WO2023221844A1

WO2023221844A1 - Resource scheduling method, device, and storage medium

Info

Publication number: WO2023221844A1
Application number: PCT/CN2023/093401
Authority: WO
Inventors: 邓荔文; 仇烨亮; 彭飞; 钟剑
Original assignee: 阿里巴巴（中国）有限公司
Priority date: 2022-05-16
Filing date: 2023-05-11
Publication date: 2023-11-23
Also published as: CN114880077A

Abstract

Embodiments of the present application provide a resource scheduling method, a device, and a storage medium. In the embodiments of the present application, by pre-planning, for a physical machine, the specifications and the quantity of exclusive virtual machines and shared virtual machines which can be deployed in a hybrid manner, hybrid deployment of exclusive virtual machines and shared virtual machines is realized, thereby facilitating the reduction in the quantity of natural resource fragments caused by the limitation to the specifications of virtual machines. In addition, since the resource specifications and quantity of first-type virtual machines and second-type virtual machines supported by a physical machine are obtained by performing simulation and scheduling on the total resources of the physical machine in advance with the maximization of the resource utilization rate of the physical machine as a target, the deployment of exclusive virtual machines and shared virtual machines in the simulated quantity and specifications in the physical machine can reduce resource fragments during actual resource scheduling of the physical machine, thereby facilitating the improvement of the resource utilization rate of the physical machine.

Description

Resource scheduling methods, equipment and storage media

This application claims priority to the Chinese patent application filed with the China Patent Office on May 16, 2022, with the application number 202210530246.4 and the application name "Resource Scheduling Method, Equipment and Storage Medium", the entire content of which is incorporated into this application by reference. middle.

Technical field

The present application relates to the field of cloud computing technology, and in particular, to a resource scheduling method, equipment and storage medium.

Background technique

Server virtualization technology is a key technology based on the infrastructure layer in cloud computing. This technology virtualizes physical servers to deploy multiple virtual machines (VMs) on a single physical machine.

In practical applications, a certain type of physical machine can only produce virtual machines of certain specifications, and the resources required by the virtual machine specifications may not be exactly equal to the resources of the physical machine, which leads to resource fragmentation in the physical machine. For example, assume that the resource ratio of memory to CPU in the physical machine is 192GB memory:104vCPU. Among them, 104vCPU means that the number of cores of the virtual CPU is 104 cores. The resource ratio required for the virtual machine specifications deployed on this physical machine is 4GB memory:1vCPU, which undoubtedly results in at least 58 vCPU resources remaining.

Contents of the invention

Various aspects of this application provide a resource processing and scheduling method, device, and storage medium to reduce resource fragmentation of physical machines and improve resource utilization of physical machines.

This embodiment of the present application provides a resource scheduling method, including:

Obtaining the first resources occupied by the first physical machine for separately deploying the first type of virtual machine; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine;

When the first resource is less than the total resources of the physical machine, set the virtual machine deployment mode of the first physical machine to a hybrid deployment mode that supports exclusive virtual machines and shared virtual machines;

With the goal of maximizing the total resource utilization, simulation scheduling is performed on the total resources of the physical machine according to the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine to determine the physical machine's total resources. The target resource specifications and quantities of the first type of virtual machine and the second type of virtual machine supported by the machine; the second type of virtual machine is the same as the first type of the exclusive virtual machine and the shared virtual machine. Different types of virtual machines;

According to the target resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the first physical machine, perform a mixed deployment of exclusive virtual machines and shared virtual machines on the physical machine;

The exclusive virtual machine refers to a virtual machine in which a virtual processor exclusively occupies the physical processor resources of the physical machine; the shared virtual machine refers to a virtual machine in which a virtual processor shares physical processor resources.

This embodiment of the present application also provides a resource scheduling method, including:

Obtain the virtual machine inventory level of the first physical machine where the first type of virtual machine is separately deployed; the first type of virtual machine The virtual machine is an exclusive virtual machine or a shared virtual machine;

When the virtual machine inventory level is less than the set first water level threshold, select a second physical machine with idle resources that meets the resource requirements of the first type virtual machine from the physical machines where the second type virtual machine is deployed separately. machine; the second type virtual machine is a type different from the first type virtual machine among the exclusive virtual machine and the shared virtual machine;

Switch the virtual machine deployment mode of the second physical machine to a hybrid deployment mode that supports exclusive virtual machines and shared virtual machines;

Utilizing the idle resources of the second physical machine, deploy the first type virtual machine in the second physical machine;

The exclusive virtual machine refers to a virtual machine in which a virtual processor exclusively occupies the physical processor resources of the physical machine; the shared virtual machine refers to a virtual machine in which a virtual processor shares the physical processor resources.

An embodiment of the present application also provides a computing device, including: a memory and a processor; wherein the memory is used to store a computer program;

The processor is coupled to the memory and is configured to execute the computer program for performing the steps in each of the above resource scheduling methods.

Embodiments of the present application also provide a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute each of the above resource scheduling methods. A step of.

In the embodiment of this application, by pre-planning the specifications and quantity of exclusive virtual machines and shared virtual machines that can be mixedly deployed on the physical machine, the mixed deployment of exclusive virtual machines and shared virtual machines is realized, which is helpful to To reduce the amount of natural resource fragmentation caused by limited virtual machine specifications. On the other hand, because the resource specifications and quantities of the first type of virtual machines and the second type of virtual machines supported by the physical machine are pre-scheduled by simulating the total resources of the physical machine with the goal of maximizing the resource utilization of the physical machine. , Therefore, deploying the simulated number and specifications of dedicated virtual machines and shared virtual machines in the physical machine can reduce resource fragmentation during actual resource scheduling of the physical machine and help improve the resource utilization of the physical machine.

Description of the drawings

The drawings described here are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the attached picture:

Figure 1 is a schematic structural diagram of a computing system provided by an embodiment of the present application;

Figure 2 is a schematic diagram of a separate deployment mode provided by the embodiment of the present application;

Figure 3 is a schematic diagram of the workflow of the computing system provided by the embodiment of the present application;

Figure 4 is a schematic diagram of the effects of the individual deployment mode and the mixed deployment mode provided by the embodiment of the present application;

Figure 5 is a schematic diagram of resource scheduling in the hybrid deployment mode provided by the embodiment of the present application;

Figure 6 is a schematic diagram of the weak mixing part mode and the strong mixing part mode provided by the embodiment of the present application;

Figure 7 is a schematic flow chart of virtual machine deployment by the computing system provided by the embodiment of the present application;

Figures 8 and 9 are schematic diagrams of the virtual machine deployment mode switching process provided by the embodiment of the present application;

Figures 10 and 11 are flowcharts of the resource scheduling method provided by the embodiment of the present application;

Figure 12 is a schematic structural diagram of a computing device provided by an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

In some embodiments of the present application, mixed deployment of exclusive virtual machines and shared virtual machines can be realized by pre-planning the specifications and quantities of exclusive virtual machines and shared virtual machines that can be mixed deployed on the physical machine. Helps reduce the amount of natural resource fragmentation caused by limited virtual machine specifications. On the other hand, because the resource specifications and quantities of the first type of virtual machines and the second type of virtual machines supported by the physical machine are pre-scheduled by simulating the total resources of the physical machine with the goal of maximizing the resource utilization of the physical machine. , Therefore, deploying the simulated number and specifications of dedicated virtual machines and shared virtual machines in the physical machine can reduce resource fragmentation during actual resource scheduling of the physical machine and help improve the resource utilization of the physical machine.

The technical solutions provided by each embodiment of the present application will be described in detail below with reference to the accompanying drawings.

It should be noted that the same reference numerals refer to the same object in the following figures and embodiments. Therefore, once an object is defined in one figure or embodiment, it does not need to be defined in subsequent figures and embodiments. which is discussed further.

Figure 1 is a schematic structural diagram of a computing system provided by an embodiment of the present application. As shown in Figure 1, the computing system includes: a management node 10 and a computing cluster 11. The computing cluster 11 includes: one or more physical machines 20. Preferably, the number of physical machines 20 is multiple. In this embodiment, the physical machine 20 refers to a computing device that provides resources. The amount of resources contained in each physical machine 20 may be the same or different. Among them, the same amount of resources contained in the physical machines 20 means that the amount of resources of each resource type of the physical machines is the same; the different amounts of resources contained in the physical machines 20 means that different physical machines have resources of a certain dimension or certain dimensions. The amount is different. Among them, the resources of the physical machine include but are not limited to: processor resources (such as CPU, GPU, etc.), storage resources (such as memory, etc.) and network resources (bandwidth, etc.).

The management and control node 10 refers to a computing device that performs resource management and resource scheduling on the physical machine 20 . The management and control node 10 can be a single server device, a cloud server array, or a virtual machine (Virtual Machine, VM) running in a cloud server array. In addition, the management and control node 10 may also refer to other computing devices with corresponding service capabilities, such as computers and other terminal devices (running service programs).

The connection between the management node 10 and the physical machine 20 may be wireless or wired. Optionally, the management and control node 10 can be connected to the physical machine 20 through a mobile network. Correspondingly, the network standard of the mobile network can be 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS). ), 4G(LTE), 4G+(LTE+), 5G, WiMax, etc. Optionally, the management node 10 and the physical machine 20 can also be communicated and connected through Bluetooth, WiFi, infrared, etc.

In practical applications, cloud vendors can use virtualization technology to deploy an exclusive virtual machine 201 or a shared virtual machine 202 on the physical machine 20 to achieve exclusive or shared use of the processor resources of the physical machine 20 . The processor of the physical machine 20 includes but is not limited to: CPU, GPU, etc. The processor resources of the physical machine 20 may be represented by the number of physical cores of the processor. In the embodiment of this application, the core of the processor of the physical machine is called the physical core, and the core of the processor used by the virtual machine is called the logical core. When allocating processor resources to the virtual machine, it is necessary to establish the physical core and the logical core. mapping relationship between them.

In the embodiment of this application, the processor of the virtual machine is called a virtual processor; the processor of the physical machine 20 is called Physical processor. Among them, an exclusive virtual machine refers to a virtual machine in which a virtual processor (such as a vCPU) exclusively occupies the physical processor resources of a physical machine. That is, one physical core can only be mapped to one logical core or one virtual machine. At this time, the virtual processor resources of the physical machine are relatively low, and the virtual processors of the virtual machines will not compete for physical processor resources. Therefore, the performance index of the Service Level Agreement (SLA) of an exclusive virtual machine is relatively high. Exclusive virtual machines have higher service performance.

Among them, the virtual ratio of processor resources refers to the ratio of the number of logical cores obtained by virtualizing a physical core to the number of physical cores, and is used to reflect the number of logical cores that can be virtualized by one physical core. For example, if the virtual ratio of processor resources is 4 (number of logical cores): 1 (number of physical cores), it means that one physical core is virtualized into 4 logical cores.

A shared virtual machine refers to a virtual machine in which a virtual processor shares physical processor resources, that is, one physical core can be mapped to multiple logical cores or multiple virtual machines. At this time, one physical core is virtualized into multiple logical cores. The virtualization ratio of the physical machine's processor resources is high, and the utilization rate of the physical machine's processor resources is high. However, because multiple virtual processors in a shared virtual machine share the same physical processor resource, multiple virtual machines compete for processor resources in a shared virtual machine. Therefore, the performance index of the SLA of a shared virtual machine is relatively low. Low. Shared virtual machines have lower service performance.

As shown in Figure 1, in some solutions, the exclusive virtual machine 201 and the shared virtual machine 202 are not deployed on the same physical machine 20. That is, the exclusive virtual machine 201 and the shared virtual machine 202 do not coexist on the same physical machine 20, but are physically isolated, and their resources are not shared at all. Users can apply for shared virtual machines or dedicated virtual machines according to their own needs. The management and control node 10 can respond to the user's virtual machine deployment request, obtain the virtual machine demand information, and perform resource scheduling in the physical machine 20 based on the virtual machine demand information, and deploy a virtual machine adapted to the virtual machine deployment request. As shown in Figure 2, if a virtual machine deployment request applies to deploy a shared virtual machine, the management and control node 10 can schedule adapted resources from the physical machine used to deploy the shared virtual machine; and use the scheduled resources to Deploy shared virtual machines on physical machines. Similarly, if a virtual machine deployment request applies to deploy an exclusive virtual machine, the management and control node 10 can schedule adapted resources from the physical machine used for the exclusive virtual machine, and use the scheduled resources to in the physical machine Deploy dedicated virtual machines.

Regardless of whether it is a shared virtual machine or a dedicated virtual machine, the specifications of the virtual machines provided by cloud vendors are certain. Among them, the resource specifications of the virtual machine include: the resource type of the virtual machine and the amount of resources of each resource type. For example, the resource specifications of the virtual machine may include: the number of virtual processors, the amount of memory, bandwidth, the number of network cards, the number of packets per second (Packets Per Second, PPS), etc., but are not limited to this. Users can select appropriate specifications to apply for virtual machine resources based on resource requirements. Since the specifications of virtual machines provided by cloud vendors are certain and limited, the resources required by the virtual machine specifications may not be completely equal to the resources of the physical machine, which leads to resource fragmentation in the physical machine. For example, it is known that the resource specifications of exclusive virtual machines produced by cloud vendors are: vCPU and memory ratios are: 1vCPU: 2GB memory, 1vCPU: 4GB memory and 1vCPU: 8GB memory; but the physical machine 20 The specifications are: 104vCPU: 192GB memory, 104vCPU: 384GB memory and 104vCPU: 768GB memory. Regardless of the physical machine's resource specifications, the ratio of vCPU to memory is not strictly 1vCPU: 2GB memory, 1vCPU: 4GB memory, or 1vCPU: 8GB. This results in resources appearing in the physical machine no matter how it is deployed or scheduled. Fragmentation causes this part of resources to be idle and the resource utilization rate is low.

In the embodiment of this application, in order to reduce physical machine resource fragmentation and improve resource utilization, a virtual machine co-location solution is proposed. Among them, virtual machine hybrid deployment refers to supporting the deployment of dedicated virtual machines and shared virtual machines on the same physical machine. The following is an exemplary description of the virtual machine co-location method provided by the embodiment of the present application in conjunction with Figure 3.

As shown in Figure 3, the management and control node 10 can manage and schedule the resources of the physical machine 20. For the physical machine 20 connected to the computing system, when connecting to the computing system, the identity of the physical machine 20 and the resources of the physical machine 20 can be registered in the management and control section. Point 10. Correspondingly, the management and control node 10 can obtain the total resources of the physical machine. In the embodiment of the present application, in order to improve resource utilization, the management and control node 10 can schedule and plan the resource usage of the physical machine 20, that is, predetermine the type and specification of the virtual machine deployed on the physical machine 20, so that the resources of the physical machine 20 can Maximize utilization. In this embodiment of the present application, the virtual machine type is used to indicate whether the virtual machine is a shared virtual machine or an exclusive virtual machine.

Based on the above goal of improving physical machine resource utilization, in this embodiment of the present application, the management and control node 10 can obtain the resources required by the physical machine to separately deploy the first type of virtual machine. Among them, the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. In the embodiment of the present application, the specific implementation manner in which the management and control node 10 obtains the resources required by the physical machine to deploy the first type of virtual machine is not limited.

In some embodiments, the management and control node 10 can greedily box the total resources of the physical machine 20 based on the preset resource specifications of the first type of virtual machine to determine the resources occupied by the greedy boxed first type of virtual machine. And determine the resources occupied by the greedy boxed first-type virtual machine, and deploy the resources occupied by the first-type virtual machine for the physical machine. The preset resource specification of the first type of virtual machine may be one specification or multiple specifications. Multiple means 2 or more types. The amount of resources in at least one dimension is different in different resource specifications.

Optionally, the management and control node 10 can simulate deployment based on the preset resource specifications of the first type of virtual machine, in descending order of resource specifications, and in accordance with the target specifications of the first type of virtual machine that the physical machine can currently deploy. For a first-type virtual machine with target specifications, the free resources of the physical machine do not meet the specification requirements of any first-type virtual machine. The resources occupied by the first-type virtual machine that has been simulated and deployed to the physical machine are obtained, which is greedy boxing. The resources occupied by the first type of virtual machine. Among them, the resource specifications of the first type of virtual machines in greedy boxing may be all the same, partially the same, or completely different.

For example, if the first type of virtual machine is an exclusive virtual machine, the total resources of the physical machine 20 can be greedily boxed based on the preset resource specifications of the exclusive virtual machine to determine the greedy boxed exclusive type. The resources occupied by the virtual machine. Assume that as shown in Figure 4, the resource specification of the exclusive virtual machine is 1vCPU:2GB memory; the total resource ratio of the physical machine is 104vCPU:192GB memory, then the total resource ratio of the exclusive virtual machine to the physical machine with this resource specification After greedy boxing of resources, the physical machine can deploy up to 96 exclusive virtual machines with resource specifications of 1vCPU:2GB memory. That is, the resources occupied by the exclusive virtual machine with greedy boxing are 96 vCPUs and 192GB of memory.

Since the specifications of virtual machines provided by cloud vendors are certain and limited, the resources required by the virtual machine specifications may not be completely equal to the resources of the physical machine, which leads to resource fragmentation in the physical machine. For example, in Figure 4 above, correspondingly, after greedy boxing, the virtual machine will generate resource fragments of 8 vCPUs. Based on this, when it is determined that the resources occupied by the first type of virtual machine alone are deployed on the physical machine, it can also be determined whether resource fragmentation will occur if the physical machine alone deploys the first type of virtual machine. Specifically, it can be compared whether the resources occupied by the first type of virtual machine deployed alone on the physical machine are less than the total resources of the physical machine; if the resources occupied by the first type of virtual machine deployed alone on the physical machine are less than the total resources of the physical machine, it can be determined that the resources occupied by the physical machine alone are less than the total resources of the physical machine. Deploying the first type of virtual machine will create resource fragmentation.

In this embodiment, in order to reduce the resource fragmentation of the physical machine, when the resources occupied by the first type of virtual machine deployed separately on the physical machine are less than the total resources of the physical machine, the management and control node 10 can change the virtual machine deployment mode of the physical machine. Set to support mixed deployment mode of dedicated virtual machines and shared virtual machines. Correspondingly, if the physical machine is deployed separately with the first type The resources occupied by the virtual machine are equal to the total resources of the physical machine. The virtual machine deployment mode of the virtual machine can be set to support independent deployment of the first type of virtual machine.

Furthermore, the management and control node 10 can aim at maximizing the total resource utilization of the physical machine, and perform simulated scheduling on the total resources of the physical machine according to the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine to determine the physical machine's total resource utilization. The target resource type and quantity of the first type of virtual machines and the second type of virtual machines supported by the machine. The above-mentioned second type of virtual machine is a type different from the first type of virtual machine among exclusive virtual machines and shared virtual machines. For example, the first type of virtual machine is a dedicated virtual machine, and the second type of virtual machine is a shared virtual machine. Correspondingly, if the first type of virtual machine is a shared virtual machine, the second type of virtual machine is an exclusive virtual machine.

Optionally, the management and control node 10 can combine the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine according to the total resource allocation of the physical machine to obtain multiple first type virtual machines and second type virtual machines. A combination of virtual machines; at least one type of virtual machine in each combination has different resource specifications, and the amount of resources required by each combination is less than or equal to the total resources of the physical machine. Further, from the combinations of multiple first-type virtual machines and second-type virtual machines, the combination with the smallest difference between the amount of resources required and the total resources of the physical machine can be obtained as the target combination. Further, the resource specifications and quantity of the first type virtual machine and the second type virtual machine in the target combination may be determined as the resource specifications and quantity of the first type virtual machine and the second type virtual machine in the simulation scheduling result.

Alternatively, in some embodiments, the management and control node 10 can greedily box the total resources of the physical machine based on the preset resource specifications of the first type of virtual machine to determine the resources occupied by the greedy boxed first type of virtual machine. ; And determine the resource fragments of the physical machine based on the total resources of the physical machine and the resources occupied by the greedy boxed first-type virtual machine. Afterwards, the management and control node 10 aims at maximizing the utilization of the resource fragments and performs simulation scheduling on the resource fragments according to the resource specifications of the second type virtual machine to determine the resource specifications and quantity of the second type virtual machines deployed in the resource fragments. Of course, the management and control node 10 can also determine the resource specifications and quantity of the first type of virtual machine for greedy boxing. Further, determine the resource specifications and quantity of the second type of virtual machines deployed in the resource fragments, and the resource specifications and quantity of the first type of virtual machines in greedy boxing, as the first type of virtual machines and the third type of virtual machines in the simulation scheduling results. Resource specifications and quantity of type 2 virtual machines.

Preferably, the resources required by the first type virtual machine and the second type virtual machine supported by the physical machine 20 are equal to the total resources of the physical machine 20 . For example, in Figure 4, in the mixed deployment mode of dedicated virtual machines and shared virtual machines, it is determined that the resource specifications of the dedicated virtual machine are 1vCPU: 2GB memory, and the quantity is 88; the resource specifications of the shared virtual machine are 2vCPU :0.5GB memory, the quantity is 32, and the virtual ratio corresponding to the shared virtual machine is 4:1, that is, one physical core virtualizes 4 logical cores. In this way, the determined resources occupied by the dedicated virtual machine and the shared virtual machine are equal to the total resources of the physical machine.

After determining the resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the physical machine 20, the physical machine 20 can be configured according to the resource specifications and quantity of the first type virtual machine and the second type virtual machine. Mixed deployment of dedicated virtual machines and shared virtual machines.

In the embodiment of this application, by pre-planning the specifications and quantity of exclusive virtual machines and shared virtual machines that can be mixedly deployed on the physical machine, the mixed deployment of exclusive virtual machines and shared virtual machines is realized, which is helpful to To reduce the amount of natural resource fragmentation caused by limited virtual machine specifications. On the other hand, since the physical machine supports the first type of virtual machine and the second type The resource specifications and quantity of type virtual machines are pre-scheduled by simulating the total resources of the physical machine with the goal of maximizing the resource utilization of the physical machine. Therefore, the simulated quantity and specifications are deployed in the physical machine. Virtual machines and shared virtual machines can reduce resource fragmentation during actual resource scheduling of physical machines and help improve resource utilization of physical machines.

In the embodiment of this application, after determining the specifications and quantity of dedicated virtual machines and shared virtual machines that can be mixedly deployed on physical machines, the resources occupied by the dedicated virtual machines and shared virtual machines can also be isolated. . Specifically, the management and control node 10 may divide the total resources into a first resource pool and a second resource pool according to the resource specifications and quantities of the first type virtual machines and the second type virtual machines supported by the physical machine. The first resource pool is used to support the deployment of a first type of virtual machine, and the second resource pool is used to support the deployment of a second type of virtual machine. Among them, the resource pool that supports the deployment of shared virtual machines can be called a shared resource pool; the resource pool that supports the deployment of exclusive virtual machines can be called an exclusive resource pool. Further, the management and control node 10 can isolate the first resource pool and the second resource pool for subsequent deployment of the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool respectively. In this way, a weak mixture of shared virtual machines and exclusive virtual machines is achieved. That is, as shown in Figure 5 and Figure 6, dedicated virtual machines and shared virtual machines have separate and independent resource pools respectively. Each type of virtual machine can only schedule resource pools that support the deployment of that type of virtual machine for virtualization. machine deployment.

In other embodiments, a forced mixing of shared virtual machines and exclusive virtual machines can also be implemented. As shown in Figure 6 , in the implementation of the strong mixing department, the resources of the physical machine 20 are dynamically divided in real time rather than physically isolated, and the scheduler can be used to ensure the SLA requirements of the exclusive virtual machine.

For the above embodiment of the weak hybrid part, for the shared resource pool in the first resource pool and the second resource pool that supports shared virtual machine deployment, the physical cores in the shared resource pool can be virtualized to obtain Multiple logical cores are used for shared virtual machines that are subsequently deployed in the shared resource pool to share the physical cores in the shared resource pool.

In the embodiment of this application, the specific implementation manner of virtualizing the physical cores in the shared resource pool is not limited. In some embodiments, the number M of logical cores that need to be virtualized can be determined based on the set virtualization ratio. M≥2, and is an integer. Optionally, the corresponding relationship between the model of the physical machine and the virtual scale can be preset. Based on this corresponding relationship, in this embodiment, the management and control node 10 can determine the virtual proportion corresponding to the physical machine 20 according to the model of the machine. Furthermore, the number M of logical cores that need to be virtualized can be determined according to the virtual ratio corresponding to the machine model. Further, as shown in FIG. 6 , the management and control node 10 can use hyper-threading technology to virtualize each physical core in the second resource into M logical cores. In this way, for the shared virtual machine in the future, the number of logical cores required by the shared virtual machine can be determined based on the resource specifications of the shared virtual machine; and the corresponding number of logical cores can be used to deploy the shared virtual machine in the shared resource pool.

It is worth noting that for the physical machine 20 with a virtual machine hybrid deployment mode, the number of shared resource pools may be one or more. Multiple means 2 or more. Multiple shared resource pools correspond to different virtual ratios, that is, the number M of logical cores virtualized from physical cores in different shared resource pools is different. The number of shared resource pools in the same physical machine and the virtual ratio supported by each shared resource pool can be flexibly set according to actual needs. For example, the number of shared resource pools and the virtual ratio corresponding to each shared resource pool can be set according to the specifications of the shared virtual machine that the user actually applies to deploy.

The above embodiment predetermines the virtual machine deployment method of the physical machine, aiming at supporting shared virtual machines and exclusive virtual machines. For physical machines deployed with mixed virtual machines, the resource specifications and quantity of shared virtual machines supported by the physical machine can also be determined in advance, as well as the resource specifications and quantity of exclusive virtual machines. After determining these, the subsequent management and control node 10 can perform resource scheduling on the physical machine 20 according to the virtual machine deployment request, so as to deploy a virtual machine that satisfies the virtual machine deployment request in the physical machine 20 . The following is an exemplary description of the resource scheduling method in the virtual machine deployment process provided by the embodiments of this application.

As shown in Figure 7, the client 30 can provide a virtual machine deployment request to the management and control node 10. The virtual machine deployment request may include the type of virtual machine to be deployed (exclusive virtual machine or shared virtual machine), target resource specifications and quantity, etc. The management and control node 10 can obtain a virtual machine deployment request. Further, the management and control node 10 can obtain the target type and target resource specifications of the virtual machine to be deployed from the virtual machine deployment request; and then select a physical machine from the physical machines 20 that supports the target type and target resource specifications. Further, the management and control node 10 can select a target physical machine whose idle resources meet the target resource specification requirements based on the idle resources in the target resource pool in the physical machine that supports the target type and target resource specification; and deploy the target physical machine in the target resource pool of the target physical machine. Deploy virtual machines. The target resource pool is a resource pool that supports the deployment of virtual machines of the target type. For example, the target type is an exclusive virtual machine and the target resource pool is an exclusive resource pool. If the target type is a shared virtual machine, the target resource pool is a shared resource pool.

In some embodiments, the target type is a shared virtual machine. Correspondingly, the target resource pool is a shared resource pool. The resource specifications of shared virtual machines supported by different physical machines may be the same or different. Of course, the virtual ratios corresponding to the shared resource pools of different physical machines may be the same or different. In some embodiments, the correspondence between the resource specifications and the virtual ratio of the shared virtual machine may be preset. Based on the corresponding relationship, the management and control node 10 can determine the target virtual ratio corresponding to the target resource specification included in the virtual machine deployment request, and select a shared resource pool that supports the target virtual ratio from the shared resource pool as the target resource pool.

Further, the target logical core that meets the processor core number requirement of the target resource specification can be determined from the logical cores of the shared resource pool as the target resource pool; and other resources that meet the target resource specification can be obtained from other resources in the shared resource pool. Other types of resources according to resource requirements; then, shared virtual machines can be deployed in the shared resource pool using the target logical core and other types of resources mentioned above.

In another embodiment, the target type is a dedicated virtual machine. Accordingly, the target resource pool is an exclusive resource pool. Correspondingly, the management and control node 10 can select a target physical machine whose idle resources meet the target resource specification requirements from the idle resources in the exclusive resource pool among the physical machines that support the target type and target resource specifications; and in the exclusive resource pool of the target physical machine Deploy the virtual machine to be deployed in a resource pool.

Specifically, the management and control node 10 can determine the target physical core that meets the core number requirement of the processor in the target resource specification from the exclusive resource pool; and obtain other resources that meet the target resource specification from other resources in the shared resource pool. Other types of resources according to resource requirements; then, the target logical core and other types of resources mentioned above can be used to deploy exclusive virtual machines in the exclusive resource pool.

In this embodiment of the present application, in addition to pre-setting the virtual machine deployment mode of the physical machine, the management and control node 10 can also flexibly switch the virtual machine deployment mode of the physical machine according to the online real-time load condition of the physical machine. The following is an exemplary description of the specific implementation manner in which the management and control node 10 switches the virtual machine deployment mode of the physical machine.

As shown in Figure 8, in some embodiments, there are physical machines in the computing cluster 11 whose current virtual machine deployment mode is a mixed deployment mode of exclusive virtual machines and shared virtual machines. Accordingly, the management and control node 10 can monitor the calculation set Inventory levels of dedicated virtual machines and shared virtual machines in group 11. Among them, the inventory level of the exclusive virtual machine is used to represent the idle resources in the computing cluster 11 that can still deploy the exclusive virtual machine. It can be calculated that the idle resources of the exclusive resource pool in the cluster 11 account for the total number of exclusive resource pools. Expressed in proportion to resources. Of course, the inventory level of exclusive virtual machines can also be calculated by calculating the number of exclusive virtual machines that can be deployed in the exclusive resource pool in cluster 11, and the total number of exclusive virtual machines that can be deployed in the exclusive resource pool. Express the ratio between quantities, etc. In the same way, the inventory level of the shared virtual machine is used to indicate that the computing cluster 11 can still deploy the idle resources of the shared virtual machine. The ratio of the idle resources of the shared resource pool in the cluster 11 to the total resources of the shared resource pool can be calculated. express. Of course, the inventory level of shared virtual machines can also be calculated as the ratio between the number of shared virtual machines that can still be deployed in the shared resource pool of cluster 11 and the total number of shared virtual machines that can be deployed in the shared resource pool. Express etc.

If there are virtual machines of the target type whose inventory level is less than the set first water level threshold among the dedicated virtual machines and shared virtual machines of the computing cluster 11, and the inventory level of another type of virtual machine is greater than the set second water level threshold. , the management and control node 10 can select a target physical machine whose current virtual machine deployment mode is the virtual machine hybrid deployment mode from the computing cluster 11 . The number of target physical machines is less than or equal to some or all of the physical machines in the current virtual machine deployment mode in the computing cluster 11 . The number of target physical machines can be 1 or multiple. Multiple units refers to 2 or more units. The specific number of target physical machines can be determined based on the actual needs of the target type virtual machines and the idle resources of each virtual machine in the computing cluster 11 .

Specifically, the management and control node 10 can obtain the physical machines whose current virtual machine deployment mode is the virtual machine hybrid deployment mode from the computing cluster; and select the supported target type from the physical machines whose current virtual machine deployment mode is the virtual machine hybrid deployment mode. A physical machine with the resource specifications of the virtual machine; then, a physical machine with idle resources can be selected from the physical machines that support the resource specifications of the target type virtual machine as the target physical machine, and so on.

Furthermore, the virtual machine deployment mode of the target physical machine can be switched to the separate deployment mode of the target type virtual machine. Wherein, the first water level threshold is smaller than the second water level threshold. Generally, the first water level threshold is less than 50%; the second water level threshold is greater than 50%. Another type of virtual machine refers to another type of virtual machine other than the target type virtual machine among exclusive virtual machines and shared virtual machines. For example, if the target type of virtual machine is an exclusive virtual machine, the other type of virtual machine is a shared virtual machine. If the target type of virtual machine is a shared virtual machine, the other type of virtual machine is an exclusive virtual machine.

After switching the virtual machine deployment mode of the target physical machine to the target type virtual machine individual deployment mode, the target type virtual machine can be deployed separately on the target physical machine. For another type of virtual machine other than the target type virtual machine that has been deployed on the physical machine, in some embodiments, the management node 10 may migrate the other type of virtual machine to support the deployment of the other type of virtual machine. physical machine. In the current embodiment, in order to facilitate expression and distinction, the target physical machine in which the above-mentioned virtual machine deployment mode is switched to a separate deployment of the target type virtual machine is defined as the first physical machine; and another type of the first physical machine will be received. The physical machine to which the virtual machine is migrated is defined as the second physical machine. In the above or following embodiments, the other type refers to another virtual machine type between the exclusive virtual machine and the shared virtual machine except the target type virtual machine.

Optionally, the management node 10 can select a physical machine that supports the deployment of another type of virtual machine from other physical machines in the computing cluster 11 according to the resource specifications of another type of virtual machine deployed on the first physical machine; and select a physical machine that supports the deployment of another type of virtual machine; Among the physical machines deployed for another type of virtual machine, select a physical machine that supports the resource specifications of the other type of virtual machine; after that, select the physical machine that supports the resource specifications of the other type of virtual machine. Among the physical machines supporting the resource specifications of another type of virtual machine, a physical machine whose idle resources can meet the resource requirements supporting the resource specifications of another type of virtual machine is selected as the second physical machine. Further, resource specifications of another type of virtual machine deployed on the first physical machine may be migrated from the first physical machine to the second physical machine.

In other embodiments, the management node 10 may not process another type of virtual machine deployed on the first physical machine. Correspondingly, after the virtual machine deployment mode of the first physical machine is switched to the separate deployment mode of the target type virtual machine, the management and control node 10 can use the idle resources of the first physical machine to deploy the virtual machine of the target type, and configure the target type virtual machine on the first physical machine. Deployed virtual machines of another type are not processed.

The virtual machine deployment method currently supported by the physical machine 20 may be a mixed deployment of the above-mentioned shared virtual machines and dedicated virtual machines, or may be separate deployment of any type of virtual machine. As shown in Figure 9, for the physical machine 20 in which the current virtual machine deployment mode is to separately deploy a certain type of virtual machine, the management and control node 10 can obtain the virtual machine inventory level of the third physical machine in which the first type of virtual machine is separately deployed. Among them, the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. The third physical machine may be: some or all physical machines in which the current virtual machine deployment mode in the computing cluster 11 is to separately deploy the first type of virtual machine. The virtual machine inventory level of the third physical machine can be the ratio of the number of first-type virtual machines that can still be deployed on the third physical machine to the total number of first-type virtual machines supported by the third physical machine, etc.

Further, if the virtual machine inventory level of the first type virtual machine is less than the set first water level threshold, a third physical machine with idle resources that meets the resource requirements of the first type virtual machine can be selected from the physical machines that separately deploy the second type virtual machine. Four physical machines. Among them, the second type of virtual machine is an exclusive virtual machine or a shared virtual machine, which is different from the first type of virtual machine. In the embodiment of the present application, the specific value of the first lower water level threshold is not limited. Generally, the first water level threshold is less than 50%.

Optionally, the management and control node 10 may select a fifth physical machine whose model supports the first type of virtual machine from the physical machines where the second type of virtual machine is separately deployed; and select idle resources from the fifth physical machine that satisfy the requirements. The resource requirements of the first type of virtual machine are the fourth physical machine.

Further, the management and control node 10 can switch the virtual machine deployment mode of the fourth physical machine to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines; after that, the management and control node 10 can use the idle resources of the fourth physical machine, in The first type of virtual machine is deployed in the fourth physical machine to realize mixed deployment of exclusive virtual machines and shared virtual machines in the fourth physical machine. Since the inventory level of the first type of virtual machine in the third physical machine is low, and there are idle resources in the fourth physical machine where the second type of virtual machine is deployed separately, the virtual machine deployment mode of the fourth physical machine is switched to support The mixed deployment mode of exclusive virtual machines and shared virtual machines realizes the mixed deployment of exclusive virtual machines and shared virtual machines in the fourth physical machine, which helps to reduce idle resources in the fourth physical machine and improve the efficiency of the fourth physical machine. Resource utilization of four physical machines.

After switching the virtual machine deployment mode of the fourth physical machine to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines, the management and control node 10 may respond to the deployment request for the first type of virtual machine from the fourth physical machine. The target resource that satisfies the resources required by the deployment request is scheduled from the idle resources of the computer; further, the target resource can be used to deploy the first type of virtual machine in the fourth physical machine.

For the embodiment of weak hybrid deployment of virtual machines, the management and control node 10 deploys the virtual machine of the fourth physical machine. After the mode is switched to a mixed deployment mode that supports dedicated virtual machines and shared virtual machines, the idle resources of the fourth physical machine can also be used to build the first resource pool; and the second type of virtual machines can be deployed in the fourth physical machine. The occupied resources are used to construct a second resource pool; after that, the first resource pool and the second resource pool can be isolated for deployment of the first type of virtual machine and the second type of virtual machine in the first resource pool and the second resource pool respectively. Type virtual machine. For specific implementation methods of respectively deploying the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool, please refer to the relevant content of the above embodiments, and will not be described again here.

In some embodiments of the present application, for the third physical machine in which the first type virtual machine is separately deployed, the inventory level of the first type virtual machine corresponding to the third physical machine is less than or equal to the set second water level threshold continuously. If the duration is greater than or equal to the set duration threshold, the virtual machine deployment mode of the third physical machine can also be switched to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines. Further, the resource specification and quantity of the second type of virtual machines supported by the third physical machine can be determined with the goal of maximizing the resource utilization of the third physical machine. After that, the second type virtual machine can be deployed in the third physical machine in response to the deployment request for the second type virtual machine, thereby realizing the mixed deployment of the exclusive virtual machine and the shared virtual machine in the third physical machine. Reducing idle resources in the third physical machine helps improve resource utilization of the third physical machine.

In addition to the above system embodiments, embodiments of the present application also provide resource scheduling methods. The following is an exemplary description of the resource scheduling method provided by the embodiment of the present application.

Figure 10 is a schematic flowchart of a resource scheduling method provided by an embodiment of the present application. As shown in Figure 10, the method includes:

S101. Obtain the first resource occupied by the first physical machine for separately deploying the first type of virtual machine; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine.

S102. When the first resource is less than the total resources of the physical machine, set the virtual machine deployment mode of the first physical machine to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines.

S103. With the goal of maximizing total resource utilization, simulate and schedule the total resources of the physical machine according to the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine to determine the first type supported by the physical machine. The target resource specifications and quantities of the virtual machine and the second type of virtual machine; the second type of virtual machine is a type of exclusive virtual machine or shared virtual machine that is different from the first type of virtual machine.

S104. According to the target resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the first physical machine, perform a mixed deployment of exclusive virtual machines and shared virtual machines on the physical machine.

In this embodiment, for descriptions of the shared virtual machine and the exclusive virtual machine, please refer to the relevant content of the above system embodiment, and will not be described again here.

In this embodiment of the present application, in order to improve resource utilization, the resource usage of the physical machine can be scheduled and planned, that is, the type and specification of the virtual machine deployed on the physical machine can be determined in advance to maximize the resource utilization of the physical machine. In this embodiment of the present application, the virtual machine type is used to indicate whether the virtual machine is a shared virtual machine or an exclusive virtual machine.

Based on the above goal of improving physical machine resource utilization, in step S101, the resources required by the physical machine to separately deploy the first type of virtual machine can be obtained. Among them, the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. In the embodiment of this application, the specific implementation method of obtaining the resources required to deploy the first type of virtual machine on a physical machine is not limited.

In some embodiments, based on the preset resource specifications of the first type virtual machine, greedy boxing can be performed on the total resources of the physical machine to determine the resources occupied by the greedy boxed first type virtual machine, and determine the greedy boxing. The resources occupied by the first type of virtual machine in the box and the resources required to deploy the first type of virtual machine to the physical machine. The preset resource specification of the first type of virtual machine may be one specification or multiple specifications. Multiple means 2 or more types. The amount of resources in at least one dimension is different in different resource specifications.

Since the specifications of virtual machines provided by cloud vendors are certain and limited, the resources required by the virtual machine specifications may not be completely equal to the resources of the physical machine, which leads to resource fragmentation in the physical machine. Based on this, when it is determined that the resources occupied by the first type of virtual machine alone are deployed on the physical machine, it can also be determined whether resource fragmentation will occur if the physical machine alone deploys the first type of virtual machine. Specifically, it can be compared whether the resources occupied by the first type of virtual machine deployed alone on the physical machine are less than the total resources of the physical machine; if the resources occupied by the first type of virtual machine deployed alone on the physical machine are less than the total resources of the physical machine, it can be determined that the resources occupied by the physical machine alone are less than the total resources of the physical machine. Deploying the first type of virtual machine will create resource fragmentation.

In this embodiment, in order to reduce the resource fragmentation of the physical machine, in step S102, when the resources occupied by the first type of virtual machine deployed separately on the physical machine are less than the total resources of the physical machine, the virtual machines of the physical machine can be The deployment mode is set to support a mixed deployment mode of dedicated virtual machines and shared virtual machines. Correspondingly, if the resources occupied by the first type of virtual machine when deployed separately on the physical machine are equal to the total resources of the physical machine, the virtual machine deployment mode of the virtual machine can be set to support the independent deployment of the first type of virtual machine.

Further, in step S103, with the goal of maximizing the total resource utilization of the physical machine, simulation scheduling is performed on the total resources of the physical machine according to the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine, so as to Determine the target resource type and quantity of the first type of virtual machine and the second type of virtual machine supported by the physical machine. The above-mentioned second type of virtual machine is a type different from the first type of virtual machine among exclusive virtual machines and shared virtual machines. Regarding the specific implementation method of determining the target resource type and quantity of the first type virtual machine and the second type virtual machine supported by the physical machine, please refer to the relevant content of the above system embodiment, and will not be described again here.

After determining the resource specifications and quantity of the first type virtual machine and the second type virtual machine supported by the physical machine, in step S104, the resource specification and quantity of the first type virtual machine and the second type virtual machine can be determined. Physical machines are deployed with a mixture of dedicated virtual machines and shared virtual machines.

In the embodiment of this application, after determining the rules of dedicated virtual machines and shared virtual machines that can be mixedly deployed on physical machines, After specifying the size and quantity, you can also isolate the resources occupied by dedicated virtual machines and shared virtual machines. Specifically, the total resources may be divided into a first resource pool and a second resource pool according to the resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the physical machine. The first resource pool is used to support the deployment of a first type of virtual machine, and the second resource pool is used to support the deployment of a second type of virtual machine. Further, the first resource pool and the second resource pool may be isolated for subsequent deployment of the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool respectively. In this way, a weak mixture of shared virtual machines and exclusive virtual machines is achieved. That is, dedicated virtual machines and shared virtual machines each have separate and independent resource pools. Each type of virtual machine can only schedule virtual machine deployment in a resource pool that supports the deployment of that type of virtual machine.

In other embodiments, a forced mixing of shared virtual machines and exclusive virtual machines can also be implemented. In the implementation of strong hybrid deployment, the resources of physical machines are dynamically divided in real time rather than physically isolated, and the scheduler can be used to ensure the SLA requirements of exclusive virtual machines.

In the embodiment of this application, the specific implementation manner of virtualizing the physical cores in the shared resource pool is not limited. In some embodiments, the number M of logical cores that need to be virtualized can be determined based on the set virtualization ratio. M≥2, and is an integer. Optionally, the corresponding relationship between the model of the physical machine and the virtual scale can be preset. Based on this corresponding relationship, in this embodiment, the virtual proportion corresponding to the physical machine model can be determined according to the model of the physical machine. Furthermore, the number M of logical cores that need to be virtualized can be determined according to the virtual ratio corresponding to the machine model. Further, hyper-threading technology can be used to virtualize each physical core in the second resource into M logical cores. In this way, for the shared virtual machine in the future, the number of logical cores required by the shared virtual machine can be determined based on the resource specifications of the shared virtual machine; and the corresponding number of logical cores can be used to deploy the shared virtual machine in the shared resource pool.

It is worth noting that for physical machines with virtual machine hybrid deployment mode, the number of shared resource pools can be one or more. Multiple means 2 or more. Multiple shared resource pools correspond to different virtual ratios, that is, the number M of logical cores virtualized from physical cores in different shared resource pools is different.

The above embodiment predetermines the virtual machine deployment method of the physical machine. For the physical machine that supports the mixed deployment of shared virtual machines and exclusive virtual machines, the resource specifications and quantity of the shared virtual machines supported by the physical machine can also be predetermined. And, the resource specifications and quantity of exclusive virtual machines. After these are determined, resource scheduling can be performed on the physical machine according to the virtual machine deployment request, so that virtual machines that meet the virtual machine deployment request can be deployed in the physical machine. The following is an exemplary description of the resource scheduling method in the virtual machine deployment process provided by the embodiments of this application.

The client can provide a virtual machine deployment request to the management and control node. The virtual machine deployment request may include the type of virtual machine to be deployed (exclusive virtual machine or shared virtual machine), target resource specifications and quantity, etc. For management and control nodes, virtual machine deployment requests can be obtained. Further, the target type and target resource specifications of the virtual machine to be deployed can be obtained from the virtual machine deployment request; then, a physical machine that supports the target type and target resource specifications can be selected from the physical machines of the computing system. Furthermore, the target resource pool in the physical machine that supports the target type and target resource specifications can be of idle resources, select a target physical machine whose idle resources meet the target resource specification requirements; and deploy the virtual machine to be deployed in the target resource pool of the target physical machine. The target resource pool is a resource pool that supports the deployment of virtual machines of the target type. For example, the target type is an exclusive virtual machine and the target resource pool is an exclusive resource pool. If the target type is a shared virtual machine, the target resource pool is a shared resource pool. Regarding the specific implementation of deploying exclusive virtual machines and shared virtual machines in the target resource pool, please refer to the relevant content of the above system embodiments, and will not be described again here.

In this embodiment of the present application, in addition to pre-setting the virtual machine deployment mode of the physical machine, the management and control node can also flexibly switch the virtual machine deployment mode of the physical machine according to the online real-time load condition of the physical machine. The following is an exemplary description of the specific implementation method of the management and control node switching the virtual machine deployment mode of the physical machine.

In some embodiments, there are physical machines in the computing cluster whose current virtual machine deployment mode is a mixed deployment mode of dedicated virtual machines and shared virtual machines. Accordingly, the inventory levels of dedicated virtual machines and shared virtual machines in the computing cluster can be monitored. For the description of the inventory water level, please refer to the relevant content of the above embodiments, and will not be described again here.

If there are virtual machines of the target type in the dedicated virtual machines and shared virtual machines of the computing cluster whose inventory level is less than the set first water level threshold, and the inventory level of another type of virtual machine is greater than the set second water level threshold, From the computing cluster, you can select a target physical machine whose current virtual machine deployment mode is virtual machine mixed deployment mode; and switch the virtual machine deployment mode of the target physical machine to the target type virtual machine separate deployment mode. Wherein, the first water level threshold is smaller than the second water level threshold. Generally, the first water level threshold is less than 50%; the second water level threshold is greater than 50%. Another type of virtual machine refers to another type of virtual machine other than the target type virtual machine among exclusive virtual machines and shared virtual machines. For example, if the target type of virtual machine is an exclusive virtual machine, the other type of virtual machine is a shared virtual machine. If the target type of virtual machine is a shared virtual machine, the other type of virtual machine is an exclusive virtual machine.

After switching the virtual machine deployment mode of the target physical machine to the target type virtual machine individual deployment mode, the target type virtual machine can be deployed separately on the target physical machine. For a virtual machine of another type other than the target type virtual machine that has been deployed on the target physical machine (defined as the first physical machine), in some embodiments, the virtual machine of the other type may be migrated to support the other type of virtual machine. A virtual machine of one type is deployed on another physical machine (defined as the second physical machine). Optionally, according to the resource specifications of another type of virtual machine deployed on the first physical machine, a physical machine that supports the deployment of another type of virtual machine can be selected from other physical machines in the computing cluster; and from other physical machines that support the deployment of another type of virtual machine; Among the physical machines for virtual machine deployment, select a physical machine that supports the resource specifications of another type of virtual machine; then, select idle resources from the physical machines that support the resource specifications of another type of virtual machine to meet the above requirements of supporting another type of virtual machine. The resource specifications of the type of virtual machine are the resource requirements of the physical machine, as the second physical machine. Further, resource specifications of another type of virtual machine deployed on the first physical machine may be migrated from the first physical machine to the second physical machine.

In other embodiments, another type of virtual machine deployed on the first physical machine may not be processed. Correspondingly, after the virtual machine deployment mode of the first physical machine is switched to the separate deployment mode of the target type virtual machine, the idle resources of the first physical machine can be used to deploy the target type virtual machine, and another virtual machine deployed on the first physical machine can be deployed. One type of virtual machine is not processed.

The virtual machine deployment methods currently supported by physical machines may be a mixed deployment of the above shared virtual machines and dedicated virtual machines, or they may be deployed separately for either type of virtual machine. The following is an exemplary description of the resource scheduling method of a physical machine in which a certain type of virtual machine is separately deployed in the current virtual machine deployment mode with reference to Figure 11.

Figure 11 is a schematic flowchart of another resource scheduling method provided by an embodiment of the present application. As shown in Figure 11, the method includes:

S111. Obtain the virtual machine inventory level of the third physical machine in which the first type of virtual machine is separately deployed from the computing cluster; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine.

S112. When the virtual machine inventory level is less than the set first water level threshold, select a fourth physical machine whose idle resources meet the resource requirements of the first type virtual machine from the physical machines where the second type virtual machine is deployed separately; The second type of virtual machine is a type of exclusive virtual machine or shared virtual machine that is different from the first type of virtual machine.

S113. Switch the virtual machine deployment mode of the fourth physical machine to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines.

S114. Use the idle resources of the fourth physical machine to deploy the first type of virtual machine in the fourth physical machine.

In this embodiment, for a physical machine in which a certain type of virtual machine is deployed separately in the current virtual machine deployment mode, the virtual machine inventory level of the physical machine (defined as the third physical machine) on which the first type of virtual machine is deployed separately can be obtained. . Among them, the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. The third physical machine is all or part of the physical machine on which the first type of virtual machine is separately deployed in the computing cluster.

Further, if the virtual machine inventory level of the first type virtual machine is less than the set first water level threshold, in step S112, idle resources can be selected from physical machines that separately deploy the second type virtual machine to satisfy the requirements of the first type virtual machine. The resource requirements of the fourth physical machine. Among them, the second type of virtual machine is an exclusive virtual machine or a shared virtual machine, which is different from the first type of virtual machine. In the embodiment of the present application, the specific value of the first lower water level threshold is not limited. Generally, the first water level threshold is less than 50%.

Optionally, a fifth physical machine whose model supports the first type of virtual machine can be selected from the physical machines that separately deploy the second type of virtual machine; and from the fifth physical machine, select idle resources that meet the requirements of the first type of virtual machine. The resource requirements of a virtual machine are those of a fourth physical machine. Among them, the fifth physical machine includes: the fourth physical machine.

Further, in step S113, the virtual machine deployment mode of the fourth physical machine can be switched to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines; and then, in step S114, the virtual machine deployment mode of the fourth physical machine can be used. Use idle resources to deploy the first type of virtual machine in the fourth physical machine to achieve mixed deployment of exclusive virtual machines and shared virtual machines in the fourth physical machine. Since the inventory level of the first type of virtual machine in the third physical machine is low, and there are idle resources in the fourth physical machine where the second type of virtual machine is deployed separately, the virtual machine deployment mode of the fourth physical machine is switched to support The mixed deployment mode of exclusive virtual machines and shared virtual machines realizes the mixed deployment of exclusive virtual machines and shared virtual machines in the fourth physical machine, which helps to reduce idle resources in the fourth physical machine and improve the efficiency of the fourth physical machine. Resource utilization of four physical machines.

After the virtual machine deployment mode of the fourth physical machine is switched to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines, in response to a deployment request for the first type of virtual machine, the idle state of the fourth physical machine can be Target resources that meet the resources required by the deployment request are scheduled among the resources; further, the target resources can be used to deploy the first type of virtual machine in the fourth physical machine.

For the embodiment of weak hybrid deployment of virtual machines, after switching the virtual machine deployment mode of the fourth physical machine to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines, the fourth physical machine can also be used. Idle resources are used to construct a first resource pool; and the resources occupied by the second type of virtual machine deployed in the fourth physical machine are used to construct a second resource pool; after that, the first resource pool and the second resource pool can be isolated to The first type virtual machine and the second type virtual machine are respectively deployed in the first resource pool and the second resource pool. For specific implementation methods of respectively deploying the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool, please refer to the relevant content of the above embodiments, and will not be described again here.

In some embodiments of the present application, for the above-mentioned third physical machine where the first type virtual machine is separately deployed, the inventory level of the first type virtual machine in the third physical machine is less than or equal to the duration of the set second water level threshold. If the duration is greater than or equal to the set duration threshold, the virtual machine deployment mode of the third physical machine can also be switched to a mixed deployment mode that supports exclusive virtual machines and shared virtual machines. Further, the resource specification and quantity of the second type of virtual machines supported by the first physical machine can be determined with the goal of maximizing the resource utilization of the third physical machine. After that, the second type virtual machine can be deployed in the third physical machine in response to the deployment request for the second type virtual machine, thereby realizing the mixed deployment of the exclusive virtual machine and the shared virtual machine in the third physical machine. Reducing idle resources in the third physical machine helps improve resource utilization of the third physical machine.

It should be noted that the execution subject of each step of the method provided in the above embodiments may be the same device, or the method may also be executed by different devices. For example, the execution subject of steps S101 and S102 may be device A; for another example, the execution subject of step S101 may be device A, the execution subject of step S102 may be device B; and so on.

In addition, some of the processes described in the above embodiments and drawings include multiple operations that appear in a specific order, but it should be clearly understood that these operations may not be performed in the order in which they appear in this document or may be performed in parallel. , the sequence numbers of operations, such as S101, S102, etc., are only used to distinguish different operations. The sequence numbers themselves do not represent any execution order. Additionally, these processes may include more or fewer operations, and the operations may be performed sequentially or in parallel.

Correspondingly, embodiments of the present application also provide a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute each of the above resource scheduling methods. step.

Figure 12 is a schematic structural diagram of a computing device provided by an embodiment of the present application. As shown in Figure 12, the computing device includes: a memory 12a and a processor 12b; wherein the memory 12a is used to store computer programs.

The processor 12b is coupled to the memory 12a and is used to execute a computer program for: obtaining the first resource occupied by the first physical machine to separately deploy the first type of virtual machine; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. machine; when the first resource is less than the total resources of the physical machine, set the virtual machine deployment mode of the first physical machine In order to support the mixed deployment mode of dedicated virtual machines and shared virtual machines; with the goal of maximizing total resource utilization, the physical machine is configured based on the resource specifications of the first type of virtual machine and the resource specifications of the second type of virtual machine. The total resources are simulated and scheduled to determine the target resource specifications and quantities of the first type of virtual machine and the second type of virtual machine supported by the physical machine; the second type of virtual machine is an exclusive virtual machine and a shared virtual machine that are the same as the first type. Different types of virtual machines; according to the target resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the first physical machine, perform a mixed deployment of exclusive virtual machines and shared virtual machines on the physical machine; Among them, an exclusive virtual machine refers to a virtual machine in which a virtual processor exclusively occupies the physical processor resources of a physical machine; a shared virtual machine refers to a virtual machine in which a virtual processor shares physical processor resources.

Optionally, when acquiring the first resources required by the first physical machine to deploy the first type virtual machine, the processor 12b is specifically configured to: based on the preset resource specifications of the first type virtual machine, configure the first physical machine based on the preset resource specifications of the first type virtual machine. The total resources are greedily boxed to determine the resources occupied by the first type of virtual machine in greedy boxing; the resources occupied by the first type of virtual machine in greedy boxing are determined as the first resources.

Optionally, the processor 12b is also configured to: divide the total resources into a first resource pool and a second resource pool according to the target resource specifications and quantity of the first type of virtual machine supported by the first physical machine; Perform isolation processing with the second resource pool to deploy the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool respectively.

Further, the physical processor resources are physical cores of the physical processor. The processor 12b is also configured to perform virtualization processing on the physical cores in the shared resource pool in which the shared virtual machine is deployed in the first resource pool and the second resource pool to obtain multiple logical cores for use. Shared virtual machines deployed in a shared resource pool share the physical cores in the shared resource pool.

Optionally, when the processor 12b performs virtualization processing on the physical cores in the shared resource pool, it is specifically used to: determine the number M of logical cores that need to be virtualized according to the set virtual ratio; the virtual ratio represents 1 physical core. The ratio of cores to virtualized logical cores; hyper-threading technology is used to virtualize each physical core in the second resource into M logical cores; M≥2, and is an integer.

Among them, the number of shared resource pools is 1 or more. Multiple means 2 or more. Multiple shared resource pools correspond to different virtual ratios.

In some embodiments, the processor 12b is also configured to: monitor the inventory levels of exclusive virtual machines and shared virtual machines in the computing cluster; When the inventory water level is less than the set first water level threshold and the inventory level of another type of virtual machine is greater than the set second water level threshold, select the target of the current virtual machine deployment mode as the virtual machine hybrid deployment mode from the computing cluster Physical machine; switch the virtual machine deployment mode of the target physical machine to the virtual machine independent deployment mode of the first target type.

Optionally, the processor 12b is also configured to: in the process of switching the virtual machine deployment mode of the target physical machine to the virtual machine independent deployment mode of the first target type, migrate another type of virtual machine deployed on the target physical machine. to a second physical machine that supports the deployment of another type of virtual machine; or, after switching the virtual machine deployment mode of the target physical machine to a separate virtual machine deployment mode of any type, use the idle resources of the target physical machine to deploy the first Target type of virtual machine.

In other embodiments, the processor 12b is also used to: obtain a virtual machine deployment request; obtain the second target type and target resource specification of the virtual machine to be deployed from the virtual machine deployment request; obtain the second target type and target resource specifications of the virtual machine to be deployed; and obtain the target resource specification from the physical machine of the computing cluster. Determine that there is a target resource pool that supports the second target type and a target physical machine that supports the target resource specifications; and deploy the to-be-deployed virtual machine of the second target type in the target resource pool of the target physical machine according to the target resource specifications.

The computing device provided in this embodiment realizes the mixed deployment of exclusive virtual machines and shared virtual machines by pre-planning the specifications and quantities of exclusive virtual machines and shared virtual machines that can be mixedly deployed on the physical machine. Helps reduce the amount of natural resource fragmentation caused by limited virtual machine specifications. On the other hand, because the resource specifications and quantities of the first type of virtual machines and the second type of virtual machines supported by the physical machine are pre-scheduled by simulating the total resources of the physical machine with the goal of maximizing the resource utilization of the physical machine. , Therefore, deploying the simulated number and specifications of dedicated virtual machines and shared virtual machines in the physical machine can reduce resource fragmentation during actual resource scheduling of the physical machine and help improve the resource utilization of the physical machine.

In the following embodiment of the present application, the processor 12b is also used to: obtain the virtual machine inventory level of the first physical machine in the computing cluster where the first type of virtual machine is separately deployed; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine. type of virtual machine; when the virtual machine inventory level is less than the set first water level threshold, select a second physical machine with idle resources that meets the resource requirements of the first type of virtual machine from the physical machines that separately deploy the second type of virtual machine. machine; the second type of virtual machine is a type of exclusive virtual machine and shared virtual machine that is different from the first type of virtual machine; switch the virtual machine deployment mode of the second physical machine to support exclusive virtual machines and shared virtual machines Hybrid deployment mode of virtual machines; utilizing the idle resources of the second physical machine, deploying the first type of virtual machine in the second physical machine; wherein, the exclusive virtual machine refers to a virtual processor that exclusively occupies the physical processor resources of the physical machine. Virtual machine; a shared virtual machine refers to a virtual machine in which a virtual processor shares physical processor resources.

Optionally, when selecting a second physical machine whose idle resources meet the resource requirements of the first type of virtual machine, the processor 12b is specifically configured to: select a model that supports the first type of physical machine from the physical machines that separately deploy the second type of virtual machine. a third physical machine of type virtual machine; and select a second physical machine whose idle resources meet the resource requirements of the first type virtual machine from the third physical machine.

In some embodiments, when the processor 12b deploys the first type of virtual machine in the second physical machine, it is specifically configured to: in response to a deployment request for the first type of virtual machine, schedule from idle resources of the second physical machine that satisfies the requirements. Deploy the target resource required by the requested resource; use the target resource to deploy the first type of virtual machine in the second physical machine.

Optionally, the processor 12b is also configured to: utilize the idle resources of the second physical machine to construct a first resource pool;

Using the resources occupied by the second type of virtual machine deployed in the second physical machine, a second resource pool is constructed; the first resource pool and the second resource pool are isolated for use in the first resource pool and the second resource pool respectively. Deploy the first type of virtual machine and the second type of virtual machine.

In the computing device provided by this embodiment, since the inventory level of the first type virtual machine in the first physical machine is low and there are idle resources in the second physical machine where the second type virtual machine is separately deployed, the second The virtual machine deployment mode of the physical machine is switched to a mixed deployment mode that supports dedicated virtual machines and shared virtual machines. This enables mixed deployment of dedicated virtual machines and shared virtual machines in the second physical machine, which helps reduce the cost of the second physical machine. The idle resources in the second physical machine improve the resource utilization of the second physical machine.

In some optional implementations, as shown in Figure 12, the computing device may also include: a communication component 12c, a power supply component 12d and other optional components. Only some components are schematically shown in FIG. 12 , which does not mean that the computing device must include all the components shown in FIG. 12 , nor does it mean that the computing device can only include the components shown in FIG. 12 .

In the embodiment of the present application, the memory is used to store computer programs, and can be configured to store various other data to support operations on the device where it is located. Among them, the processor can execute the computer program stored in the memory to implement the corresponding The logic should be controlled. Memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable memory Read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

In this embodiment of the present application, the processor may be any hardware processing device that can execute the logic of the above method. Optionally, the processor can be a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) or a microcontroller unit (Microcontroller Unit, MCU); it can also be a field programmable gate array (Field -Programmable Gate Array (FPGA), Programmable Array Logic (PAL), General Array Logic (GAL), Complex Programmable Logic Device (CPLD) and other programmable devices ; Or advanced reduced instruction set (RISC) processors (Advanced RISC Machines, ARM) or system on chip (System on Chip, SoC), etc., but are not limited to this.

In the embodiment of the present application, the communication component is configured to facilitate wired or wireless communication between the device where it is located and other devices. The device where the communication component is located can access a wireless network based on communication standards, such as WiFi, 2G or 3G, 4G, 5G or their combination. In an exemplary embodiment, the communication component receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component may also be based on near field communication (NFC) technology, radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology or other technologies to achieve this.

In the embodiment of the present application, the power supply component is configured to provide power to various components of the device where it is located. A power component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which the power component resides.

It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent the order, nor do they limit "first" and "second" are different types.

Those skilled in the art will understand that embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce a use A device for realizing the functions specified in one process or multiple processes of the flowchart and/or one block or multiple blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions The device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device. Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-permanent storage in computer-readable media, random access memory (RAM) and/or non-volatile memory in the form of read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

The storage medium of a computer is a readable storage medium, which can also be called a readable medium. Readable storage media includes permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology to store information. Information may be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), and read-only memory. (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cassettes, disk storage or other magnetic storage devices, or any other non-transmission medium, can be used to store information that can be accessed by a computing device. As defined in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprises," or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements not only includes those elements, but also includes Other elements are not expressly listed or are inherent to the process, method, article or equipment. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article, or device that includes the stated element.

The above descriptions are only examples of the present application and are not intended to limit the present application. To those skilled in the art, various modifications and variations may be made to this application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application shall be included in the scope of the claims of this application.

Claims

A resource scheduling method, characterized by including:

Obtaining the first resources occupied by the first physical machine for separately deploying the first type of virtual machine; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine;

When the first resource is less than the total resources of the physical machine, set the virtual machine deployment mode of the first physical machine to a hybrid deployment mode that supports exclusive virtual machines and shared virtual machines;

With the goal of maximizing the total resource utilization, simulation scheduling is performed on the total resources of the physical machine according to the resource specifications of the first type virtual machine and the resource specifications of the second type virtual machine to determine the physical machine's total resources. The target resource specifications and quantities of the first type of virtual machine and the second type of virtual machine supported by the machine; the second type of virtual machine is the same as the first type of the exclusive virtual machine and the shared virtual machine. Different types of virtual machines;

According to the target resource specifications and quantities of the first type virtual machine and the second type virtual machine supported by the first physical machine, perform a mixed deployment of exclusive virtual machines and shared virtual machines on the physical machine;

The exclusive virtual machine refers to a virtual machine in which a virtual processor exclusively occupies the physical processor resources of the physical machine; the shared virtual machine refers to a virtual machine in which a virtual processor shares physical processor resources.
The method of claim 1, wherein obtaining the first resources required by the first physical machine to deploy the first type of virtual machine includes:

Based on the preset resource specifications of the first type virtual machine, perform greedy boxing on the total resources of the first physical machine to determine the resources occupied by the greedy boxed first type virtual machine;

The resources occupied by the greedy boxed first type virtual machine are determined as the first resources.
The method according to claim 1, further comprising:

Divide the total resources into a first resource pool and a second resource pool according to the target resource specifications and quantity of the first type of virtual machine supported by the first physical machine;

Perform isolation processing on the first resource pool and the second resource pool to deploy the first type virtual machine and the second type virtual machine in the first resource pool and the second resource pool respectively. machine.
The method according to claim 3, characterized in that the physical processor resource is a physical core of a physical processor; for the shared resources of the shared virtual machine deployed in the first resource pool and the second resource pool pool, the method further includes:

The physical cores in the shared resource pool are virtualized to obtain multiple logical cores, so that the shared virtual machines deployed in the shared resource pool can share the physical cores in the shared resource pool.
The method according to claim 4, characterized in that said virtualizing the physical cores in the shared resource pool to obtain multiple logical cores includes:

According to the set virtual ratio, determine the number M of logical cores that need to be virtualized; the virtual ratio represents the ratio of one physical core to the virtualized logical core;

Hyper-threading technology is used to virtualize each physical core in the second resource into M logical cores; M≥2, and is an integer.
The method according to claim 5, characterized in that the number of the shared resource pools is multiple; the virtual ratios corresponding to the multiple shared resource pools are different.
The method according to claim 1, further comprising:

Monitor the inventory levels of dedicated virtual machines and shared virtual machines in the computing cluster;

In the exclusive virtual machine and the shared virtual machine, the virtual machine inventory level of the first target type is less than the set first water level threshold, and the virtual machine inventory level of the other type is greater than the set second water level threshold. In the case of , select the target physical machine whose current virtual machine deployment mode is the virtual machine hybrid deployment mode from the computing cluster;

Switch the virtual machine deployment mode of the target physical machine to the virtual machine independent deployment mode of the first target type.
The method according to claim 7, further comprising:

In the process of switching the virtual machine deployment mode of the target physical machine to the virtual machine independent deployment mode of the first target type, migrating another type of virtual machine deployed on the target physical machine to support the other type. A second physical machine for deploying a type of virtual machine;

or,

After the virtual machine deployment mode of the target physical machine is switched to the virtual machine independent deployment mode of the first target type, the virtual machine of the first target type is deployed using idle resources of the target physical machine.
The method according to claim 3, further comprising:

Get the virtual machine deployment request;

Obtain the second target type and target resource specifications of the virtual machine to be deployed from the virtual machine deployment request;

Determine from the physical machines of the computing cluster that there is a target resource pool that supports the second target type and a target physical machine that supports the target resource specifications;

According to the target resource specification, the to-be-deployed virtual machine of the second target type is deployed in the target resource pool of the target physical machine.
A resource scheduling method, characterized by including:

Obtaining the virtual machine inventory level of the first physical machine in which the first type of virtual machine is separately deployed in the computing cluster; the first type of virtual machine is an exclusive virtual machine or a shared virtual machine;

When the virtual machine inventory level is less than the set first water level threshold, select a second physical machine with idle resources that meets the resource requirements of the first type virtual machine from the physical machines where the second type virtual machine is deployed separately. machine; the second type virtual machine is a type different from the first type virtual machine among the exclusive virtual machine and the shared virtual machine;

Switch the virtual machine deployment mode of the second physical machine to a hybrid deployment mode that supports exclusive virtual machines and shared virtual machines;

Utilizing the idle resources of the second physical machine, deploy the first type virtual machine in the second physical machine;

The exclusive virtual machine refers to a virtual machine in which a virtual processor exclusively occupies the physical processor resources of the physical machine; the shared virtual machine refers to a virtual machine in which a virtual processor shares the physical processor resources.
The method according to claim 10, characterized in that selecting a second physical machine whose idle resources meet the resource requirements of the first type virtual machine from the physical machines in which the second type virtual machine is deployed separately includes:

Select a third physical machine whose model supports the first type virtual machine from the physical machines on which the second type virtual machine is separately deployed;

Select a second physical machine whose idle resources meet the resource requirements of the first type virtual machine from the third physical machine.
The method according to claim 10, further comprising:

Utilize the idle resources of the second physical machine to construct a first resource pool;

Utilize the resources occupied by the second type of virtual machine deployed in the second physical machine to construct a second resource pool;

Perform isolation processing on the first resource pool and the second resource pool for respectively deploying the first type virtual machine and the second type in the first resource pool and the second resource pool. virtual machine.
A computing device, characterized by comprising: a memory and a processor; wherein the memory is used to store a computer program;

The processor is coupled to the memory for executing the computer program for performing the steps of the method of any one of claims 1-12.
A computer-readable storage medium storing computer instructions, characterized in that, when the computer instructions are executed by one or more processors, they cause the one or more processors to execute any one of claims 1-12 steps in the method.