CN106445631B - Method and system for deploying virtual machine and physical server - Google Patents

Abstract

The invention discloses a method, a system and a physical server for deploying virtual machines. The method includes: among multiple physical servers, determining a physical server with idle resources; When all the physical servers of the first binding rule, determine the target physical server, the target physical server satisfies some of the first binding rules in the at least one first binding rule, or the target physical server does not meet the Any one of at least one first binding rule; deploying the virtual machine to be deployed in the target physical server; adopting the method, system and physical server for deploying a virtual machine provided by the present invention , which can improve the utilization of physical servers.

Description

A method, system and physical server for deploying virtual machine

technical field

The present invention relates to the technical field of virtual machines, and in particular, to a method, a system and a physical server for deploying a virtual machine.

Background technique

With the rapid development of the Internet, cloud computing came into being; the so-called cloud computing is a computing model, as shown in Figure 1, the cloud computing system is mainly composed of a cloud computing platform and a client; the cloud computing platform is used to provide a unified physical The virtualized resources of the server can be allocated to the client according to the needs of the client. As shown in FIG. 2 , a scheduling system is usually deployed on the cloud computing platform, and the specific implementation of deploying virtual machines on the cloud computing platform is: the scheduling system deploys virtual machines in physical servers.

At present, before the virtual machine is deployed, a binding rule is usually set for each virtual machine in advance; the scheduling system of the cloud computing platform generally adopts the following method to arrange the virtual machine: First, determine whether there are idle resources in the cloud computing system. physical server, and then obtain the binding rules of the virtual machine to be deployed, according to the binding rules, from the physical servers with idle resources, find the physical server that satisfies the binding rules; finally, the virtual machine to be deployed is It is deployed in the found physical server; of course, if no physical server that satisfies the binding rule is found at this time, the deployment of the virtual machine to be deployed fails, and the virtual machine is no longer deployed.

However, in practical applications, using the above deployment method, there may be a situation in which physical servers that meet the conditions may not be found, but there are physical servers with idle resources, resulting in a waste of resources. For example, as shown in Figure 3, the cloud computing system provides 3 physical servers, and each physical server can deploy up to 4 virtual machines, and virtual machine 1 has been deployed in physical server 1, and there are still three idle resources left ; Virtual machine 2 has been deployed in physical server 2, and three idle resources are still left, while virtual machine 3, virtual machine 4, virtual machine 5 and virtual machine 6 have been deployed in physical server 3, and there are no idle resources; At this time, if the physical server 1 fails, the virtual machine 1 also fails. In order to ensure the normal operation of the virtual machine 1, the scheduling system needs to redeploy the virtual machine 1, and the preset binding rule of the virtual machine 1 is the same as that of the virtual machine. 2 cannot be deployed on the same physical server, then at this time, the scheduling system, using the above deployment method, will query the physical server with idle resources as physical server 2, but since physical server 2 has already deployed virtual machine 2, The binding rules of the virtual machine 1 are not satisfied. Therefore, the virtual machine 1 cannot be deployed on the physical server 2 at this time, and the reconstruction process of the virtual machine 1 fails.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method, a system and a physical server for deploying a virtual machine, so as to improve the utilization rate of the physical server.

In a first aspect, a method for deploying a virtual machine is provided, wherein at least one first binding rule is used as a first basis for a scheduling system to select a physical server for deploying a virtual machine to be deployed, and the method includes: among a plurality of physical servers , determine that there are physical servers with idle resources; when no physical servers that satisfy all the first binding rules are found in the physical servers that have idle resources, determine the target physical server, and the target physical server satisfies at least one of the first binding rules. Some of the first binding rules, or the target physical server does not satisfy any one of the at least one first binding rule; deploy the virtual machine to be deployed on the target physical server.

Because in this embodiment of the present invention, when a physical server that satisfies all the first binding rules is not found among the physical servers that have idle resources, the target physical server will be determined among the physical servers that have idle resources, and then the target physical server will be determined. The virtual machine is deployed in the target physical server; then, the method of the present invention improves the success rate of deploying the virtual machine on the physical server, thereby improving the resource utilization rate of the physical server.

With reference to the first aspect, in a first possible implementation manner, the number of the first binding rules is one, and when no physical servers that satisfy all the first binding rules are found in the physical servers with idle resources, determine The target physical server includes: selecting any physical server from the physical servers with idle resources as the target physical server when no physical server satisfying the first binding rule is found in the physical servers with idle resources. When a first restrictive rule is set, the above method can improve the resource utilization of the physical server.

With reference to the first aspect, in a second possible implementation manner, the number of the first binding rules is M, where M is a positive integer greater than or equal to 2, and the M first binding rules have priority; When a physical server that satisfies all the first binding rules is not found in the physical servers with idle resources, determining the target physical server, including: among the physical servers with idle resources, according to the priority of the M first binding rules, one by one The physical server is filtered by matching the first binding rule; if the first binding rule matching the first binding rule is filtered to the first physical server, and the first binding rule matching the second priority of the first binding rule is not selected from the first binding rule If a physical server is selected from the first physical servers, any physical server is selected from the first physical servers as the target physical server. When multiple first restrictive rules are set, the above method can improve the resource utilization of the physical server.

With reference to the first aspect, the first possible implementation manner of the first aspect, or the second possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the above method further includes: Among the physical servers with idle resources, search for physical servers that satisfy all the first restrictive rules in the at least one first restrictive rule; when finding out the physical servers that satisfy all the first restrictive rules in the at least one first restrictive rule, The virtual machines to be deployed are deployed in physical servers that satisfy all the first binding rules.

In combination with the first aspect, the first possible implementation manner of the first aspect, the second possible implementation manner of the first aspect, or the third possible implementation manner of the first aspect, the fourth possible implementation manner of the first aspect In the implementation of the method, when a physical server that satisfies all the first binding rules is not found in the physical servers that have idle resources, before determining the target physical server, the above method further includes: in the physical servers that have idle resources, searching for a physical server that satisfies all the first binding rules The physical server of the second binding rule, the second binding rule is used as the second basis for selecting the physical server for deploying the virtual machine to be deployed; in the physical server that satisfies the second binding rule, find the physical server that satisfies the first binding rule physical server. In the embodiment of the present invention, when the virtual machine is deployed, the user can also set the second binding rule that the physical server must satisfy according to the requirements, thereby making the setting of the binding rule of the virtual machine more flexible.

In a second aspect, the present invention provides a scheduling system for deploying virtual machines. A cloud computing platform is deployed on at least one physical server, and the cloud computing platform includes a scheduling system; at least one first binding rule is selected for deploying virtual machines to be deployed. The first basis for the physical server of the computer, the above-mentioned scheduling system includes: an idle resource determination unit, used for determining a physical server with idle resources among a plurality of physical servers; a target physical server determination unit, used for physical servers with idle resources. When a physical server that satisfies all the first binding rules is not found in the servers, the target physical server is determined, the target physical server satisfies some of the first binding rules in at least one first binding rule, or the target physical server does not meet at least one Any one of the first binding rules; a first deployment unit, configured to deploy the virtual machine to be deployed in the target physical server.

Because in this embodiment of the present invention, when a physical server that satisfies all the first binding rules is not found among the physical servers that have idle resources, the target physical server will be determined among the physical servers that have idle resources, and then the target physical server will be determined. The virtual machine is deployed in the target physical server; then, the system of the present invention improves the success rate of deploying the virtual machine on the physical server, thereby improving the resource utilization rate of the physical server.

With reference to the second aspect, in a first possible implementation manner, when the number of the first binding rules is one, the above-mentioned target physical server determination unit is specifically used for: not finding a physical server with idle resources When a physical server satisfies the first binding rule, select any physical server as the target physical server among the physical servers with idle resources. When a first restrictive rule is set, the above-mentioned system can improve the resource utilization rate of the physical server.

With reference to the second aspect, in the second possible implementation manner, the number of the first binding rules is M, where M is a positive integer greater than or equal to 2, and the M first binding rules have priority; the above goal The physical server determination unit is specifically used for: in the physical server with idle resources, according to the priority of the M first binding rules, one by one matching the first binding rules to filter the physical servers; The first physical server is selected from the first physical server, and the first binding rule matching the second priority of the first first binding rule is not selected from the first physical server, then any physical server is selected from the first physical server. A physical server serves as the target physical server. When multiple first restrictive rules are set, the above-mentioned system can improve the resource utilization rate of the physical server.

In combination with the second aspect, the first possible implementation manner of the second aspect, and the second possible implementation manner of the second aspect, in a third possible implementation manner, the scheduling system further includes: a first search unit, is used to search for physical servers that satisfy all the first restrictive rules in the at least one first restrictive rule among the physical servers that have idle resources; the second deployment unit is used to find out the physical servers that satisfy the at least one first restrictive rule When the physical servers that meet all the first binding rules are in, deploy the virtual machine to be deployed on the physical servers that satisfy all the first binding rules.

In combination with the second aspect, the first possible implementation manner of the second aspect, the second possible implementation manner of the second aspect, or the third possible implementation manner of the second aspect, in the fourth possible implementation manner , the above-mentioned scheduling system further includes: a second search unit, used to search for a physical server that satisfies a second binding rule among the physical servers with idle resources, and the second binding rule is used to select the physical server for deploying the virtual machine to be deployed. The second basis of the server; the third search unit is configured to search for a physical server that satisfies the first binding rule among the physical servers that meet the second binding rule. In the embodiment of the present invention, when the virtual machine is deployed, the user can also set the second binding rule that the physical server must satisfy according to the requirements, thereby making the setting of the binding rule of the virtual machine more flexible.

In a third aspect, the present invention also provides a physical server, the physical server includes at least one processor, memory and communication interface; at least one processor, memory and communication interface are all connected through a bus; the memory is used for storing computer execution instructions; At least one processor for executing computer-executed instructions stored in the memory, so that the physical server executes the method for deploying a virtual machine provided by the first aspect or various possible designs of the first aspect, or enables the physical server to communicate with cloud computing through a communication interface Other physical servers included in the system perform data interaction to execute the first aspect or the method for deploying a virtual machine provided by various possible designs of the first aspect. By using the above physical server, the success rate of deploying virtual machines on the physical server can be improved, thereby improving the resource utilization rate of the physical server.

A fourth aspect provides a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when at least one processor of a physical server executes the computer-executable instructions, the physical server executes the first aspect or the first Various possible designs of aspects provide methods of deploying virtual machines. By using the above computer-readable medium, the success rate of deploying a virtual machine on a physical server can be improved, thereby improving the resource utilization rate of the physical server.

In a fifth aspect, a computer program product is provided, the computer program product comprising computer-executable instructions stored in a computer-readable storage medium. At least one processor of the physical server can read the computer-executable instructions from the computer-readable storage medium, and the at least one processor executes the computer-executable instructions to cause the physical server to implement the first aspect or the configuration provided by various possible designs of the first aspect. method for deploying virtual machines. By using the above computer program product, the success rate of deploying a virtual machine on a physical server can be improved, thereby improving the resource utilization rate of the physical server.

Because in this embodiment of the present invention, when a physical server that satisfies all the first binding rules is not found among the physical servers that have idle resources, the target physical server will be determined among the physical servers that have idle resources, and then the target physical server will be determined. The virtual machine is deployed in the target physical server; then, the method of the present invention improves the success rate of deploying the virtual machine in the physical server, thereby improving the resource utilization rate of the physical server.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained according to these drawings.

1 is a schematic diagram of a cloud computing system provided by an embodiment of the present invention;

2 is another schematic diagram of a cloud computing system provided by an embodiment of the present invention;

3 is another schematic diagram of a cloud computing system provided by an embodiment of the present invention;

4 is a schematic flowchart of a method for deploying a virtual machine according to an embodiment of the present invention;

5 is another schematic flowchart of a method for deploying a virtual machine according to an embodiment of the present invention;

6 is another schematic diagram of a cloud computing system provided by an embodiment of the present invention;

FIG. 7 is a schematic structural diagram of a scheduling system provided by an embodiment of the present invention;

FIG. 8 is a physical server provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments.

The present invention first provides a method for deploying virtual machines, and the application scenarios of the method are as follows:

Resource virtualization is performed on physical resources such as physical servers and physical networks, and cloud computing technology is used to build a cloud computing system. As the management center of the cloud computing system, the cloud computing platform can deploy virtual machines (Virtual Machines) on physical servers. VM for short), deploying virtual networks, deploying virtual switches, etc.

As shown in Figure 1, a cloud computing system is mainly composed of a cloud computing platform and a client; the cloud computing platform is used to uniformly provide resources after physical server virtualization, and can allocate resources to clients according to the needs of clients.

As shown in Figure 2, a scheduling system is deployed on the cloud computing platform; the specific implementation of deploying virtual machines on the cloud computing platform is: the scheduling system deploys virtual machines in physical servers. For example, the scheduling system may receive an instruction from the client, and deploy a corresponding virtual machine for the client to use according to the instruction from the client.

Usually, the scheduling system needs to deploy the virtual machine on the physical server during the process of creating, migrating, and HA (High Availability) of the virtual machine. The process of creating a virtual machine refers to the process of deploying a newly created virtual machine to a physical server; the process of migrating a virtual machine refers to the process of re-planning a physical server, which needs to be stored in the physical server. In the process of migrating the virtual machine to other physical servers in the cloud computing system resource pool, during the virtual machine migration process, the virtual machine to be migrated can run normally, and the client cannot know the migration process; the HA process of the virtual machine refers to Due to the failure of a physical server in the cloud computing system, all virtual machines on the physical server are faulty. The process of selecting other physical servers in the resource pool of the cloud computing system and deploying the failed virtual machines. It can be seen that in the above creation process, the newly created virtual machine needs to be deployed on the physical server, and in the above migration and HA process, the to-be-migrated and faulty virtual machines need to be redeployed in the physical server. In the prior art, the deployment process of the above-mentioned newly-created virtual machine, the virtual machine to be migrated, and the faulty virtual machine are the same; The virtual machines are collectively referred to as the virtual machines to be deployed.

In the prior art, in order to meet service performance or reliability requirements, a constraint rule is generally set for each virtual machine, and the constraint rule can constrain the deployment of the virtual machine. At present, the scheduling system of the cloud computing platform generally adopts the following method to arrange the virtual machines: first, determine the physical servers with idle resources in the cloud computing system, and then obtain the preset binding rules for the virtual machines to be deployed, and according to the constraints According to the binding rules, find the physical servers that satisfy the binding rules from the physical servers with idle resources; finally, deploy the virtual machine to be deployed on the physical servers found; If the specified physical server is used, the deployment of the virtual machine to be deployed fails, and the virtual machine is no longer deployed. With the above deployment method, it is possible that no physical server that satisfies the binding rule is found, but there are still physical servers with idle resources, resulting in a waste of resources. For example, as shown in Figure 3, a cloud computing system provides 3 physical servers, and each physical server can deploy up to 4 virtual machines, and virtual machine 1 has been deployed in physical server 1, and there are still three idle Resources; virtual machine 2 has been deployed in physical server 2, and there are still three idle resources left, while physical server 3 has deployed virtual machine 3, virtual machine 4, virtual machine 5, and virtual machine 6, and there are no idle resources. ; At this time, if the physical server 1 fails, the virtual machine 1 also fails, and the scheduling system starts the above HA process to ensure the normal operation of the virtual machine 1, and re-deploys the virtual machine 1 in the physical server; The preset binding rule of machine 1 is that it cannot be deployed on the same physical server as virtual machine 2, and the scheduling system uses the above deployment method to query the physical server with idle resources as physical server 2. The deployment of virtual machine 2 does not satisfy the binding rules of virtual machine 1. Therefore, virtual machine 1 cannot be deployed on physical server 2 at this time, and the HA process of virtual machine 1 fails.

Example 1

Based on the above application scenario, the present invention provides a method for deploying a virtual machine in a physical server. In the method, a first binding rule is set for the virtual machine to be deployed, and the first binding rule can be specifically used as The scheduling system selects the first basis for deploying physical servers to be deployed virtual machines, and the number of the first binding rules may be one or more; here, for the convenience of distinguishing different binding rules, binding The rules are divided into first binding rules and the following second binding rules, and the first binding rules can be specifically binding rules that the physical server does not have to meet. For the specific content of the first binding rules, users can Set it up yourself. As shown in Figure 4, the method is as follows:

Step S41: The scheduling system determines, among the multiple physical servers in the cloud computing system, a physical server with idle resources.

Specifically, the resource requirement of the virtual machine to be deployed is determined, and a physical server whose idle resources of the physical server meet the requirement is selected. The resource, or the idle resource, may specifically include any one or any combination of software and hardware resources such as computing resources, storage resources, and network resources.

For example, before deploying virtual machines to physical servers in a cloud computing system, the central processing unit (CPU, Central Processing Unit) of each physical server is virtualized into multiple vCPUs (virtual Central Processing Unit, virtualized central processing unit). For example, the CPU of a physical server can be virtualized into 90 vCPUs; in the embodiment of the present invention, for any physical server in the cloud computing system, it can be first determined whether the remaining vCPU resources in the physical server can meet the current to-be-deployed The vCPU requirements of the virtual machine to be deployed are determined. If the vCPU resources of the physical server can be satisfied, it is determined that the vCPU resources of the physical server meet the requirements. Otherwise, it is determined that the physical server does not meet the requirements. For example, if the current virtual machine to be deployed needs to occupy 8G of memory, you need to check whether the remaining memory of the physical server is greater than 8G. If it is greater than 8G, determine that the current physical server memory meets the requirements; When the remaining vCPU resources and remaining memory resources meet the requirements of the virtual machine, it is determined that the physical server has idle resources.

Step S42: The scheduling system judges whether there are physical servers that satisfy all the first binding rules among the physical servers with idle resources; if so, go to step S43; otherwise, go to step S44.

In this embodiment of the present invention, the first binding rule may be any one of the following rules or a combination of any of the following rules: a virtual machine vCPU core binding rule, a virtual machine NUMA (Non UniformMemory AccessAchitecture, non-uniform access architecture) memory model) rules, virtual machine group mutual exclusion rules, or association rules between virtual machines and physical servers, etc. The CPU of the physical server in the cloud computing system may be multi-core, and the multi-core means that the CPU has multiple processing cores, and each core processes a task independently at any time; while the vCPU generally needs to be bound to the CPU of the physical server. It runs on a core, and once a vCPU is bound to a certain core of the CPU, it has exclusive rights to the core; the binding rules for virtual machine vCPU cores are set to ensure that there are virtual machines available for vCPU binding on the physical server where the virtual machine is deployed. In the physical server of the cloud computing system, the operating resources of the central processing unit (CPU) are generally divided into multiple blocks, and the memory resources are also divided into multiple blocks, and the operating resources are paired with the memory resources. For example, the operating resources are divided into Running resource A1, running resource B1 and running resource C1; divide the memory resources into memory resource A1, memory resource B1 and memory resource C1, then set running resource A1 and memory resource A1 as a pair, and set running resource B1 and The memory resource B1 is set as a pair, and the running resource C1 and the memory resource C1 are set as a pair; and when the virtual machine is running, it needs to occupy one or more pairs of resources in the physical server, and the NUMA rule of the virtual machine is set to ensure the deployment. In the physical server of the virtual machine, there are paired resources that satisfy the operation of the virtual machine; the virtual machine group mutual exclusion rule refers to the rule that two virtual machines cannot be deployed in the same physical server, which is mainly to ensure that the virtual machine runs. reliability. For example, the group mutual exclusion rule of virtual machine 1 and virtual machine 2 cannot run on the same physical server, because virtual machine 1 and virtual machine 2 provide the same functions; during the creation process of virtual machine, virtual machine The virtual machine 1 and the virtual machine 2 are deployed on the same physical server, so that a situation in which both the virtual machine 1 and the virtual machine 2 fail due to the failure of a physical server can be avoided, thereby improving the reliability of the virtual machine operation. The association rules between virtual machines and physical servers are mainly to constrain the rules of physical servers that can be deployed by virtual machines. For example, the association rules between virtual machines and physical servers are physical server 1, physical server 2, and physical server 3. A machine can only be deployed in Physical Server 1, Physical Server 2, or Physical Server 3.

Step S43: The scheduling system deploys the virtual machines to be deployed in physical servers that satisfy all the first binding rules.

Step S44: The scheduling system determines a target physical server in a physical server with idle resources, and the target physical server satisfies some of the first binding rules, or does not meet any of the first binding rules.

Step S45: The scheduling system deploys the virtual machine to be deployed in the target physical server.

Because in this embodiment of the present invention, when a physical server that satisfies all the first binding rules is not found among the physical servers that have idle resources, the target physical server will be determined among the physical servers that have idle resources, and then the target physical server will be determined. The virtual machine is deployed in the target physical server; then, the method of the present invention improves the success rate of deploying the virtual machine on the physical server, thereby improving the resource utilization rate of the physical server.

Embodiment 2

In this embodiment of the present invention, only one first binding rule may be preset for the virtual machine to be deployed. Then, the step S44 in the above-mentioned embodiment of "determining a target physical server in a physical server with idle resources" may be specifically: : The scheduling system selects any physical server as the target physical server among the physical servers with idle resources.

For example, assuming that the above-mentioned only one first binding rule preset for the virtual machine to be deployed is the first binding rule A, in practical applications, the entire processing flow of the scheduling system can be specifically: firstly, in the cloud computing system All the physical servers in , search for physical servers with idle resources; then, in physical servers with idle resources, search for physical servers that satisfy the above-mentioned first restrictive rule A; if a physical server that satisfies the first restrictive rule A can be found , then deploy the virtual machine to be deployed in the physical server; if no physical server that satisfies the first binding rule A is found, then select any physical server with idle resources, and then The deployment virtual machine is deployed on this physical server.

It can be seen from the above that the above method improves the success rate of deploying virtual machines on the physical server, thereby improving the resource utilization rate of the physical server.

Embodiment 3

In this embodiment of the present invention, M first binding rules can be preset for the virtual machine to be deployed, which are respectively the first binding rule 1, the first binding rule 2, and so on, until the first binding rule M , M is an integer greater than or equal to 2; the above-mentioned M first binding rules may be sorted by priority, that is, the priority of the first binding rule 1 is higher than that of the second binding rule 2, the first binding rule The priority of a binding rule 2 is higher than that of the first binding rule 3, and so on, the priority of the first binding rule I-1 is higher than that of the first binding rule I, and the priority of the different binding rules above The level can be specifically set by users according to their own needs.

Specifically, the step S44 in the above-mentioned embodiment of "determining the target physical server among the physical servers with idle resources" may be specifically as follows: the scheduling system may, among the physical servers with idle resources, determine the target physical server according to the M first The priority of the restrictive rules, one by one matching the first restrictive rules to filter physical servers; if the first restrictive rule matches the first physical server, and matches the first restriction If the first restrictive rule of the sub-priority of the sexual rule does not select a physical server from the first physical server, the scheduling system selects any physical server from the first physical server as the target physical server.

More specifically, with respect to the first binding rule 1, the step S44 in the above-mentioned embodiment "determine the target physical server in the physical server with idle resources" may be specifically: the scheduling system is in the physical server with idle resources. , find a physical server that satisfies the first binding rule 1; when the scheduling system does not find a physical server that satisfies the first binding rule 1, among the physical servers with idle resources, select any physical server as the the target physical server;

For the first binding rule I, the I is a positive integer greater than or equal to 2 and less than or equal to M, the step S44 in the above embodiment "determine the target physical server in the physical server with idle resources" may be specifically: the scheduling The system searches for a physical server that satisfies the first restrictive rule I among the physical servers that satisfy the first restrictive rule I-1; when the scheduling system does not find a physical server that meets the first restrictive rule I, it will Any one of the physical servers in the first binding rule I-1 is used as the target physical server.

Specifically, for the to-be-deployed virtual machine with M first binding rules, the entire processing flow of the scheduling system in the cloud computing platform is as follows: First, from multiple physical servers in the cloud computing system, determine the physical server that has idle resources ; Then, from the physical servers that have idle resources, find the physical server that satisfies the first restrictive rule 1; , find a physical server that satisfies the first binding rule 2; if no physical server that satisfies the first binding rule 1 is found, then arbitrarily select a physical server with idle resources, and then deploy the virtual machine to be deployed in the selected In the physical server; if the physical server that satisfies the first binding rule 1 can be found, then in the physical server that satisfies the first binding rule 1, continue to search for the physical server that satisfies the first binding rule 2, if you can find the physical server that satisfies the first binding rule 2 If a physical server that satisfies the first binding rule 2 is found, continue to search for a physical server that satisfies the first binding rule 3 among the physical servers that meet the first binding rule 2, and so on, until the first binding rule is found. M; of course, at this time, if the above method can be used to find out the physical servers that satisfy the first binding rules 1 to M, it means that there are physical servers that satisfy all the first binding rules, and then the virtual machine to be deployed will be It can be deployed on physical servers that satisfy all the first binding rules.

Assume that three first binding rules are preset for a virtual machine to be deployed, namely first binding rule 1, first binding rule 2 and first binding rule 3, as shown in FIG. 5, the entire scheduling system The processing flow is as follows:

Step S51: from a plurality of physical servers of the cloud computing system, search for physical servers with idle resources;

Step S52: determine whether a physical server that satisfies the first binding rule 1 can be found among the physical servers with idle resources; if it can be found, go to step S53, if not, go to step S56;

Step S53: Determine whether a physical server that satisfies the first binding rule 2 can be found among the physical servers that satisfy the first binding rule 1; if it can be found, go to step S54; otherwise, go to step S57;

Step S54: determine whether a physical server that satisfies the first binding rule 2 can be found among the physical servers that satisfy the first binding rule 2; if it can be found, go to step S55; otherwise, go to step S58;

Step S55: in the physical servers that satisfy the first binding rule 3, select a physical server, and deploy the virtual machine to be deployed in the selected physical server;

Step S56: in the physical servers with idle resources, select a physical server, and deploy the virtual machine to be deployed in the selected physical server;

Step S57: Among the physical servers that satisfy the first binding rule 1, a physical server is arbitrarily selected, and the virtual machine to be deployed is deployed in the physical server;

Step S58: Among the physical servers that satisfy the first binding rule 2, a physical server is arbitrarily selected, and the virtual machine to be deployed is deployed in the physical server.

It can be seen from the foregoing that the first binding rule is a binding rule that does not have to be satisfied. When no physical server that satisfies all the first binding rules is found, the above method can be used to find the physical server that satisfies the most first binding rules. server, so as to further optimize the deployment of virtual machines on the basis of saving resources.

Embodiment 4

In this embodiment of the present invention, in addition to setting a first restrictive rule for the virtual machine to be deployed, a second restrictive rule may also be set for the virtual machine to be deployed, and the second restrictive rule is used for selecting the scheduling system. The second basis for deploying the physical server of the virtual machine to be deployed, the second binding rule is a rule that must be satisfied by the physical server when deploying the virtual machine; the number of the second binding rule can be one or one For multiple, users can set by themselves. In this embodiment of the present invention, the second binding rule may also be any one of the following rules or a combination of any of the following rules: virtual machine vCPU core binding rule, virtual machine NUMA rule, virtual machine group mutual exclusion A rule or an association rule between a virtual machine and a physical server; and whether the above rule is set as the first binding rule or the second binding rule, users can set it according to their own needs.

The methods described in the first embodiment, the second embodiment and the third embodiment may further include: the scheduling system searches for a physical server that satisfies the second binding rule among the physical servers with idle resources; From the physical servers that satisfy the second binding rule, search for physical servers that satisfy all the first binding rules.

Specifically, for the to-be-deployed virtual machine set with the first binding rule and the second binding rule, the scheduling system may deploy the to-be-deployed virtual machine in the following manner: In the physical server, search for a physical server with idle resources; then, in a physical server with idle resources, search for a physical server that satisfies the second binding rule; if no physical server that satisfies the second binding rule is found, it is to be deployed The deployment of the virtual machine fails, and the deployment of the virtual machine to be deployed is stopped; when there are multiple second binding rules, because the second binding rules are binding rules that must be satisfied by the physical server, find out The physical server needs to meet all the second binding rules. When the physical server does not meet any second binding rule, the deployment of the virtual machine to be deployed fails; if a physical server that meets the second binding rule is found, then On the physical server that satisfies the second binding rule, continue to search for the physical server that satisfies the first binding rule; when there is one first binding When there are more than one, please refer to the description of the third embodiment.

It can be seen from the above that in this embodiment of the present invention, when deploying a virtual machine, the user can also set the second binding rule that the physical server must satisfy according to requirements, thereby making the setting of the binding rule for the virtual machine more flexible.

Embodiment 5

The present invention also provides a second method for deploying virtual machines, the details are as follows, still referring to FIG. 6 , the failure of server 1 causes virtual machine 1 in server 1 to also fail. At this time, the above-mentioned HA mechanism is activated, and in the resource pool Select another physical server, and redeploy the virtual machine 1; and the first binding rule preset by the virtual machine. Please refer to Table 1 for details.

Table 1 First Binding Rules for Virtual Machine 1

In the embodiment of the present invention, the scheduling system will first determine the physical servers with idle resources. From FIG. 6 , it can be concluded that there are idle resources in both the physical server 2 and the physical server 3; Binding rules, according to the priority, search in physical server 2 and physical server 3 in turn, and meet the first priority "virtual machine vCPU core binding rule" and the second priority "virtual machine NUMA rule" Physical server, Assume that only physical server 2 satisfies the above two rules; at this time, the scheduling system will continue to check whether physical server 2 satisfies the "virtual group mutual exclusion rule" of the third priority, but physical server 2 has already deployed virtual machine 2 , does not meet the "Virtual Group Mutual Exclusion Rule" of the third priority, but at this time physical server 2 is the physical server that satisfies the most high-priority first binding rules, therefore, virtual machine 1 is deployed in physical server 2 , thus protecting the reliability of the business.

Embodiment 6

Corresponding to the above method, the present invention also provides a scheduling system for deploying virtual machines, the cloud computing platform is deployed on a plurality of physical servers, and the cloud computing platform includes a scheduling system; at least one first binding rule is selected as the The first basis for deploying the physical server of the virtual machine to be deployed, as shown in FIG. 7 , the scheduling system 700 includes:

An idle resource determining unit 71, configured to determine a physical server with idle resources among the plurality of physical servers;

The target physical server determining unit 72 is configured to determine a target physical server when no physical server that satisfies all the first constraint rules is found in the physical servers with idle resources, and the target physical server satisfies the at least one first constraint part of the first binding rules in the binding rules, or the target physical server does not satisfy any one of the at least one first binding rules;

The first deployment unit 73 is configured to deploy the to-be-deployed virtual machine in the target physical server.

Optionally, when the number of the first constraint rules is one, the target physical server determination unit 72 is specifically configured to: find no physical server that satisfies the first constraint in the physical servers with idle resources. When selecting the physical server of the rule, select any physical server as the target physical server among the physical servers with idle resources.

Optionally, the number of the first binding rules is M, the M is a positive integer greater than or equal to 2, and the M first binding rules have priorities; the target physical server determining unit 72, It can be specifically used for: in a physical server with idle resources, according to the priority of the M first binding rules, one by one matching the first binding rules to filter physical servers; A binding rule is screened to the first physical server, and the first binding rule matching the sub-priority of the first one of the first binding rules is not screened to a physical server from the first physical server, then Any physical server is selected from the first physical servers as the target physical server.

Optionally, the scheduling system 700 further includes: a first search unit, configured to search for physical servers that satisfy all the first restrictive rules in the at least one first restrictive rule among physical servers with idle resources; a second The deployment unit is configured to deploy the virtual machine to be deployed in the physical server that satisfies all the first binding rules in the at least one first binding rule in the physical server.

Optionally, the scheduling system 700 may further include: a second search unit, configured to search for a physical server that satisfies a second binding rule among physical servers with idle resources, where the second binding rule is selected for deployment a second basis for the physical server on which the virtual machine is to be deployed; and a third search unit, configured to search for a physical server that satisfies the first binding rule among the physical servers that meet the second binding rule.

Because in this embodiment of the present invention, when a physical server that satisfies all the first binding rules is not found among the physical servers that have idle resources, the target physical server will be determined among the physical servers that have idle resources, and then the target physical server will be determined. The deployed virtual machine is deployed in the target physical server; then, using the scheduling system of the present invention improves the success rate of deploying the virtual machine on the physical server, thereby improving the resource utilization rate of the physical server.

Embodiment 7

The cloud computing platform is deployed on multiple physical servers, so the scheduling system can be deployed on at least one physical server. Therefore, the hardware subject for implementing the method for deploying virtual machines is the physical server, and the hardware subject implementing the above scheduling system is the physical server. Also a physical server. Therefore, corresponding to the above embodiments, the present invention also discloses a physical server, the physical server is shown in FIG. 8 .

The physical server 800 includes at least one processor 801 , a memory 802 and a communication interface 803 ; the at least one processor 801 , the memory 802 and the communication interface 803 are all connected through a bus 804 .

The memory 802 is used to store computer-executed instructions.

The at least one processor 801 is configured to execute computer-executed instructions stored in the memory 802, so that the physical server 800 executes the method for deploying a virtual machine provided by the foregoing method embodiments, or enables the physical server 800 to implement scheduling some or all of the functionality of the system; or,

The at least one processor 801 is configured to execute the computer-executed instructions stored in the memory 802, so that the physical server 800 performs data interaction with other physical servers included in the cloud computing system through the communication interface 803 to execute the above method implementation. The method for deploying a virtual machine provided in the example, or enabling the physical server 800 to perform data interaction with other physical servers included in the cloud computing system through the communication interface 803 to implement part or all of the functions of the scheduling system.

At least one processor 801 may include processors 801 of different types, or include processors 801 of the same type; the processor 801 may be any of the following: a central processing unit (Central Processing Unit, CPU for short), an ARM processor , Field Programmable Gate Array (Field Programmable Gate Array, FPGA for short), dedicated processor and other devices with computing processing capabilities. In an optional implementation manner, the at least one processor 801 may also be integrated into a many-core processor.

The memory 802 may be any one or any combination of the following: random access memory (Random Access Memory, RAM for short), read only memory (ROM for short), non-volatile memory (non-volatile memory, NVM for short), Solid State Drives (SSD for short), mechanical hard disks, magnetic disks, disk arrays and other storage media.

The communication interface 803 is used for data interaction between the physical server 800 and other devices (eg, other physical servers in the virtual machine system). The communication interface 803 may be any one or any combination of the following: a network interface (eg, an Ethernet interface), a wireless network card, and other devices with a network access function.

The bus 804 may include an address bus, a data bus, a control bus, and the like. For ease of illustration, FIG. 8 represents the bus with a thick line. The bus 804 may be any one or any combination of the following: an Industry Standard Architecture (ISA for short) bus, a Peripheral Component Interconnect (PCI for short) bus, an Extended Industry Standard Architecture (Extended Industry) Standard Architecture, referred to as EISA) bus and other wired data transmission devices.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows 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 the 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 Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be construed to include the preferred embodiment and all changes and modifications that fall within the scope of the present invention.

Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, provided that these modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (9)

1. A method for deploying a virtual machine, wherein at least one first binding rule is used as a first basis for selecting a physical server for deploying a virtual machine to be deployed, the method comprising: Among multiple physical servers, determine the physical server with idle resources; When a physical server that satisfies all the first binding rules is not found in the physical servers with idle resources, a target physical server is determined, and the target physical server satisfies a part of the first binding rules among the plurality of first binding rules, or the target physical server does not satisfy any one of the at least one first binding rule; deploying the virtual machine to be deployed in the target physical server; The number of the first restrictive rules is M, the M is a positive integer greater than or equal to 2, and the M first restrictive rules have priorities; the physical servers that have idle resources are not found. Target physical servers are determined when all physical servers that satisfy all first binding rules include: In the physical servers with idle resources, according to the priorities of the M first binding rules, the physical servers are selected by matching the first binding rules one by one; If the first binding rule matching the first one is filtered to the first physical server, and the first binding rule matching the second priority of the first binding rule is not selected from the first physical server If a physical server is selected from the servers, select any physical server from the first physical server as the target physical server, the I is a positive integer greater than or equal to 2, and the I is a positive integer less than or equal to M Integer. 2 . The method according to claim 1 , wherein the number of the first binding rules is one, and no physical server that satisfies all the first binding rules is found in the physical servers with idle resources. 3 . When server, determine the target physical server, including: When no physical server satisfying the first binding rule is found among the physical servers with idle resources, any physical server is selected as the target physical server among the physical servers with idle resources. 3. The method according to claim 1 or 2, wherein the method further comprises: Among physical servers with idle resources, searching for physical servers that satisfy all the first binding rules in the at least one first binding rule; When finding out the physical servers that satisfy all the first restrictive rules in the at least one first restrictive rule, deploy the to-be-deployed virtual machine in the physical servers that satisfy all the first restrictive rules. 4. The method according to claim 1 or 2, wherein, when no physical servers satisfying all the first binding rules are found in the physical servers with idle resources, before determining the target physical server, the Methods also include: Among the physical servers with idle resources, a physical server that satisfies a second binding rule is searched for, and the second binding rule is used as a second method for selecting a physical server from the plurality of physical servers for deploying the virtual machine to be deployed. in accordance with; Among the physical servers that satisfy the second binding rule, a physical server that satisfies the first binding rule is searched. 5. A scheduling system for deploying virtual machines, wherein at least one first binding rule is used as a first basis for selecting a physical server for deploying virtual machines to be deployed, and the scheduling system comprises: an idle resource determination unit, configured to determine a physical server with idle resources among the plurality of physical servers; A target physical server determination unit, configured to determine a target physical server when a physical server that satisfies all the first binding rules is not found in the physical servers with idle resources, and the target physical server satisfies one of the plurality of first binding rules part of the first binding rule, or the target physical server does not satisfy any one of the at least one first binding rule; a first deployment unit, configured to deploy the to-be-deployed virtual machine in the target physical server; The number of the first binding rules is M, the M is a positive integer greater than or equal to 2, and the M first binding rules have priorities; the target physical server determination unit is specifically used for: In the physical servers with idle resources, according to the priorities of the M first binding rules, the physical servers are selected by matching the first binding rules one by one; If the first binding rule matching the first one is filtered to the first physical server, and the first binding rule matching the second priority of the first binding rule is not selected from the first physical server If a physical server is selected from the servers, select any physical server from the first physical server as the target physical server, the I is a positive integer greater than or equal to 2, and the I is a positive integer less than or equal to M Integer. 6. The scheduling system according to claim 5, wherein when the number of the first binding rules is one, the target physical server determination unit is specifically used for: When no physical server satisfying the first binding rule is found among the physical servers with idle resources, any physical server is selected as the target physical server among the physical servers with idle resources. 7. The scheduling system according to claim 5 or 6, wherein the scheduling system further comprises: a first search unit, configured to search for physical servers that satisfy all the first binding rules in the at least one first binding rule among the physical servers with idle resources; The second deployment unit is configured to deploy the virtual machine to be deployed in a physical server that satisfies all the first binding rules in the at least one first binding rule in the physical server. 8. The scheduling system according to claim 5 or 6, wherein the scheduling system further comprises: A second search unit, configured to search for a physical server that satisfies a second binding rule among physical servers with idle resources, the second binding rule being used as a second basis for selecting a physical server for deploying a virtual machine to be deployed ; The third search unit is configured to search for physical servers that satisfy the first restrictive rule among the physical servers that satisfy the second restrictive rule. 9. A physical server, comprising at least one processor, a memory and a communication interface; the at least one processor, the memory and the communication interface are all connected through a bus; the memory for storing computer-executed instructions; The at least one processor is configured to execute computer-executed instructions stored in the memory, so that the physical server executes the method for deploying a virtual machine according to any one of claims 1 to 4, or causes the physical server to pass The communication interface performs data interaction with other physical servers to execute the method for deploying a virtual machine according to any one of claims 1 to 4.

Images