instances in the public cloud platform, for example, the price of the reserved instances is lower than that of the on-demand instances, and a user purchases a large number of reserved instances in advance under the condition that the resource demand is not tight, so that the cost can be reduced.

In a specific implementation, the virtual instance may be a virtual resource, such as a virtual machine and a container, or a physical resource, such as a physical server, or a service resource, such as a process, that may be provided by other public cloud data centers, which is not limited in this application.

In a specific implementation, the resource pool information may include resource usage and instance types of a plurality of physical servers, and the resource usage of each physical server includes usage of resources of different types, for example, the resource usage of the physical server 1 may include CPU resource usage, memory resource usage, cache resource usage, memory bandwidth usage, network bandwidth resource usage, disk capacity resource usage, GPU resource usage, and the like. The usage of each category of resource may also be described by using a plurality of parameter information, for example, the usage of the CPU resource may include the number of cores of the CPU, the model of the CPU, and the like, and the usage of the memory resource may further include the size of the memory, the remaining capacity of the memory, and the like, which are not illustrated herein.

In the method described in the first aspect, the cloud management platform may record resources that need to be occupied by the virtual instance that each physical server needs to run in the hardware resource pool of the public cloud data center and the instance type of the virtual instance that each physical server needs to run, and the system may perform centralized scheduling on the reserved instance recorded in the resource pool information, so that the physical server associated with the virtual instance of the reserved instance type before scheduling may be used by the virtual instance of the on-demand instance type, which may reduce resource fragments of the public cloud and improve resource utilization of the public cloud without affecting user usage.

In a possible implementation manner of the first aspect, the cloud management platform may receive a first virtual instance creation request, where the first virtual instance creation request includes an instance type of a first virtual instance, and the instance type of the first virtual instance is an on-demand instance type; the cloud management platform updates the resource pool information, the updated resource pool information comprises a record of a first virtual instance created by the first physical server and an instance type of the first virtual instance, and the cloud management platform sends a request for creating the first virtual instance to the first physical server.

Optionally, the cloud management platform receives a second virtual instance creation request, where the second virtual instance creation request includes an instance type of a second virtual instance, and the instance type of the second virtual instance is a reserved instance type; and the cloud management platform updates the resource pool information, wherein the updated resource pool information comprises a record of the second virtual instance created by the second physical server and the instance type of the second virtual instance.

In the implementation manner, if the instance type of the virtual instance is the reserved instance, the cloud management platform does not need to send the request for creating the virtual instance to the physical server for creating the virtual instance, and if the instance type of the virtual instance is the on-demand instance, the cloud management platform needs to send the request for creating the virtual instance to the physical server for creating the virtual instance. It should be understood that, since the reserved instance is not used by the user within the reserved time, the reserved instance may not be created on the physical server within the reserved time, which may avoid resource waste and improve the resource utilization rate of the public cloud.

In a possible implementation manner of the first aspect, when the cloud management platform schedules the reserved instances in the resource pool information, one reserved instance may be scheduled, a part of reserved instances may also be scheduled, and all reserved instances may also be scheduled, which is not specifically limited in this application. The reserved instances participating in the centralized scheduling at this time can be determined according to the number of the reserved instances and the number of resource shards, so that the number of the resource shards recorded in the resource pool information after the centralized scheduling is reduced, wherein the resource shards refer to resources which cannot be used for creating the virtual instances.

In the implementation manner, the cloud management platform can screen out the physical servers with the resource fragments and the reserved examples from the resource pool information, and then schedule the reserved examples to other physical servers capable of creating the reserved examples, so that the number of the resource fragments of the public cloud is reduced, and the operation cost of the public cloud is reduced.

In a possible implementation manner of the first aspect, before the cloud management platform centrally schedules the virtual instances of the reserved instance type to the at least one physical server in the resource pool information, a first number of remaining resources recorded in the resource pool information may be determined, the cloud management platform obtains, from the hardware resource pool, a second number of remaining resources of the plurality of physical servers in the hardware resource pool, and when a ratio between the first number and the second number is less than or equal to a first threshold, centrally schedules the virtual instances of the reserved instance type to the at least one physical server in the resource pool information.

The number of the remaining resources may refer to the number of standard virtual instances that can be created, the specification of the standard virtual instances is determined according to the historical virtual machine creation record of the public cloud, and the specification of the standard virtual instances may include the specification of one or more virtual instances. For example, it is determined that the issuing frequency of VM1 and VM2 is higher according to the history virtual machine creation record of the public cloud, then the total number of VM1 and VM2 that can be issued by the remaining resource may be determined according to the remaining resource recorded in the resource pool information, so as to obtain the first number, then the remaining resource of the plurality of physical servers in the hardware resource pool is obtained from the hardware resource pool, and the total number of VM1 and VM2 that can be issued by the remaining resource is determined, so as to obtain the second number, and if the ratio of the first number to the second number is smaller than or equal to the first threshold, the cloud management platform may perform centralized scheduling on the reserved instance in the resource pool information.

Optionally, the cloud management platform periodically determines the first amount of the remaining resources in the resource pool information record, and then performs centralized scheduling on the reserved instances, where the periodicity may be scheduling of the reserved instances at a fixed time interval, or scheduling of the reserved instances after receiving the virtual instance creation request for each M times, which is not limited in this application.

Optionally, the cloud management platform determines a first amount of the remaining resources recorded in the resource pool information in the case that the virtual instance is failed to be created, and then performs centralized scheduling on the reserved instances. The number of resource fragments in the resource pool information is reduced, the virtual instance creation request which is failed to be created before is retried to be processed, the use experience of a user is improved, and the resource utilization rate of the public cloud is improved.

In the above implementation manner, the first number is determined according to the resource pool information, the second number is determined according to the remaining resources of the plurality of physical servers in the hardware resource pool, and the resource pool information additionally records the creation record of the reserved instances, so that the first number is smaller than the second number, and the smaller the ratio of the first number to the second number is, the larger the number of resource fragments in the current resource pool information is, and therefore when the ratio of the first number to the second number is smaller than or equal to the second threshold, the cloud management platform schedules the reserved instances, and centrally schedules the virtual instances of the reserved instance types in the resource pool information to at least one physical server, so that the physical servers associated with the virtual instances of the reserved instance types before scheduling can be used by the virtual instances of the on-demand instance types, thereby avoiding resource waste and improving the resource utilization rate of the public cloud.

In a possible implementation manner of the first aspect, the reserved instance recorded in the resource pool information includes not only the instance specification, but also a reserved time of the reserved instance. When the reserved time of the reserved instance is about to expire, the cloud management platform may send a virtual instance creation request to the physical server where the reserved instance is recorded in the resource pool information, and update the resource pool information.

By the implementation mode, the situation that the user cannot use the reserved instance after the reserved time is over can be avoided, and the use experience of the user is improved.

In a second aspect, a cloud management platform is provided, which is applied to a cloud management platform of a public cloud data center, and includes: the system comprises a recording unit, a resource pool information recording unit and a resource management unit, wherein the recording unit is used for recording resource pool information, the resource pool information is used for recording resources occupied by virtual instances needing to be operated by each physical server in a hardware resource pool of a public cloud data center and instance types of the virtual instances needing to be operated by each physical server, and the instance types comprise on-demand instance types and reserved instance types; and the scheduling unit is used for scheduling the virtual instances of the reserved instance type to at least one physical server in the resource pool information in a centralized manner, so that the physical server related to the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type.

Implementing the cloud management platform described in the second aspect, the cloud management platform records resources that need to be occupied by a virtual instance that each physical server needs to run in a hardware resource pool of a public cloud data center and an instance type of the virtual instance that each physical server needs to run, and the resource pool information recorded by the system includes creation information of a reserved instance, but the physical server of the hardware resource pool does not create the reserved instance, thereby avoiding resource waste caused by the fact that the reserved instance is not used by a user within the reserved time; meanwhile, the system can also perform centralized scheduling on the reserved instances recorded in the resource pool information, so that the physical server associated with the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type, and therefore resource fragments of the public cloud can be reduced and the resource utilization rate of the public cloud can be improved under the condition that the use of users is not influenced.

Any implementation manner of the second aspect or the second aspect is a cloud management platform implementation corresponding to any implementation manner of the first aspect or the first aspect, and the description in any implementation manner of the first aspect or the first aspect is applicable to any implementation manner of the second aspect or the second aspect, and is not described herein again.

In a third aspect, a computer program product is provided which, when run on a computer, causes the computer to perform the method of the above aspects.

In a fourth aspect, a computer-readable storage medium having stored therein instructions, which when executed on a computer, cause the computer to perform the method of the above aspects is provided.

In a fifth aspect, a computing device is provided that includes a processor configured to perform the method described in the above aspects.

In a sixth aspect, a computer-readable storage medium having stored thereon instructions for implementing the method described in the above aspects when run on a processor.

The present application may further combine to provide more implementation manners on the basis of the implementation manners provided by the above aspects.

Drawings

Fig. 1 is a schematic structural diagram of a public cloud system provided in the present application;

FIG. 2 is a schematic diagram of an architecture of a cloud management platform provided herein;

FIG. 3 is an exemplary representation of resource pool information provided herein;

fig. 4 is a flowchart illustrating centralized scheduling of reserved instances by resource pool information according to the present application;

FIG. 5 is a flowchart illustrating steps of a method for resource management in a public cloud data center according to the present disclosure;

fig. 6 is a schematic structural diagram of a computing device provided in the present application.

Detailed Description

In order to facilitate understanding of the technical solution of the present invention, first, an application scenario of the "public cloud" related to the present invention is explained.

Public cloud refers to computing services provided by third party providers to users over a public network. Under the public cloud architecture, a user can rent resources, such as a virtual machine, from the public cloud after paying a certain fee according to the service application requirement of the user, and after successful payment, the user can use different terminal equipment to be connected with the public cloud at any time and any place to achieve the purposes of data access, operation and the like.

Fig. 1 is a schematic structural diagram of a public cloud system provided in the present application, where the system may include a terminal device 110 and a public cloud data center 130, and the terminal device 110 and the public cloud data center 130 are connected through a network 120. The network 120 may be a public network such as the internet (internet), among others.

The terminal device 110 may be an electronic device with streaming media playing capability, such as a personal computer, a tablet computer, a mobile notebook, a smart phone, a handheld processing device, a virtual reality device, a wearable device, an all-in-one handheld device, and the like, and fig. 2 illustrates an example in which the terminal device 110 is a personal computer, but this application is not limited to this.

The public cloud data center 130 may provide shared resource services for users, where the shared resource services may include an OBS service, an SFS service, a cloud phone (cloud phone) service, a Content Delivery Network (CDN), a cloud backup service (CBR), a Data Admin Service (DAS), and the like, and the application does not limit the types of shared resource services that the public cloud data center 130 may provide.

For example, the data center 130 of the public cloud may include a cloud management platform 131 and a hardware resource pool 132, and it should be understood that the division manner shown in fig. 2 is for illustration, and the data center 130 of the public cloud may also be divided in other manners, and the present application does not limit the division manner of the data center 130 of the public cloud.

The cloud management platform 131 may be implemented by a general physical server, for example, an ARM server or an X86 server, or may be a Virtual Machine (VM) implemented by combining a Network Function Virtualization (NFV) technology, and the cloud management platform 131 may also be a virtual machine or a physical server in the hardware resource pool 132, which is not limited in this application.

The hardware resource pool 132 may include at least one physical server (fig. 2 exemplifies that the resource pool includes a physical server 1, a physical server 2, a physical server 3, and a physical server 4), wherein the physical server may be a general physical server, such as an ARM server or an X86 server, and the present application is not limited in particular. The physical servers in the hardware resource pool 132 may communicate with other physical servers or the cloud management platform 131 through an internal network. Each physical server at least includes hardware resources (for example, the physical server 1 includes the hardware resources 1, and the physical server 2 includes the hardware resources 2) and an operating system (for example, the operating system 1 and the operating system 2), some physical servers may further include multiple virtual instances, where the virtual instances may be containers (for example, the physical server 1 includes the containers 11 and 12) or virtual machines (for example, the virtual machines 21 and 22), and the multiple virtual instances in the physical server may share the operating system and the hardware resources in the physical server.

The hardware resources (e.g., hardware resource 1 and hardware resource 2) may include various available hardware resources of the physical server, such as processor 1, memory 1, PCI device 1, and the like, and may also include other hardware resources that may be needed by the user, which is not specifically limited in this application. The operating systems (such as the operating system 1 and the operating system 2) may be containers, virtual machines, or physical server-suitable operating systems, such as an Android operating system, a windows operating system, a linux operating system, and the like, and the present application is not limited in particular. It should be noted that the operating system may be an official complete operating system, or may be an operating system obtained by modifying an individual driver module of the official complete operating system in order to adapt to the operation mode of the server, and the application is not particularly limited. The number of physical servers, the number of virtual machines, the number of containers, and the types and numbers of hardware resources shown in fig. 2 are only for illustration, and the present application is not particularly limited.

A user may send a virtual instance creation request to the cloud management platform 131 through the terminal device 110, where the virtual instance creation request carries specification information of a virtual instance that the user requests to create, and the cloud management platform 131 may determine, according to a resource idle condition of the hardware resource pool 132 and the specification information, a physical server used for creating the virtual instance, and then send the specification information to the physical server to request for creating a corresponding virtual instance.

The virtual instance may be a virtual resource, such as a virtual machine and a container, a physical resource, such as a physical server, or a service resource, such as a process, available from other public cloud data centers, which is not limited in this application.

The specification information of the virtual instance may include specification information under multiple resource categories, for example, the specification information of the first virtual instance may include CPU resource specification information, memory resource specification information, cache resource specification information, memory bandwidth resource specification information, network bandwidth resource specification information, disk capacity resource specification information, GPU resource specification information, and the like. The specification information under each resource category may further include a plurality of parameter information, for example, the CPU resource specification may include a CPU core number, a CPU model, and the like, and the memory resource specification may further include a memory size, a memory remaining capacity, and the like, which are not illustrated here. Moreover, the specification information may further include a constraint condition, for example, a first virtual instance created by a user request, where a CPU resource and a memory resource of the first virtual instance are on the same physical server, and the CPU resource and a GPU resource are on the same non-uniform memory access (NUMA) architecture, which is not specifically limited in this application.

It should be noted that, the cloud management platform 131 determines, according to a virtual instance creation request sent by a user, a physical server used for creating the virtual instance, and sends the specification information to the physical server, where the physical server may create an instance with a corresponding specification according to the specification information, where the virtual instance may specifically be a container, a virtual machine, a physical server, or a process, and then sends a connection manner of the virtual instance, such as a connection password, to the terminal device 110 of the user, and the user may use the terminal device 110 to connect with the virtual instance at any time and at any place according to the connection password, run a workload on the virtual instance, and achieve the purposes of data access, operation, and the like. It is to be understood that the above description is illustrative, and that the present application is not limited to this description.

However, because the number of resources on each physical server is relatively fixed, and the specification information of the virtual instance requested to be obtained by the user is not fixed, when resource deduction is performed, after partial resources on partial physical servers are deducted, the remaining part of resources cannot be used for creating other resource carriers, and will be retained in the physical servers in the form of resource fragments.

For example, if the CPU and memory allocation ratio of the physical server 1 is 1, and the memory allocation ratio of the virtual instance created by the user request is 1; similarly, if the memory ratio of the virtual instance created by the user request is 2. The foregoing examples are illustrative, and the present application is not limited to these examples.

In summary, the physical servers in the public cloud generally have resource fragments, and for the public cloud having a large number of physical servers, the resource fragments may cause resource waste of the public cloud, and increase the operation cost of the public cloud.

In order to solve the problems of resource fragment of a physical server in a public cloud, which causes resource waste of the public cloud and high operation cost, a cloud management platform in the resource management system records resources required to be occupied by a virtual instance required to be operated by each physical server in a hardware resource pool of a public cloud data center and instance types of the virtual instance required to be operated by each physical server, wherein the instance types comprise an on-demand instance type and a reserved instance type, the on-demand instance type is an instance used immediately after a user purchases the on-demand instance type is an instance which cannot be used by the user within a reserved time, and the on-reserve instance type is an instance which cannot be used by the user after the reserved time elapses. The resource pool information recorded by the system comprises the creation information of the reserved instance, but the physical server in the hardware resource pool does not create the reserved instance, so that resource waste caused by the fact that a user does not use the resource within the reserved time after the reserved instance is created on the physical server is avoided, the system can centrally schedule the virtual instance of the reserved instance type to at least one physical server in the resource pool information, the physical server related to the virtual instance of the reserved instance type before the scheduling can be used by the virtual instance of the on-demand instance type, the resource management system can reduce resource fragments of the public cloud under the condition that the use of the user is not influenced, the resource waste of the public cloud is avoided, the resource utilization rate of the public cloud is improved, and the operation cost of the public cloud is reduced.

Fig. 2 is a schematic diagram of an architecture of a cloud management platform provided in the present application, and as shown in fig. 2, the architecture may include a terminal device 110, a cloud management platform 131, and a hardware resource pool 132, where descriptions and connection manners of the terminal device 110, the cloud management platform 131, and the hardware resource pool 132 may refer to the embodiment in fig. 1, and are not repeated here.

The cloud management platform 131 may be further divided into a plurality of unit modules, such as the scheduling unit 1311, the receiving unit 1312, the recording unit 1313, and the sending unit 1314 shown in fig. 2, it should be understood that fig. 2 is an exemplary division manner, and each module may be a software module, a hardware module, or a part of a software module and a part of a hardware module, which is not limited in this application.

The receiving unit 1312 is configured to receive a virtual instance creation request sent by a user, where the virtual instance creation request includes specification information of a virtual instance and an instance type, and the instance type includes an on-demand instance type and a reserved instance type. The on-demand instance type refers to a virtual instance which can be used by a user after payment, the reserved instance type refers to a virtual instance which can be used after the reserved time passes, for example, the user pays for purchasing a virtual machine with the specification of X on 6 months and 1 day, and selects the reserved time to be 3 months, so that the user can use the virtual machine with the specification of X after 9 months and 1 day. It should be understood that there is a certain deviation between the prices of the reserved instances and the on-demand instances in the public cloud platform, for example, the price of the reserved instances is lower than that of the on-demand instances, and a user purchases a large number of reserved instances in advance under the condition that the resource demand is not tight, so that the cost can be reduced. The detailed description of the specification information of the virtual instance may refer to the embodiment in fig. 1, and is not repeated here.

The recording unit 1313 is configured to record resource pool information, where the resource pool information includes resources that need to be occupied by a virtual instance that needs to be run by each physical server in the hardware resource pool 132 of the public cloud data center and an instance type of the virtual instance that needs to be run by each physical server.

For example, fig. 3 is an exemplary representation of resource pool information provided in the present application, and it is assumed that the recording unit 1313 records CPU resource usage and memory resource usage of 4 physical servers, and according to the resource pool information shown in fig. 3, part of the CPU resources and memory resources in the physical servers 1 to 4 are used to create a reserved instance, and part of the CPU resources and memory resources in the physical servers 2 and 3 are used not only to create a reserved instance, but also to create an on-demand instance.

It should be understood that fig. 3 is only an example to show the content included in the resource pool information, and in practical applications, the resource pool information may also be stored in other expressions. In other expression modes of the resource pool information, the use condition of each resource type can also be marked with a specific numerical value, such as the CPU resource utilization rate of 50%, the memory resource occupancy rate of 80% and the like; the resource use condition of each physical server can be marked with other information, and the other information can be deduction time, affiliated users and corresponding specification information and the like; each physical server in the resource pool information may further include more resource categories, such as cache resources, network bandwidth resources, disk capacity resources, GPU resources, and the like in the foregoing; the resource pool information may also use other ways to label the instance categories (FIG. 3 labels different instance categories with different shading); the representation of the resource pool information is not limited to the graphical interface shown in fig. 3, and the resource pool information may be recorded by a table, a character, a file, or the like, which is not specifically limited in the present application.

In this embodiment of the application, the recording unit 1313 may determine, according to the specification information in the virtual instance creation request, a physical server used for creating the virtual instance, and update the resource pool information, where the updated resource pool information includes a record of the virtual instance created by the physical server and an instance type of the virtual instance.

Specifically, the recording unit 1313 may determine, according to the specification information, at least one physical server capable of reducing the virtual instance, optionally select one physical server from the at least one physical server as a physical server for creating the virtual instance, use a physical server with the largest amount of remaining resources as a physical server for creating the virtual instance, or select a physical server for creating the virtual instance from the physical servers in other manners, which is not limited in this application.

The sending unit 1314 is configured to, when the instance type of the virtual instance is the on-demand instance type, after the recording unit 1313 determines a physical server used for creating the virtual instance and updates the resource pool information, the sending unit 1314 sends the specification information of the virtual instance to the physical server to request for creating the virtual instance.

In a specific implementation, the receiving unit 1312 may receive a first virtual instance creation request, where the first virtual instance creation request includes an instance specification and an instance type of a first virtual instance, if the instance type of the first virtual instance is an on-demand instance type, the recording unit 1313 may update resource pool information according to the instance specification of the first virtual instance, where the updated resource pool information includes a record that the first physical server creates the first virtual instance and an instance type (on-demand instance) of the first virtual instance, and the sending unit 1314 may send the request to create the first virtual instance to the first physical server, where the first physical server may create the first virtual instance locally in response to the request.

In this embodiment of the application, the receiving unit 1312 may receive a second virtual instance creation request, where the second virtual instance creation request includes an instance specification and an instance type of a second virtual instance, and if the instance type of the second virtual instance is a reserved instance type, the recording unit 1313 may update the resource pool information according to the instance specification of the second virtual instance, where the updated resource pool information includes a record of the second virtual instance created by the second physical server and the instance type (reserved instance) of the second virtual instance. Briefly, if the instance type of the virtual instance is a reserved instance, the sending unit 1314 needs not to send a request for creating the virtual instance to the physical server for creating the virtual instance, and if the instance type of the virtual instance is an on-demand instance, the sending unit 1314 needs to send a request for creating the virtual instance to the physical server for creating the virtual instance. It should be understood that, since the reserved instance is not used by the user within the reserved time, the reserved instance may not be created on the physical server within the reserved time, which may avoid resource waste and improve the resource utilization rate of the hardware resource pool 132.

Scheduling unit 1311 may schedule the virtual instance set of reserved instance types to at least one physical server in the resource pool information such that the physical server with which the reserved instance type is associated before scheduling is available for the virtual instance of the on-demand instance type.

Specifically, one reserved instance may be scheduled, a part of reserved instances may be scheduled, and all reserved instances may also be scheduled, which is not specifically limited in this application. The reserved instances participating in the centralized scheduling at this time may be determined according to the number of the reserved instances and the number of the resource fragments, so that the number of the resource fragments recorded in the resource pool information after the centralized scheduling is reduced, where the resource fragments refer to resources that cannot be used for creating virtual instances, and the specific description may refer to the embodiment in fig. 1, and will not be repeated here.

Specifically, the scheduling unit 1311 may screen out a physical server that has a resource fragment and a reserved instance from the resource pool information, and then schedule the reserved instance to another physical server that can create the reserved instance, so as to reduce the number of resource fragments of the public cloud and reduce the operation cost of the public cloud.

For example, fig. 4 is a flowchart of centralized scheduling of reserved instances by resource pool information, as shown in fig. 4, assuming that an on-demand instance and a reserved instance exist in a physical server X1, and the remaining resources are resource fragments that cannot be used for creating a virtual instance, a scheduling unit 1311 may determine, according to the resource pool information, the physical server Y1 that can create the reserved instance, and then schedule the reserved instance to the physical server Y1, after resource scheduling, the remaining resources of the physical server X1 'may be used for creating a virtual instance, and the physical server X1' no longer has resource fragments, so that the number of resource fragments of a public cloud is increased, and the operation cost of the public cloud is reduced. Meanwhile, since the scheduling unit 1311 does not schedule the on-demand instance being used by the user, it is possible to avoid the user from being affected during the resource scheduling process, and improve the user experience, and the resource scheduling updates the resource pool information, instead of performing virtual instance migration on the physical server, which can reduce the overhead of network and memory bandwidth, and further reduce the operation cost of the public cloud.

In an embodiment, before performing resource scheduling on the reserved instances, the scheduling unit 1311 may determine a first amount of the remaining resources recorded in the resource pool information, then obtain a second amount of the remaining resources of the multiple physical servers in the hardware resource pool from the hardware resource pool 132, and in a case that a ratio between the first amount and the second amount is smaller than or equal to a first threshold, the scheduling unit 1311 performs the foregoing resource scheduling step, performs resource scheduling on the reserved instances in the resource pool information, and the amount of resource fragments recorded in the resource pool information after centralized scheduling becomes smaller.

The number of the remaining resources may refer to the number of standard virtual instances that can be created, the specification of the standard virtual instance is determined according to a history virtual machine creation record of the public cloud, and the specification of the standard virtual instance may include one or more specifications of the virtual instances. For example, it is determined that the issuing frequency of VM1 and VM2 is higher according to the history virtual machine creation record of the public cloud, then the total number of the remaining resources that can issue VM1 and VM2 may be determined according to the remaining resources recorded in the resource pool information, so as to obtain the first number, then the remaining resources of the multiple physical servers in the hardware resource pool 132 are obtained from the hardware resource pool 132, the total number of the remaining resources that can issue VM1 and VM2 is determined, so as to obtain the second number, and if the ratio of the first number to the second number is smaller than or equal to the first threshold, the scheduling unit 1311 may perform centralized scheduling on the reserved instances in the resource pool information.

It should be understood that the first quantity is determined according to the resource pool information, the second quantity is determined according to the remaining resources of the multiple physical servers in the hardware resource pool 132, the resource pool information additionally records a creation record of the reserved instance, so that the first quantity is smaller than the second quantity, and the smaller the ratio of the first quantity to the second quantity is, the larger the quantity of the resource fragments in the current resource pool information is, therefore, when the ratio of the first quantity to the second quantity is smaller than or equal to the second threshold, the scheduling unit 1311 schedules the reserved instance, and centrally schedules the virtual instances of the reserved instance type in the resource pool information to at least one physical server, so that the physical server associated with the virtual instance of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type, thereby avoiding resource waste and improving the resource utilization rate of the public cloud.

Optionally, the scheduling unit 1311 may also schedule the reserved instance periodically, where the periodicity may be the scheduling of the reserved instance at a fixed time interval, or may be the scheduling of the reserved instance after receiving the virtual instance creation request for M times, which is not limited in this application.

Optionally, the scheduling unit 1311 may also determine the first number and the second number periodically, and schedule the reserved instance in the resource pool information when a ratio of the first number to the second number is smaller than or equal to a second threshold. The foregoing may be referred to for the periodic description, and the description is not repeated here.

Optionally, the scheduling unit 1311 may further determine the first number and the second number through the above steps when the virtual instance creation fails, then perform scheduling of the reserved instance, reduce the number of resource fragments in the resource pool information, and retry to process the virtual instance creation request that was created unsuccessfully before, so as to improve the use experience of the user and improve the resource utilization rate of the public cloud.

In one embodiment, the reserved instance recorded in the resource pool information includes not only the instance specification, but also the reserved time of the reserved instance. When the reserved time of the reserved instance is about to expire, the sending unit 1314 may send a virtual instance creation request to the physical server where the reserved instance is recorded in the resource pool information, update the resource pool information, and convert the instance type of the virtual instance from the reserved instance to an on-demand instance, thereby avoiding that the user cannot use the reserved instance after the reserved time is over, and improving the user experience.

In summary, the cloud management platform provided by the application records the resources occupied by the virtual instances required to be operated by each physical server in the hardware resource pool of the public cloud data center and the instance types of the virtual instances required to be operated by each physical server, and the resource pool information recorded by the system includes the creation information of the reserved instances, but the physical servers of the hardware resource pool do not create the reserved instances, so that the resource waste caused by the fact that the reserved instances are not used by users in the reserved time is avoided; meanwhile, the system can also perform centralized scheduling on the reserved instances recorded in the resource pool information, so that the physical server associated with the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type, and therefore resource fragments of the public cloud can be reduced and the resource utilization rate of the public cloud can be improved under the condition that the use of users is not influenced.

Fig. 5 is a schematic flowchart of steps of a resource management method provided in the present application, where the method may be applied to the resource management system shown in fig. 2, and as shown in fig. 5, the method may include the following steps:

s510: and the cloud management platform records the resource pool information. The resource pool information is used for recording resources occupied by virtual instances needing to be operated by each physical server in a hardware resource pool of the public cloud data center and instance types of the virtual instances needing to be operated by each physical server, and the instance types comprise on-demand instance types and reserved instance types.

In a specific implementation, the resource pool information may include resource usage and instance types of a plurality of physical servers, and the resource usage of each physical server includes usage of resources of different types, for example, the resource usage of the physical server 1 may include CPU resource usage, memory resource usage, cache resource usage, memory bandwidth usage, network bandwidth resource usage, disk capacity resource usage, GPU resource usage, and the like. The usage of each category of resource may also be described by using a plurality of parameter information, for example, the usage of the CPU resource may include the number of cores of the CPU, the model of the CPU, and the like, and the usage of the memory resource may further include the size of the memory, the remaining capacity of the memory, and the like, which are not illustrated herein. The description of the resource pool information may refer to the foregoing embodiments of fig. 2 to fig. 3, which are not repeated herein.

In an embodiment of the present application, a cloud management platform may receive a virtual instance creation request sent by a user, where the virtual instance creation request includes specification information of a virtual instance and an instance type of the virtual instance, and the instance type includes an on-demand instance type and a reserved instance type. The cloud management platform may determine, according to the specification information and the resource pool information in the virtual instance creation request, a physical server used for creating the virtual instance, and then update the resource pool information, where the updated resource pool information includes a record of the virtual instance created by the physical server and an instance type of the virtual instance.

The specification information may include specification information of different resource types, such as the number of cores of the CPU, the type of the CPU, the size of the memory, the size of the cache, and the like, and reference may be made to the description of the specification information in the embodiments of fig. 1 to 4, which is not repeated herein. The on-demand instance refers to an instance which can be used by the user after payment, the reserved instance refers to an instance which cannot be used within the reserved time after the user pays, and the instance can be used after the reserved time passes, for example, the user pays for purchasing a virtual machine with the specification of X on 6 months and 1 day, and selects to reserve for 3 months, and then the user can use the virtual machine with the specification of X after 9 months and 1 day. The virtual instance may be a virtual machine, a physical machine, a container, a thread, and the like, and it should be understood that the description of the virtual instance, the on-demand instance, and the reserved instance may refer to the embodiments in fig. 2 to fig. 4, which is not repeated herein.

Optionally, if a first virtual instance creation request is received, where an instance type in the first virtual instance creation request is an on-demand instance type, after resource pool information is updated according to the first virtual instance creation request, the updated resource pool information includes a record that a first physical server creates a first virtual instance and an instance type (on-demand instance) of the first virtual instance, the cloud management platform sends a request for creating the first virtual instance to a first physical machine used for creating the first virtual instance, and the first physical machine may create a virtual instance in response to the request.

Optionally, if a second virtual instance creation request is received, where an instance type in the second virtual instance creation request is a reserved instance type, after resource pool information is updated according to the second virtual instance creation request, the updated resource pool information includes a record of the second virtual instance created by the second physical server and an instance type (reserved instance) of the second virtual instance. But the second physical server does not create the second virtual instance. It should be understood that, since the reserved instance is not used by the user within the reserved time, the reserved instance may not be created on the physical server within the reserved time, thereby avoiding resource waste and improving the resource utilization rate of the hardware resource pool 132.

Further, the reserved instance recorded in the resource pool information includes not only the instance specification but also the reserved time of the reserved instance. When the reserved time of the reserved instance is about to expire, the cloud management platform can send a virtual instance creation request to the physical server where the reserved instance is recorded in the resource pool information, update the resource pool information, and convert the instance type of the virtual instance from the reserved instance to an on-demand instance, so that the situation that a user cannot use the reserved instance after the reserved time is over is avoided, and the use experience of the user is improved.

S520: the cloud management platform schedules the virtual instances of the reserved instance type to at least one physical server in the resource pool information in a centralized manner, so that the physical server associated with the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type.

In specific implementation, when the cloud management platform performs centralized scheduling on the virtual instances of the reserved instance type, one reserved instance may be scheduled, a part of reserved instances may also be scheduled, and all reserved instances may also be scheduled, which is not specifically limited in the present application. The reserved instances participating in the centralized scheduling at this time can be specifically determined according to the number of the reserved instances and the number of the resource fragments, and it should be understood that the amount of the resource fragments recorded in the resource pool information is reduced after the centralized scheduling. Here, the resource fragment refers to a resource that cannot be used for creating a virtual instance, and the description of the resource fragment may refer to the embodiments in fig. 1 to fig. 4, which is not repeated herein.

In specific implementation, the cloud management platform can screen out the physical servers with resource fragments and reserved examples from the resource pool information, and then schedule the reserved examples to other physical servers capable of creating the reserved examples, so that the number of the resource fragments of the public cloud is reduced, and the operation cost of the public cloud is reduced. Specifically, reference may be made to the example described in the embodiment of fig. 4, which is not repeated herein.

In an embodiment, before performing resource scheduling on the reserved instance, the cloud management platform may determine a first quantity of remaining resources recorded in the resource pool information, then obtain a second quantity of remaining resources of the plurality of physical servers in the hardware resource pool from the hardware resource pool, perform resource scheduling on the reserved instance when a ratio between the first quantity and the second quantity is less than or equal to a first threshold, and reduce a resource fragment amount in the resource pool information after the resource scheduling.

The number of the remaining resources may refer to the number of standard virtual instances that can be created, the specification of the standard virtual instance is determined according to a history virtual machine creation record of the public cloud, and the specification of the standard virtual instance may include one or more specifications of the virtual instances. For example, it is determined that the issuing frequency of VM1 and VM2 is higher according to the history virtual machine creation record of the public cloud, then the total number of VM1 and VM2 that can be issued by the remaining resource may be determined according to the remaining resource recorded in the resource pool information, so as to obtain the first number, then the remaining resource of the plurality of physical servers in the hardware resource pool is obtained from the hardware resource pool, and the total number of VM1 and VM2 that can be issued by the remaining resource is determined, so as to obtain the second number, and if the ratio of the first number to the second number is smaller than or equal to the first threshold, the cloud management platform may perform centralized scheduling on the reserved instance in the resource pool information.

It should be understood that the first quantity is determined according to the resource pool information, the second quantity is determined according to the remaining resources of the plurality of physical servers in the hardware resource pool, the resource pool information additionally records creation records of the reserved instances, and therefore the first quantity is smaller than the second quantity, and the smaller the ratio of the first quantity to the second quantity is, the larger the quantity of resource fragments in the current resource pool information is, when the ratio of the first quantity to the second quantity is smaller than or equal to the second threshold value, the cloud management platform schedules the reserved instances, centrally schedules the virtual instances of the reserved instance types in the resource pool information to at least one physical server, so that the physical servers associated with the virtual instances of the reserved instance types before scheduling can be used by the virtual instances of the on-demand instance types, thereby avoiding resource waste and improving the resource utilization rate of the public cloud.

Optionally, the cloud management platform may also schedule the reserved instance periodically, where the periodicity may be scheduling of the reserved instance at a fixed time interval, or scheduling of the reserved instance after receiving the virtual instance creation request for M times, which is not limited in this application.

Optionally, the cloud management platform may further periodically determine the first number and the second number, and schedule the reserved instance in the resource pool information when a ratio of the first number to the second number is less than or equal to a second threshold. The foregoing may be referred to for the periodic description, and the description is not repeated here.

Optionally, the cloud management platform may further determine the first number and the second number through the above steps under the condition that the virtual instance creation fails, then perform scheduling of the reserved instance, reduce the number of resource fragments in the resource pool information, and retry to process the virtual instance creation request that was created with the failure before, so as to improve the use experience of the user and improve the resource utilization rate of the public cloud.

In summary, by using the resource management method provided by the application, the cloud management platform records the resources occupied by the virtual instances required to be operated by each physical server in the hardware resource pool of the public cloud data center and the instance types of the virtual instances required to be operated by each physical server, and the resource pool information recorded by the system includes the creation information of the reserved instances, but the physical servers of the hardware resource pool do not create the reserved instances, so that the resource waste caused by the fact that the reserved instances are not used by users in the reserved time is avoided; meanwhile, the system can also perform centralized scheduling on the reserved instances recorded in the resource pool information, so that the physical servers associated with the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type, thereby reducing resource fragments of the public cloud and improving the resource utilization rate of the public cloud under the condition of not influencing the use of users.

Fig. 6 is a schematic structural diagram of a computing device 600 provided herein. Among other things, the computing device 600 may be the cloud management platform 131 in fig. 1-5. As shown in fig. 6, computing device 600 includes: a processor 610, a communication interface 620, and a memory 630. The processor 610, the communication interface 620, and the memory 630 may be connected to each other through an internal bus 640, or may communicate with each other through other means such as wireless transmission. In the embodiment of the present application, the bus may be a PCI bus, for example, by using a bus connection. The bus 640 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 6, but that does not indicate only one bus or one type of bus.

The processor 610 may be constituted by at least one general-purpose processor, such as a CPU, or a combination of a CPU and a hardware chip. The hardware chip may be an ASIC, PLD, or a combination thereof. The aforementioned PLD may be a CPLD, an FPGA, a GAL, or any combination thereof. Processor 230 executes various types of digitally stored instructions, such as software or firmware programs stored in memory 630, which enable computing device 600 to provide a variety of services.

The memory 630 is used for storing program codes and is controlled by the processor 610 to execute the processing steps of the cloud management platform 131 in the above embodiments. The program code may include one or more software modules, which may be software modules provided by the cloud management platform 131 in the embodiment of fig. 2, such as a recording unit and a scheduling unit, where the recording unit is configured to record resource pool information, the resource pool information is used to record resources that need to be occupied by a virtual instance that needs to be run by each physical server in a hardware resource pool of a public cloud data center and an instance type of a virtual instance that needs to be run by each physical server, and the scheduling unit is configured to centrally schedule the virtual instance of the reserved instance type to at least one physical server in the resource pool information, so that a physical server to which the virtual instance of the reserved instance type is associated before scheduling is available for a virtual instance of an on-demand instance type. Specifically, the method may be used to perform steps S510 to S520 in the embodiment of fig. 5 and optional steps thereof, and may also be used to perform other steps performed by the cloud management platform 131 described in the embodiments of fig. 1 to 5, which are not described herein again.

Memory 630 may include volatile memory such as RAM, DRAM, SRAM, SDRAM, DDR, cache, etc., memory 630 may also include combinations of the above.

The communication interface 620 may be a wired interface (e.g., an ethernet interface), may be an internal interface (e.g., a PCI bus interface), a wired interface (e.g., an ethernet interface), or a wireless interface (e.g., a cellular network interface or using a wireless local area network interface) for communicating with other devices or modules.

It should be noted that, this embodiment may be implemented by a general physical server, for example, an ARM server or an X66 server, or may also be implemented by a virtual machine implemented based on the general physical server and combining with an NFV technology, where the virtual machine refers to a complete computer system that has a complete hardware system function and runs in a completely isolated environment through software simulation, for example, this embodiment may be implemented on a cloud computing infrastructure, and a specific process implemented on the cloud computing infrastructure may refer to the embodiment in fig. 3, which is not repeated.

It should be noted that fig. 6 is only one possible implementation manner of the embodiment of the present application, and in practical applications, the computing device 600 may further include more or less components, which is not limited herein. For contents that are not shown or described in the embodiments of the present application, reference may be made to the related explanations in the embodiments of fig. 1 to fig. 5, which are not described here again.

It should be understood that the computing device shown in fig. 6 may also be a computer cluster formed by at least one physical server, and reference may be made to the embodiments in fig. 1 to fig. 4 specifically, so that details are not described here again to avoid repetition.

Embodiments of the present application further provide a computer-readable storage medium, which stores instructions that, when executed on a processor, implement the method flows shown in fig. 5 and its optional steps.

Embodiments of the present application also provide a computer program product, wherein the method flows shown in fig. 5 and its optional steps are implemented when the computer program product is run on a processor.

The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes at least one computer instruction. The procedures or functions according to the embodiments of the invention are wholly or partly generated when the computer program instructions are loaded or executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage node, such as a server, a data center, or the like, that contains at least one collection of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital Video Disk (DVD), or a semiconductor medium.

While the invention has been described with reference to specific embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A resource management method of a public cloud data center is characterized by being applied to a cloud management platform of the public cloud data center, and the method comprises the following steps:

the cloud management platform records resource pool information, wherein the resource pool information is used for recording resources occupied by virtual instances needing to be operated by each physical server in a hardware resource pool of the public cloud data center and instance types of the virtual instances needing to be operated by each physical server, and the instance types comprise on-demand instance types and reserved instance types;

the cloud management platform centrally schedules the virtual instances of the reserved instance type to at least one physical server in the resource pool information, so that the physical server associated with the virtual instances of the reserved instance type before scheduling can be used by the virtual instances of the on-demand instance type.

2. The method of claim 1, further comprising:

the cloud management platform receives a first virtual instance creation request, wherein the first virtual instance creation request comprises an instance type of a first virtual instance, and the instance type of the first virtual instance is an on-demand instance type;

the cloud management platform updates the resource pool information, wherein the updated resource pool information comprises a record of the first virtual instance created by the first physical server and an instance type of the first virtual instance;

the cloud management platform sends a request to create the first virtual instance to the first physical server.

3. The method according to claim 1 or 2, characterized in that the method further comprises:

the cloud management platform receives a second virtual instance creation request, wherein the second virtual instance creation request comprises an instance type of a second virtual instance, and the instance type of the second virtual instance is a reserved instance type;

and the cloud management platform updates the resource pool information, wherein the updated resource pool information comprises a record of the second virtual instance created by the second physical server and the instance type of the second virtual instance.

4. The method of claim 3, wherein before the cloud management platform centrally schedules virtual instances of a reserved instance type to at least one physical server in the resource pool information, the method further comprises:

the cloud management platform determines a first number of remaining resources of the resource pool information record;

the cloud management platform acquires a second quantity of the remaining resources of the plurality of physical servers in the hardware resource pool from the hardware resource pool;

the cloud management platform centrally scheduling virtual instances of the reserved instance type to at least one physical server in the resource pool information comprises:

the cloud management platform centrally schedules the virtual instances of the reserved instance type to at least one physical server in the resource pool information if a ratio between the first number and the second number is less than or equal to a first threshold.

5. The method of claim 3 or 4, wherein the cloud management platform determining the first amount of remaining resources of the resource pool information record comprises:

the cloud management platform periodically determines a first number of remaining resources of the resource pool information record; or,

and the cloud management platform determines a first number of the remaining resources of the resource pool information record under the condition that the virtual instance is failed to be created.

6. The method of any of claims 1 to 5, wherein the virtual instance comprises a virtual machine, a container, and/or a bare metal server.

7. The cloud management platform is applied to a cloud management platform of a public cloud data center, and comprises:

a recording unit, configured to record the resource pool information, where the resource pool information is used to record resources that need to be occupied by a virtual instance that needs to be run by each physical server in a hardware resource pool of the public cloud data center and an instance type of a virtual instance that needs to be run by each physical server, and the instance types include an on-demand instance type and a reserved instance type;

a scheduling unit, configured to schedule the virtual instances of the reserved instance type to at least one physical server in the resource pool information in a centralized manner, so that the physical server associated with the virtual instances of the reserved instance type before scheduling is available for the virtual instances of the on-demand instance type.

8. The cloud management platform of claim 7, wherein the cloud management platform further comprises a receiving unit and a transmitting unit,

the receiving unit is configured to receive a first virtual instance creation request, where the first virtual instance creation request includes an instance type of a first virtual instance, and the instance type of the first virtual instance is an on-demand instance type;

the recording unit is configured to update the resource pool information, where the updated resource pool information includes a record of the first virtual instance created by the first physical server and an instance type of the first virtual instance;

the sending unit is configured to send a request for creating the first virtual instance to the first physical server.

9. The cloud management platform of claim 8,

the receiving unit is configured to receive a second virtual instance creation request, where the second virtual instance creation request includes an instance type of a second virtual instance, and the instance type of the second virtual instance is a reserved instance type;

the recording unit is configured to update the resource pool information, where the updated resource pool information includes a record of the second virtual instance created by the second physical server and an instance type of the second virtual instance.

10. The cloud management platform of claim 9,

the scheduling unit is configured to determine a first number of remaining resources recorded in the resource pool information before scheduling the virtual instances of the reserved instance types to at least one physical server in a centralized manner in the resource pool information;

the scheduling unit is configured to obtain, from the hardware resource pool, a second number of remaining resources of the plurality of physical servers in the hardware resource pool;

the scheduling unit is configured to, by the cloud management platform, collectively schedule the virtual instances of the reserved instance type to at least one physical server in the resource pool information when a ratio between the first number and the second number is less than or equal to a first threshold.

11. The cloud management platform of claim 9 or 10, wherein the scheduling unit is configured to periodically determine a first number of remaining resources of the resource pool information record; or,

the scheduling unit is configured to determine a first number of remaining resources of the resource pool information record in case of a failure in creating a virtual instance.

12. The cloud management platform of any of claims 7 to 11, wherein the resources comprise virtual machines, containers, and/or bare metal servers.

13. A computing device comprising a processor and a memory, the memory storing code, the processor executing the code to implement the method of any of claims 1 to 6.

14. A computer-readable storage medium having stored therein instructions which, when executed on a processor, carry out the method of any one of claims 1 to 6.