WO2022143714A1

WO2022143714A1 - Server system, and virtual machine creation method and apparatus

Info

Publication number: WO2022143714A1
Application number: PCT/CN2021/142282
Authority: WO
Inventors: 龚磊; 龙鹏; 周谊珉; 黄智超
Original assignee: 华为云计算技术有限公司
Priority date: 2020-12-29
Filing date: 2021-12-29
Publication date: 2022-07-07
Also published as: CN114691286A

Abstract

Disclosed are a server system, and a virtual machine creation method and apparatus, which belong to the technical field of cloud computing. The server system comprises a server and an uninstalling card inserted into the server, wherein a connection channel is established between the server and the uninstalling card; a virtual machine is arranged in the server, the virtual machine is implemented by means of a virtual machine simulator, the virtual machine simulator comprises a virtual machine management module, and the virtual machine management module comprises virtual machine management proxy sub-modules and virtual machine management sub-modules; the virtual machine management proxy sub-modules are deployed on the server, and the virtual machine management sub-modules are deployed on the uninstalling card; and the virtual machine management proxy sub-modules communicate with the virtual machine management sub-modules by means of the connection channel, and the virtual machine management sub-modules are used for implementing a management function of the virtual machine. By means of the present application, the occupation of server resources by the overheads of a virtual machine simulator itself is reduced.

Description

Server system, virtual machine creation method and device

technical field

The present application relates to the field of cloud computing technologies, and in particular, to a server system, a method and apparatus for creating a virtual machine.

Background technique

As the core technology of cloud scenarios, virtualization technology shares the physical resources of a server to multiple users with virtual machines as the granularity, so that users can use the physical resources of the server conveniently and flexibly under the premise of security isolation, which can greatly Improve the utilization of physical resources of the server.

However, in the traditional virtualization technology, when using the physical resources of the server to provide virtual machine services to users through the virtual machine simulator, certain physical resources need to be consumed.

Therefore, it is impossible to provide all physical resources of the server to users. In addition, with the improvement of IO performance, the physical resources consumed by the virtual machine simulator will increase linearly, which will have a great impact on the business performance of the virtual machine provided to the user.

SUMMARY OF THE INVENTION

The present application provides a server system, a method and apparatus for creating a virtual machine, which can reduce the occupation of the physical resources of the server by the virtual machine simulator and ensure the service performance of the virtual machine.

In a first aspect, the present application provides a server system. The server system includes a server and an unloading card inserted in the server. A connection channel is established between the server and the unloading card; a virtual machine is set in the server, and the virtual machine passes through a virtual machine simulator. Implementation, the virtual machine simulator includes a virtual machine management module, and the virtual machine management module includes a virtual machine management agent sub-module and a virtual machine management sub-module; a virtual machine management agent sub-module is deployed on the server, and a virtual machine management sub-module is deployed on the uninstall card. module, wherein the virtual machine management agent sub-module communicates with the virtual machine management sub-module through the connection channel, and the virtual machine management sub-module is used to realize the management function of the virtual machine.

In the server system, by dividing the virtual machine management module into a virtual machine management agent sub-module and a virtual machine management sub-module, and deploying the virtual machine management sub-module on the unloading card, the virtual machine simulator is realized. The functions that consume more server resources are deployed in the uninstall card, which reduces the overhead of the virtual machine emulator itself and the occupation of server resources, and reduces the cost of the virtual machine emulator itself and the virtual machine that provides services to users due to competition for resources. The impact of the virtual machine business performance effectively guarantees the virtual machine business performance. In addition, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the business performance of the virtual machine (such as the scheduling performance of the virtual processor and the I/O read and write performance of the virtual machine) is also It will not produce side effects, so it has a good application prospect.

Optionally, the management function includes: one or any combination of a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function and a parameter parsing function.

As can be seen from the above, the functions provided by the virtual machine management sub-module can be regarded as functions that have little relevance to the virtual machine business and/or involve asynchronous operations. Such functions have no impact on the internal performance of the virtual machine itself, but consume less server resources. larger. Therefore, by deploying the virtual machine management sub-module on the uninstall card, the occupation of server resources by the virtual machine simulator can be effectively reduced, and the service performance of the virtual machine can be effectively guaranteed.

In an implementation manner, the virtual machine simulator further includes a virtual machine hardware simulation module, the virtual machine hardware simulation module is provided in the server, and the virtual machine hardware simulation module is used to simulate the hardware of the virtual machine according to the hardware of the server.

Optionally, the hardware of the virtual machine includes one or more of the following: a virtual processor, a virtual memory, a virtual motherboard chip, a virtual input and output device, and a virtual bus.

The virtual machine hardware emulation module is used to implement a function strongly related to the virtual machine function, and the function basically does not consume server resources. Therefore, even if the virtual machine hardware emulation module is deployed in the server, the virtual machine hardware emulation module will not affect the virtual machine. The probability of affecting the business performance of the virtual machine is also very low, which can effectively ensure the business performance of the virtual machine.

In an implementation manner, the virtual machine management sub-module is further configured to access data of the virtual machine from the memory of the server in a direct memory access DMA manner through the connection channel.

Obtain data from virtual memory through DMA. Since DMA allows direct data transfer between external devices and storage resources, the CPU does not need to be involved in the data transfer process, which can effectively reduce the impact on the server of obtaining data from storage resources. Occupation of resources.

Optionally, the server and the offload card are connected through a virtual high-speed peripheral component interconnect PCIe bus or a computing express link CXL bus.

In a second aspect, the present application provides a method for creating a virtual machine. The method is applied to a server system. The server system includes a server and an offload card inserted in the server. A connection channel is established between the server and the offload card. The method includes: the server receives The virtual machine creation request sent by the cloud management platform; the server creates a virtual machine in the server according to the virtual machine creation request, starts the virtual machine management agent sub-module set in the server, and starts the virtual machine management set in the uninstall card through the connection channel submodule; wherein, the virtual machine is implemented by a virtual machine simulator, the virtual machine simulator includes a virtual machine management module, the virtual machine management module includes a virtual machine management agent submodule and a virtual machine management submodule, and the virtual machine management agent submodule is connected by The channel communicates with the virtual machine management sub-module, and the virtual machine management sub-module is used to implement the management function of the virtual machine.

Optionally, the virtual machine simulator further includes a virtual machine hardware simulation module, the virtual machine hardware simulation module is set in the server, and the virtual machine hardware simulation module is used to simulate the hardware of the virtual machine according to the hardware of the server.

Optionally, the virtual machine management sub-module is further configured to access data of the virtual machine from the memory of the server in a direct memory access DMA manner through the connection channel.

In a third aspect, the present application provides an apparatus for creating a virtual machine, the apparatus is applied to a server, an uninstall card is inserted on the server, a connection channel is established between the server and the uninstall card, and the apparatus includes: a receiving module for receiving cloud management The virtual machine creation request sent by the platform; the processing module is used to create a virtual machine in the server according to the virtual machine creation request, start the virtual machine management agent sub-module set in the server, and start the virtual machine set in the uninstall card through the connection channel. machine management sub-module; wherein, the virtual machine is implemented by a virtual machine simulator, the virtual machine simulator includes a virtual machine management module, the virtual machine management module includes a virtual machine management agent sub-module and a virtual machine management sub-module, and the virtual machine management agent sub-module It communicates with the virtual machine management sub-module through the connection channel, and the virtual machine management sub-module is used to realize the management function of the virtual machine.

In a fourth aspect, a computer device is provided. The computer device includes: a processor and a memory, and a computer program is stored in the memory; when the processor executes the computer program, the computer device implements the method provided in the second aspect.

In a fifth aspect, a storage medium is provided, and when instructions in the storage medium are executed by a processor, the method provided in the second aspect is implemented.

Description of drawings

FIG. 1 is a schematic diagram of an implementation scenario involved in a server system provided by an embodiment of the present application;

2 is a schematic diagram of the deployment of a server system provided by an embodiment of the present application;

3 is a schematic structural diagram of a server system provided by an embodiment of the present application;

4 is a schematic structural diagram of another server system provided by an embodiment of the present application;

5 is a schematic diagram of an implementation process of a virtual machine management agent sub-module and a virtual machine management sub-module collaboratively implementing a VNC function according to an embodiment of the present application;

6 is a schematic diagram of an implementation process of performing input on a VNC client through a mouse and a keyboard to realize remote control of a virtual machine provided by an embodiment of the present application;

7 is a flowchart of a method for creating a virtual machine provided by an embodiment of the present application;

8 is a schematic structural diagram of an apparatus for creating a virtual machine provided by an embodiment of the present application;

FIG. 9 is a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

For ease of understanding, the terms involved in the embodiments of the present application are first explained below.

Cloud computing: Cloud computing is a type of distributed computing, which refers to a network that uniformly manages and schedules a large number of computing and storage resources and provides on-demand services to users. Both the computing resources and the storage resources are provided by a cluster of computing devices arranged in the data center. Moreover, cloud computing can provide users with various types of services, such as infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (software as a service). a service, SaaS), etc.

Virtualization is a resource management technology, which abstracts and transforms various physical resources of the host, such as servers, network resources, memory and storage resources, to break the inseparability between the physical structures of the host. barriers, allowing users to apply these resources in a better way than the original configuration. The resources used through virtualization are called virtualized resources, and the virtualized resources are not limited by the way of setting up existing resources, the geographical area or the physical configuration. Generally, virtualized resources include computing resources and storage resources.

Virtual machine (virtual machine, VM): refers to a complete computer system with complete hardware system functions simulated by virtualization technology and running in a completely isolated environment. Part of the instruction subset of the virtual machine can be processed in the host machine, and other part of the instruction can be executed in an emulated manner. Users can purchase cloud services by renting virtual machines.

Host (also called physical machine): A physical resource (such as a virtual machine or physical machine) used to host virtualization technology. Typically, a host for deploying a virtual machine is a physical server.

Direct memory access (DMA): It can also be called direct memory operation or group data transfer, which means that external devices do not go through the central processing unit (CPU) of the computer, but directly access the memory from the computer's memory. Data interaction mode for fetching data. When transferring data in DMA mode, the CPU of the computer issues an instruction to the DMA controller to instruct the DMA controller to control the data transfer, and the DMA controller feeds back the transfer completion information to the CPU after completing the data transfer. It can be seen that in the process of transferring data by DMA, there is no need for the computer's CPU to perform the transfer operation, which can save the computer's CPU operations such as fetching instructions, fetching numbers, and sending numbers, reducing the resource occupancy rate of the computer's CPU and saving system resource.

DMA can include remote DMA (remote direct memory access, RDMA) and local DMA. RDMA refers to a data transmission method that transfers data directly from the memory of one computer to another computer through the network without the intervention of the operating systems of both parties. Local DMA refers to the transfer of data without going through the network.

Peripheral Component Interconnect Express (PCIe) bus: A high-speed serial computer expansion bus.

Compute Express Link (CXL) bus: A fast connection bus for high-speed computers.

Network card: Also known as a network interface controller (NIC), network adapter, or LAN receiver, is a piece of computer hardware designed to allow a host or computing device to communicate over a network.

Memory: It can also be called internal memory or main memory. Its function is to temporarily store the operation data in the CPU and the data exchanged with external memory such as hard disks.

The embodiment of the present application provides a server system. The server system includes a server and an unloading card inserted in the server. A connection channel is established between the server and the uninstall card. A virtual machine is set in the server, and the virtual machine is realized by a virtual machine simulator. The virtual machine simulator includes a virtual machine management module, and the virtual machine management module includes a virtual machine management agent submodule and a virtual machine management submodule. A virtual machine management agent submodule is deployed on the server, and a virtual machine management submodule is deployed on the uninstall card. The virtual machine management agent sub-module communicates with the virtual machine management sub-module through the connection channel, and the virtual machine management sub-module is used to implement the management function of the virtual machine.

Since the virtual machine management sub-module is used to implement the management function of the virtual machine, the virtual machine management sub-module consumes a lot of physical resources. By splitting the virtual machine management module into a virtual machine management agent sub-module and a virtual machine management sub-module , and deploy the virtual machine management sub-module on the uninstall card, so that the functions that consume more server resources in the virtual machine emulator are deployed on the uninstall card, reducing the occupancy of server resources by the overhead of the virtual machine emulator itself. , reducing the cost of the virtual machine simulator itself and the impact of the virtual machine providing services to users on the business performance of the virtual machine due to competing resources, and effectively ensuring the business performance of the virtual machine.

FIG. 1 is a schematic diagram of an implementation scenario involved in a server system provided by an embodiment of the present application. As shown in FIG. 1 , the implementation scenario includes: a server system 1 and a management platform 2 . The management platform 2 is used to interact with the user and send the user's request to the server system 1 . The server system 1 is used to respond to the user's request. For example, the management platform 2 is used to send a request for creating a virtual machine or a request for managing a virtual machine to the server system 1, and the server system 1 is used to create a virtual machine according to the request for creating a virtual machine, or the server 1 is used to create a virtual machine according to the virtual machine creation request. The request for management of the virtual machine is performed to perform operations related to the management of the virtual machine. Among them, the virtual machine is used to implement the user's business and provide services for the user, for example, the virtual machine is used to provide the user with a language recognition service, a video review service, or an image rendering service.

In an implementation manner, the server system 1 may be deployed in a cloud platform P of a cloud data center, and the management platform may be a cloud management platform 2 deployed in the cloud platform P. The cloud platform P may be a cloud platform P of a central cloud, a cloud platform P of an edge cloud, or a cloud platform P including a central cloud and an edge cloud, which is not specifically limited in this embodiment of the present application. Wherein, FIG. 1 is a schematic diagram of the server system 1 being deployed in the cloud platform P. As shown in FIG.

In an implementation manner, the cloud platform P may be implemented by hardware such as the physical server 11 . For example, the cloud platform P can be implemented by one physical server 11, or a physical server 11 cluster composed of several physical servers 11, or a cloud computing service center. Moreover, the cloud platform P can be implemented by independently deployed resources, or can be implemented by distributed deployed resources.

A large number of basic resources owned by the cloud service provider are deployed in the cloud platform P, and the cloud platform P can use the basic resources to provide cloud services to users. The basic resources may include computing resources, storage resources, network resources, etc., and the computing resources may be a large number of computing devices (eg, the server system 1 ). As shown in FIG. 1 , after the user purchases the cloud service, the cloud platform P can create a virtual machine for the user by using basic resources such as the server system 1 to provide the user with the cloud service by using the created virtual machine.

The cloud management platform 2 and the server system 1 can be implemented according to the basic resources in the cloud platform P. Optionally, the cloud management platform 2 may be implemented by software such as virtual machines or containers on basic resources, or the cloud management platform 2 may be implemented by a physical server 11 (also called a host) such as a bare metal server 11 . The server system 1 includes a server 11 and an uninstall card 12 , and the server 11 may be implemented by a physical server 11 (also called a host) such as a bare metal server 11 . The offload card 12 can be implemented by heterogeneous hardware, and the heterogeneous hardware can be used as an external device of the server 11 . The uninstall card 12 is also called a smart card, and the uninstall card 12 can be regarded as a small server.

As shown in FIG. 2 , a virtual machine monitor (VMM) 111 and a virtual machine simulator 112 run on the server 11 . The VMM is used to provide hardware resources such as CPU and memory in the server 11 for use by virtual machines (VM1 to VMn in FIG. 2 ) through virtualization technology. In addition, the VMM is also used to directly pass the input and output devices provided by the offload card 12 to the virtual machine. By directly passing the input and output devices of the offload card 12 to the virtual machine, the virtual machine can directly access the input and output devices of the offload card 12 bypassing the VMM, so that the virtual machine can obtain performance close to that of a physical machine. The virtual machine simulator 112 is used to virtualize a virtual machine based on the hardware function of the virtual machine provided by the virtual machine manager 111 .

Optionally, as shown in FIG. 2 , the offload card 12 can be connected with storage resources and/or network resources, so that the offload card 12 can send the I/O request of the virtual machine running in the server 11 to the storage resource and/or network resources for processing.

The implementation manner of the server system 1 provided by the embodiments of the present application will be described below. As shown in FIG. 3 , the server system 1 includes a server 11 and an uninstall card 12 , and a connection channel is established between the server 11 and the uninstall card 12 .

The server 11 is provided with a virtual machine VM (such as VM1, . The management module 1121 includes a virtual machine management agent sub-module 1121a and a virtual machine management sub-module 1121b.

A virtual machine management agent sub-module 1121a is deployed on the server 11, and a virtual machine management sub-module 1121b is deployed on the uninstall card 12, wherein the virtual machine management agent sub-module 1121a communicates with the virtual machine management sub-module 1121b through a connection channel, and the virtual machine The management submodule 1121b is used to implement the management function of the virtual machine.

In an implementation manner, the virtual machine management agent sub-module 1121a is specifically configured to receive a management request requesting to manage the virtual machine, and send the management request to the virtual machine management sub-module 1121b through the connection channel, and the virtual machine management sub-module 1121b is specifically configured to perform operations related to the management of the virtual machine according to the management request, so as to realize the management of the virtual machine.

Since the virtual machine management sub-module 1121b is used to implement the management function of virtual machines, the virtual machine management sub-module 1121b consumes a lot of physical resources. The virtual machine management sub-module 1121b is deployed on the offload card 12, so that the function of the virtual machine emulator 112 that consumes more server resources can be deployed on the offload card 12, which reduces the number of virtual machine simulations. The overhead of the server 112 itself occupies server resources, which reduces the overhead of the virtual machine emulator 112 itself and the impact of the virtual machine providing services to users on the business performance of the virtual machine due to competition for resources, effectively ensuring the business performance of the virtual machine. .

In this embodiment of the present application, the server 11 may be implemented by a physical server 11 such as a bare metal server 11 . The offload card 12 can be implemented by heterogeneous hardware, and the heterogeneous hardware can be used as an external device of the server 11 . The connection channel between the offload card 12 and the server 11 can be realized through a connection bus. For example, a PCIe bus, a CXL bus, a virtual peripheral component interconnect (PCI) bus or a virtual industry standard architecture (ISA) bus is connected between the offload card 12 and the server 11. At this time, when the server 11 processes the input/output (I/O) service that needs to be offloaded to the offload card 12 for processing, it can output the relevant data to the offload card 12 through the bus between the offload card 12 and the server 11 deal with. In addition, the offload card 12 is an offload card 12 suitable for connecting a bus between the offload card 12 and the server 11. For example, when a PCIe bus is connected between the offload card 12 and the server 11, the offload card 12 is a PCIe card, which is implemented in this application. For example, the model is not specifically limited.

The implementation of the server system 1 will be described below by taking the connection between the offload card 12 and the server 11 via the PCIe bus as an example. FIG. 4 is a schematic structural diagram of a server system 1 provided by an embodiment of the present application. As shown in FIG. 4 , in the system architecture, the server system 1 includes a server 11 and an uninstall card 12 . The server 11 includes a CPU1, a memory 1 and a root complex (RC) chip. The offload card 12 includes a CPU 2, a memory 2 and an endpoint (end point, EP) chip, and the EP chip is connected to the RC chip through a PCIe bus. Among them, the RC chip is used to provide the CPU1 with a hardware interface for sending and receiving PCIe messages with external devices. The EP chip is used to provide a hardware interface for the peripheral interface to send and receive PCIe messages with the server 11 .

The embodiments of the present application do not limit the specific implementation of the RC chip and the EP chip, and any RC chip and EP chip implemented in accordance with the PCIe specification can be used. For example, the EP chip may be an application-specific integrated circuit (ASIC) chip, a smart network card, or a field-programmable gate array (field-programmable gate array, FPGA) chip, or the like. Moreover, the EP chip can be directly or indirectly connected to the PCIe bus, so that the server 11 and the EP can be connected through the PCIe bus. For example, the EP can be indirectly connected to the PCIe bus through a switch, and the switch can provide more ports for connecting the EP chips to expand the link.

Optionally, the management function of the virtual machine includes: one or any combination of a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function and a parameter parsing function. Correspondingly, as shown in FIG. 4, the virtual machine management sub-module 1121b includes: one of a virtual machine life cycle management unit b1, a migration unit b2, a virtual remote control unit b3, an asynchronous information processing unit b4 and a parameter parsing unit b5 or random combination. The functions of each unit are as follows:

The virtual machine life cycle management unit b1 is used to implement the virtual machine life cycle management function, that is, to manage the running state of the virtual machine. For example, the virtual machine life cycle management unit b1 is used to control the start and stop of the virtual machine.

The migration unit b2 is configured to implement a virtual machine migration function, that is, acquire data to be migrated of the virtual machine from the virtual memory 1122b, and perform a migration operation on the virtual machine based on the data to be migrated. In an achievable manner, the migration unit b2 may be a hot migration unit b2, which is used to perform accelerated processing of the migration logic, including acquiring the data to be migrated of the virtual machine in the virtual memory 1122b by DMA, and then sending the data to be migrated. to other servers 11 to which the virtual machine needs to be migrated. The hot migration module may send the data to be migrated to other servers 11 through the connection channel between the servers 11 . For example, a communication channel such as a socket channel between servers 11 is used to send the data to be migrated to other servers 11 .

The virtual remote control unit b3 is used to realize the virtual remote control function, that is, obtain the video memory data from the virtual memory 1122b of the virtual machine, and provide the video memory data to the remote control client, so that the remote control client can remotely control the virtual machine based on the video memory data. . In an achievable manner, the virtual remote control unit b3 can be used to implement the functions of a virtual network console (virtual network console, VNC), including performing VNC protocol analysis, and obtaining video memory data according to operations such as keyboard and mouse.

The asynchronous information processing unit b4 is used to implement the asynchronous information processing function, that is, to obtain the information required by the virtual machine management sub-module 1121b to perform asynchronous operations from the virtual machine. During the running process of the virtual machine, the virtual machine management sub-module 1121b needs to implement some asynchronous operations according to the data obtained from the virtual machine. At this time, the asynchronous information processing unit b4 is configured to acquire relevant information from the virtual machine in a non-busy state of the virtual machine. For example, the asynchronous information processing unit b4 may acquire some configuration information of the virtual machine when the virtual machine is powered on, so that the virtual machine management sub-module 1121b can perform corresponding asynchronous operations according to the configuration information.

The parameter parsing unit b5 is used for implementing a parameter parsing function, that is, parsing the communication protocol used for running the virtual machine and the parameters required for running the virtual machine. For example, the parameter parsing unit b5 is used to parse parameters required for creating a virtual machine.

As can be seen from the above, the functions provided by the virtual machine management sub-module 1121b can be regarded as functions that have little relevance to the virtual machine business and/or involve asynchronous operations. Such functions have no impact on the internal performance of the virtual machine itself, but consume server resources. but larger. Therefore, by deploying the virtual machine management sub-module 1121b on the uninstall card 12, the occupation of server resources by the virtual machine simulator 112 can be effectively reduced, and the service performance of the virtual machine can be effectively guaranteed.

It should be noted that the above contents of the management functions of the virtual machine management module 1121 are only examples of the management functions of the virtual machine management module 1121, and are not used to limit the types of management functions of the virtual machine management module 1121. The types of management functions of the 1121 can be adjusted according to the application requirements of the virtual machine.

In an implementation manner, as shown in FIG. 4 , the virtual machine simulator 112 further includes a virtual machine hardware simulation module 1122, the virtual machine hardware simulation module 1122 is set in the server 11, and the virtual machine hardware simulation module 1122 is used for according to the server The hardware of 11 emulates the hardware of the virtual machine. The virtual machine hardware emulation module 1122 is used to implement a function strongly related to the virtual machine function, and the function basically does not consume server resources. Therefore, even if the virtual machine hardware emulation module 1122 is deployed in the server 11, the virtual machine hardware emulation module The probability of 1122 affecting the business performance of the virtual machine is also very low, which can effectively ensure the business performance of the virtual machine.

Optionally, as shown in FIG. 4 , the hardware of the virtual machine includes one or more of the following: a virtual processor (virtual central processing unit) 1122a, a virtual memory 1122b, a virtual motherboard chip 1122c, a virtual input and output device 1122d, and a virtual bus 1122e .

The virtual machine hardware simulation module 1122 simulates the hardware of the virtual machine, which means that the virtual machine hardware simulation module 1122 provides the virtual machine with the function of the hardware corresponding to the virtual machine. For example, the virtual machine hardware simulation module 1122 simulates the virtual processor 1122a, which means that the virtual machine hardware simulation module 1122 provides the virtual machine with the computing capability of the virtual processor 1122a. For another example, the virtual machine hardware simulation module 1122 simulates the virtual memory 1122b, which means that the virtual machine hardware simulation module 1122 provides the virtual machine with the virtual memory 1122b to temporarily store the operation data in the virtual processor 1122a and the data exchanged with the external memory such as the hard disk. Function. For another example, the virtual machine hardware emulation module 1122 emulates the virtual motherboard chip 1122c, which means that the virtual machine hardware emulation module 1122 provides the virtual machine with the virtual motherboard chip 1122c to connect different virtual circuit elements, transmits information between different virtual circuit elements, and receives external information and pass the external information to the corresponding virtual circuit element processing and other functions. For another example, the virtual machine hardware simulation module 1122 simulates the virtual input and output device 1122d, which means that the virtual machine hardware simulation module 1122 provides the virtual machine with the input and output functions of the input and output device. For another example, the virtual machine hardware simulation module 1122 simulates the virtual bus 1122e, which means that the virtual machine hardware simulation module 1122 provides the virtual machine with the function of transferring information between the virtual processor 1122a, the virtual memory 1122b and the virtual input/output device 1122d.

Optionally, the virtual memory 1122b includes processor memory allocated to the virtual processor 1122a, and also includes virtual video memory. The virtual motherboard chip 1122c includes a virtual bus 1122e main bridge chip, a virtual bus 1122e south bridge chip, a virtual bus 1122e north bridge chip, and the like. The virtual input and output devices 1122d include a virtual mouse, a virtual keyboard, and the like. The type of the virtual bus 1122e may be determined according to the application requirements of the virtual machine. For example, the virtual bus 1122e may be a virtual PCIe bus, a CXL bus, a PCI bus, an ISA bus, or the like.

It should be noted that the above content of the hardware of the virtual machine is only an example of the hardware of the virtual machine simulated by the virtual machine hardware simulation module 1122, and is not used to limit the type of hardware of the virtual machine simulated by the virtual machine hardware simulation module 1122. The type of hardware of the virtual machine simulated by the machine hardware simulation module 1122 can be adjusted according to the application requirements of the virtual machine.

In the virtual machine simulator 112 provided in the embodiment of the present application, since the virtual machine management agent sub-module 1121a and the virtual machine hardware simulation module 1122 are deployed on the server 11, and the virtual machine management sub-module 1121b is deployed on the uninstall card 12, the The management function of the virtual machine can only be realized through the cooperation of the virtual machine VM, the virtual machine management agent sub-module 1121a and the virtual machine management sub-module 1121b. For example, in some functions for virtual machines, the virtual machine management sub-module 1121b needs to acquire data from the storage resources of the virtual machine VM, and perform corresponding operations according to the acquired data.

A connection channel is established between the server 11 and the uninstall card 12, and the virtual machine management agent sub-module 1121a can communicate with the virtual machine management sub-module 1121b through the connection channel, so that the virtual machine management agent sub-module 1121a can access the storage resources of the virtual machine from the virtual machine management agent sub-module 1121a. The acquired data is transmitted to the virtual machine management sub-module 1121b. Moreover, in order to reduce the occupation of server resources due to the overhead of acquiring data, at least two communication channels implemented by software may be established based on the connection channel.

In an implementation manner, the at least two communication channels implemented by software include: an address acquisition channel and a DMA channel. The address acquisition channel is used for the virtual machine management agent sub-module 1121a to send the storage address of the data in the virtual memory to the virtual machine management sub-module 1121b. The storage address can be represented by the first address and offset for the data. Optionally, the address acquisition channel between the virtual machine management agent sub-module 1121a and the virtual machine management sub-module 1121b may be a socket connection channel (such as a vsock channel). The DMA channel is used for the virtual machine management sub-module 1121b to obtain the required data from the virtual memory through DMA based on the storage address of the data in the virtual memory.

The storage address of the data in the virtual memory belongs to the control plane information. Since the data amount of the control plane information is small, the transmission of the control plane information through the address acquisition channel can effectively reduce the overhead of transmitting the control plane information and the occupation of server resources. In addition, data is obtained from virtual memory through DMA. Since DMA allows direct data transfer between external devices and storage resources, the CPU does not need to be involved in the data transfer process, which can effectively reduce the need to obtain data from storage resources. Occupation of server resources.

For example, a vsock channel is set between the virtual machine management sub-module 1121b and the virtual machine management agent sub-module 1121a. When implementing the VNC function, the virtual machine management submodule 1121b establishes a connection with the VNC client. When the video memory data needs to be acquired, the virtual machine management sub-module 1121b may send an address acquisition request to the virtual machine management agent sub-module 1121a through the vsock channel. The virtual machine management agent sub-module 1121a can obtain the address of the video memory data in the virtual video memory of the virtual machine according to the address obtaining request, and then sends the address to the virtual machine management sub-module 1121b through the vsock channel. Then, the virtual machine management sub-module 1121b obtains video memory data from the virtual video memory of the virtual machine through DMA according to the address, and sends the video memory data to the VNC client. The implementation of the VNC function means that the VNC client obtains video memory data from the virtual video memory of the virtual machine, displays the video memory data on the VNC client, and then remotely controls the virtual machine according to the displayed content.

In an achievable manner, as shown in FIG. 5 , the specific implementation process of the VNC function may include the following steps:

Step S1, after the virtual machine is started, the virtual machine management agent sub-module 1121a and the virtual machine management sub-module 1121b start to run, the server 11 pulls up the management agent main thread, the unload card 12 pulls up the vsk_iothread thread and the vnc_worker thread, and in the A vsock channel is established between the vsk_iothread thread and the management agent main thread. Among them, the function of the virtual machine management agent sub-module 1121a is implemented by the management agent main thread in the server 11, the function of the virtual machine management sub-module 1121b is implemented by the vsk_iothread thread and the vnc_worker thread, and the virtual memory 1122b is allocated with a virtual video memory vga.vram .

In step S2, the vnc_worker thread monitors a connection request sent by the VNC client, where the connection request carries an internet protocol (IP) address and a port address of the VNC client.

Step S3, the vnc_worker thread establishes a communication connection with the VNC client according to the IP address and port address of the VNC client, and starts the timer in the virtual machine management sub-module 1121b.

In step S4, the vsk_iothread thread sends an address acquisition request to the management agent main thread through the vsock channel under the trigger of the timer.

Step S5, the management agent main thread in the virtual machine management agent sub-module 1121a monitors the message sent through the vsock channel, parses the address acquisition request after receiving the address acquisition request, and then queries the virtual video memory vga.vram according to the parsed content to obtain Get the address of the video memory data in the virtual video memory, and send the queried video memory address to the vsk_iothread thread through the vsock channel. The operating system of the virtual machine is used to update the video memory data in the virtual video memory according to the running state of the virtual machine.

The address of the video memory data in the virtual video memory can be reflected by information such as the physical address of the virtual video memory, a dirty page bitmap, an image dirty bitmap, and the like. In addition, the virtual machine management agent sub-module 1121a also simulates a VGA virtual device. When implementing the VNC function, the video memory data can be stored in the video memory memory allocated to the VGA virtual device. Correspondingly, when querying the video memory address, you can query The video memory data is implemented at the address in the video memory memory allocated to the VGA virtual device. In addition, when the virtual machine works in graphics mode and text mode respectively, the representation of this address is slightly different. In graphics mode, the address is represented by the VGA memory physical address, memory dirty page bitmap and image dirty page bitmap and other information. In text mode, the address is represented by information such as the physical address stored in the encoding result and the image dirty page bitmap. At the same time, in order to facilitate the VNC client to display correctly according to the video memory data, in addition to obtaining the address of the video memory data in the virtual video memory, it is also necessary to obtain the basic composition information of the video memory data. For example, in the graphics mode, it is also necessary to obtain the image Basic composition information. Dirty pages are pages that have updated content in memory but have not yet been flushed to disk.

Step S6, the vsk_iothread thread sends the video memory address obtained from the vsock channel to the vnc_worker thread.

Step S7, the vnc_worker thread obtains the video memory data from the virtual video memory by means of DMA according to the received video memory address.

In addition, before the vnc_worker thread obtains the video memory data according to the video memory address, it also needs to do preparations such as applying for the memory space for storing the obtained video memory data according to the basic composition information of the video memory data.

In step S8, the vnc_worker thread encodes the acquired video memory data, and sends the encoded video memory data to the VNC client.

In step S9, the VNC client performs display according to the received video memory data, and remotely controls the virtual machine according to the displayed image.

There are various implementation scenarios of the VNC function. The following takes inputting through the mouse and keyboard on the VNC client to realize the remote control of the virtual machine as an example to describe the implementation process. Figure 6 is the application A schematic diagram of an implementation process of the implementation scenario provided by the embodiment.

As shown in FIG. 6, in response to the input operation of the mouse and the keyboard on the VNC client, the virtual input and output devices simulated by the virtual machine hardware simulation module 1122 include a virtual mouse and a virtual keyboard, and the virtual machine management sub-module 1121b is configured with the virtual input and output devices. Shadow devices for mouse and keyboard. After the mouse or keyboard on the VNC client inputs information, the VNC client will transmit the input event to the vnc_worker thread, and the vnc_worker thread will then transmit the input event to the console in the offload card 12 . The console determines whether the input event is generated by the mouse or the keyboard according to the input event, and then feeds back the identification of the input event to the vnc_worker thread. The vnc_worker thread sends the identification to the shadow device, the shadow device records the identification, and sends the identification and the input information to the vsk_iothread thread. The vsk_iothread thread sends the identification and input information to the management agent main thread through the vsock channel. The management agent main thread sends the input information to the corresponding virtual input and output device 1122d according to the identification, that is, when the identification indicates a keyboard, the input information is sent to the virtual keyboard, and when the identification indicates a mouse, the input information is sent to virtual mouse. Then, the corresponding input and output device performs the relevant input operation according to the input information, and sends the result of the input operation to the operating system of the virtual machine through an interrupt signal, so that the operating system updates the video memory according to the result of the input operation. Then, the management agent main thread obtains the updated video memory address, and sends the updated video memory address to the vsk_iothread thread through the vsock channel, so that the vnc_worker thread can obtain the updated video memory data according to the address, and send the updated video memory data to VNC client to display the result of the input operation on the VNC client. The process from receiving the input information to sending the information to the server 11 by the uninstall card 12 can be regarded as the uninstall card 12 encapsulates the input information of the VNC client and transforms it into a remote call event, and then registers the callback and the corresponding shadow device through the corresponding shadow device. vsock channel, the process of sending remote invocation events to server 11. The process from the server 11 receiving the information sent by the vsk_iothread thread through the vsock channel to updating the video memory according to the operation result can be regarded as the server 11 reversely analysing the remote call event, and finding the corresponding event processing function of the device call according to the identifier in the content obtained by the analysis process of processing.

It should be noted that there are various functions involved in the virtual machine that need to be implemented through the cooperation of the virtual machine, the virtual machine management agent sub-module 1121a and the virtual machine management sub-module 1121b, which are not listed one by one in the embodiments of the present application. However, it should be understood that, when implementing the functions that require the cooperation of the virtual machine, the virtual machine management agent sub-module 1121a and the virtual machine management sub-module 1121b, the implementation process can refer to the implementation process of the aforementioned VNC function. I won't repeat them one by one.

Optionally, as shown in FIG. 4 , the offload card 12 can be connected to the storage resource and/or the network resource through the network card 13, so that the offload card 12 can send the I/O request of the virtual machine running in the server 11 to the storage resource. and/or network resources for processing. For example, the offload card 12 is configured with a network card 13 , and the offload card 12 can be connected to storage resources and network resources in the cloud data center through the network card 13 .

In an implementation manner, as shown in FIG. 4 , a server management module 113 is also deployed in the uninstall card 12 , and the server management module 113 is used to manage the running state of the server system 1 . For example, the server management module 113 is used to manage the virtualization platform (such as implementing the functions of libvirt), be responsible for maintaining and managing the computing resources of the virtual machine (such as implementing the functions of Nova), and be responsible for system installation, upgrade, and deployment in the server 11 functions related to operation and maintenance.

In addition, when the server system 1 is deployed in the cloud data center, the server management module 113 is further configured to receive a management plane operation request sent by the cloud management platform 2, and perform corresponding operations according to the management plane operation request. The management plane operation request may be a request sent by the cloud management platform 2 to manage the virtual machine, or may be a request sent by the user to manage the virtual machine through the cloud management platform 2 .

The management plane operation request is received by the server management module 113 and the corresponding operation is performed, so that the server 11 does not need to receive the management plane operation request and perform the corresponding operation, thereby reducing the overhead of server resources due to receiving the management plane request and performing the corresponding operation. , which can further ensure the service performance of the virtual machine.

To sum up, in the server system provided by the embodiments of the present application, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the virtual machine management sub-module is deployed on the uninstall card , realizes that the functions that consume more server resources in the virtual machine emulator are deployed in the uninstall card, which reduces the occupation of server resources by the overhead of the virtual machine emulator itself, and reduces the overhead and performance of the virtual machine emulator itself. The virtual machine provided by the user effectively guarantees the business performance of the virtual machine due to the influence of competing resources on the business performance of the virtual machine. In addition, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the business performance of the virtual machine (such as the scheduling performance of the virtual processor and the I/O read and write performance of the virtual machine) is also It will not produce side effects, so it has a good application prospect.

The embodiment of the present application also provides a method for creating a virtual machine. The method for creating a virtual machine is applied to the server system provided by the embodiment of the present application. The server system includes a server and an offload card inserted in the server, and a connection channel is established between the server and the offload card. Optionally, a server management module is also deployed in the uninstall card. Also, the offload card is connected to storage resources and/or network resources through a network card. Further, the server and the offload card may be connected through a bus, for example, a PCIe bus through a virtual high-speed peripheral component interconnection.

As shown in Figure 7, the virtual machine creation method includes:

Step 701: The server receives a virtual machine creation request sent by the cloud platform.

When a user needs to use a virtual machine to implement services, the user can operate on the client, and the operation will trigger a virtual machine creation request to be sent to the cloud management platform for requesting the creation of a virtual machine, and the cloud management platform will send the virtual machine to the server. Create a request.

The virtual machine creation request carries the configuration requirements of the virtual machine. The configuration requirements for this virtual machine are used to indicate the specifications of the virtual machine requested to be created. Moreover, since the virtual machine needs to run based on the hardware of the virtual machine provided by the virtual machine simulator, the configuration requirement may indicate the specifications of the hardware of the virtual machine required for the virtual machine to run. For example, the hardware specifications of the virtual machine indicated by the configuration requirements may include: the total number of virtual processors, the topology of virtual processors, the type of virtual processors, the amount of virtual memory, the usage policy of virtual memory, the number of virtual memory slots Setting method, type of virtual input and output device, specification of virtual input and output device, storage path of the image file of the operating system of the virtual machine, storage path of the image file of the application program used by the virtual machine to realize the business, and the operating system of the virtual machine Specification parameters such as configuration information of initialization parameters.

The topology of the virtual processors is used to indicate one or more of the total number of virtual processors that the virtual machine needs to use, the total number of virtual cores included in each virtual processor, and the total number of hyperthreads included in each virtual processor. multiple. The type of virtual processor is used to indicate the word length and model of the virtual processor. The word size of a virtual processor is the number of bits that the virtual processor can process in parallel at a time. The virtual memory usage policy is used to indicate how virtual memory is used by multiple virtual processors used by the virtual machine. For example, the virtual memory usage policy is used to instruct multiple virtual CPUs used by the virtual machine to use the virtual memory in a non-uniform memory access (NUMA) manner. The specification of the virtual input/output device is used to indicate the configuration specification of the virtual input/output device. For example, when the virtual input and output device is a virtual network card, the specification of the virtual network card indicates information such as the description information of the queue of the virtual network card and the media access control (media access control, MAC) address and other information. When the virtual input/output device is a virtual disk, the specification of the virtual disk indicates information such as the capacity of the virtual disk. The image file of the operating system of the virtual machine is used to record information related to the interaction between the operating system and the hardware of the virtual machine. The image file of the application program used by the virtual machine to implement the service is used to record the relevant information about the interaction between the application program used to implement the service and the operating system of the virtual machine running on the basis of the operating system of the virtual machine. For example, when the operating system of the virtual machine is a LINUX system, the image file of the operating system of the virtual machine may be a kernel image file, and the image file of an application program used by the virtual machine to implement services may be an initrd image file.

Step 702: The server creates a virtual machine in the server according to the virtual machine creation request, starts the virtual machine management agent sub-module set in the server, and starts the virtual machine management sub-module set in the uninstall card through the connection channel.

After the VMM in the server receives the virtual machine creation request, it will pull up the virtual machine hardware simulation module according to the virtual machine creation request, and use the virtual machine hardware simulation module according to the configuration requirements of the virtual machine indicated by the virtual machine creation request. Hardware emulates the hardware of a virtual machine. In addition, the VMM will also start the virtual machine management agent sub-module set in the server, and start the virtual machine management sub-module set in the uninstall card through the connection channel, so as to facilitate the virtual machine management agent sub-module and the virtual machine management sub-module through the virtual machine management sub-module. Implements virtual machine management functions.

Optionally, the hardware of the virtual machine includes one or more of the following: a virtual processor, a virtual memory, a virtual motherboard chip, a virtual input and output device, and a virtual bus. For example, the virtual machine hardware emulation module can indicate the total number of virtual processors, the topology of virtual processors, the type of virtual processors, the amount of virtual memory, the usage policy of virtual memory, and the setting method of virtual memory slots according to the configuration requirements. , the type of virtual input and output devices and the specifications of virtual input and output devices and other specifications related to the hardware of the virtual machine, to simulate the hardware of the virtual machine.

Optionally, the management function of the virtual machine includes: one or any combination of a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function and a parameter parsing function.

To sum up, in the virtual machine creation method provided by the embodiments of the present application, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the virtual machine management sub-module is deployed in the uninstaller. On the card, the functions that consume more server resources in the virtual machine emulator are deployed in the unloading card, which reduces the occupation of server resources by the overhead of the virtual machine emulator itself, and reduces the overhead of the virtual machine emulator itself. The business performance of the virtual machine is effectively guaranteed due to the impact of competing resources with the virtual machine that provides services to the user on the business performance of the virtual machine. In addition, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the business performance of the virtual machine (such as the scheduling performance of the virtual processor and the I/O read and write performance of the virtual machine) is also It will not produce side effects, so it has a good application prospect.

It should be noted that, the sequence of steps of the method for creating a virtual machine provided by the embodiment of the present application may be adjusted appropriately, and the steps may be correspondingly increased or decreased according to the situation. Any person skilled in the art who is familiar with the technical scope disclosed in the present application can easily think of any variation of the method, which should be covered by the protection scope of the present application, and thus will not be repeated here.

Moreover, those skilled in the art can clearly understand that, for the convenience and brevity of description, the implementation process of the above-described virtual machine creation method and its related content can refer to the corresponding content in the foregoing server system embodiment, which is not described here. Repeat.

The present application provides an apparatus for creating a virtual machine, the apparatus for creating a virtual machine is applied to a server, an uninstall card is inserted on the server, and a connection channel is established between the server and the uninstall card. As shown in FIG. 8 , the apparatus 80 for creating a virtual machine includes: :

The receiving module 801 is configured to receive a virtual machine creation request sent by the cloud management platform.

The processing module 802 is configured to create a virtual machine in the server according to the virtual machine creation request, start the virtual machine management agent sub-module set in the server, and start the virtual machine management sub-module set in the uninstall card through the connection channel.

The virtual machine is implemented by a virtual machine simulator, the virtual machine simulator includes a virtual machine management module, the virtual machine management module includes a virtual machine management agent sub-module and a virtual machine management sub-module, and the virtual machine management agent sub-module communicates with the virtual machine through a connection channel The virtual machine management sub-module is used for communication, and the virtual machine management sub-module is used to realize the management function of the virtual machine.

To sum up, in the virtual machine creation apparatus provided by the embodiments of the present application, the virtual machine management module is split into a virtual machine management agent sub-module and a virtual machine management sub-module, and the virtual machine management sub-module is deployed in the uninstaller. On the card, the functions that consume more server resources in the virtual machine emulator are deployed in the unloading card, which reduces the occupation of server resources by the overhead of the virtual machine emulator itself, and reduces the overhead of the virtual machine emulator itself. The business performance of the virtual machine is effectively guaranteed due to the impact of competing resources with the virtual machine that provides services to the user on the business performance of the virtual machine. In addition, the virtual machine management module is divided into a virtual machine management agent sub-module and a virtual machine management sub-module, and the business performance of the virtual machine (such as the scheduling performance of the virtual processor and the I/O read and write performance of the virtual machine) is also It will not produce side effects, so it has a good application prospect.

Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the working principle of the virtual machine creation device and module described above can be referred to the corresponding content in the foregoing server system embodiment, which is not repeated here.

The embodiments of the present application provide a computer device. FIG. 9 exemplarily provides a possible architectural diagram of a computer device. As shown in FIG. 9 , the computer device 90 may include a processor 901 , a memory 902 , a communication interface 903 and a bus 904 . In a computer device, the number of processors 901 may be one or more, and FIG. 9 only illustrates one of the processors 901 . Optionally, the processor 901 may be a central processing unit (Central Processing Unit, CPU). If the computer device has multiple processors 901, the multiple processors 901 may be of different types, or may be the same. Optionally, multiple processors of the computer device may also be integrated into a multi-core processor.

The memory 902 stores computer instructions and data, and the memory 902 may store computer instructions and data required to implement the virtual machine creation method provided by the present application. The memory 902 can be any one or any combination of the following storage media: non-volatile memory (such as read-only memory (Read-Only Memory, ROM), solid state disk (Solid State Disk, SSD), hard disk (Hard Disk) Drive, HDD), optical disc, etc., volatile memory.

The communication interface 903 may be any one or any combination of the following devices: a network interface (eg, an Ethernet interface), a wireless network card, and other devices with a network access function. The communication interface 903 is used for data communication between the computer device and other nodes or other computer devices.

FIG. 9 also exemplarily depicts bus 904 . The bus 904 can connect the processor 901 with the memory 902 and the communication interface 903 . In this way, through the bus 904, the processor 901 can access the memory 902, and can also use the communication interface 903 to perform data interaction with other nodes or other computer devices.

In the present application, the computer device executes the computer instructions in the memory 902 to implement the virtual machine creation method provided by the embodiments of the present application. For example, receive a virtual machine creation request sent by the cloud management platform; create a virtual machine in the server according to the virtual machine creation request, start the virtual machine management agent sub-module set in the server, and start the virtual machine set in the uninstall card through the connection channel machine management sub-module. Moreover, by executing the computer instructions in the memory 902 by the computer device, the implementation process of executing the steps of the computer device may refer to the corresponding descriptions in the foregoing method embodiments.

Embodiments of the present application further provide a storage medium, where the storage medium is a non-volatile computer-readable storage medium. When instructions in the storage medium are executed by a processor, the method for creating a virtual machine provided by the embodiments of the present application is implemented. .

The embodiments of the present application also provide a computer program product containing instructions, when the computer program product is run on a computer, the computer is made to execute the virtual machine creation method provided by the embodiments of the present application.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, etc.

In the embodiments of the present application, the terms "first", "second" and "third" are only used for description purposes, and cannot be understood as indicating or implying relative importance. The term "at least one" refers to one or more, and the term "plurality" refers to two or more, unless expressly limited otherwise.

The term "and/or" in this application is only an association relationship to describe associated objects, which means that there can be three kinds of relationships, for example, A and/or B, which can mean that A exists alone, A and B exist at the same time, independently There are three cases of B. In addition, the character "/" in this document generally indicates that the related objects are an "or" relationship.

The above are only optional embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the concept and principles of the present application shall be included in the protection of the present application. within the range.

Claims

A server system, characterized in that the server system includes a server and an unloading card inserted in the server, and a connection channel is established between the server and the unloading card;

The server is provided with a virtual machine, and the virtual machine is implemented by a virtual machine simulator, the virtual machine simulator includes a virtual machine management module, and the virtual machine management module includes a virtual machine management agent sub-module and a virtual machine management module submodule;

The virtual machine management agent sub-module is deployed on the server, and the virtual machine management sub-module is deployed on the uninstall card, wherein the virtual machine management agent sub-module communicates with the virtual machine management sub-module through the connection channel. module for communication, and the virtual machine management sub-module is used to implement the management function of the virtual machine.
The system according to claim 1, wherein the virtual machine simulator further comprises a virtual machine hardware simulation module, the virtual machine hardware simulation module is provided in the server, and the virtual machine hardware simulation module is used for The hardware of the virtual machine is simulated according to the hardware of the server.
The system according to claim 1 or 2, wherein the virtual machine management sub-module is further configured to access the data of the virtual machine from the memory of the server in a direct memory access DMA manner through the connection channel .
The system according to any one of claims 2 to 3, wherein the hardware of the virtual machine includes one or more of the following: a virtual processor, a virtual memory, a virtual motherboard chip, a virtual input and output device, and a virtual bus.
The system according to any one of claims 1 to 4, wherein the management functions include: a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function, and a parameter parsing function. one or any combination.
The system according to any one of claims 1 to 5, wherein the server and the offload card are connected through a virtual high-speed peripheral component interconnect PCIe bus or a computing express link CXL bus.
A method for creating a virtual machine, characterized in that the method is applied to a server system, the server system includes a server and an offload card inserted in the server, and a connection channel is established between the server and the offload card, and the method include:

The server receives the virtual machine creation request sent by the cloud management platform;

The server creates a virtual machine in the server according to the virtual machine creation request, starts the virtual machine management agent sub-module set in the server, and starts the virtual machine set in the uninstall card through the connection channel. machine management sub-module;

Wherein, the virtual machine is implemented by a virtual machine simulator, and the virtual machine simulator includes a virtual machine management module, and the virtual machine management module includes the virtual machine management agent sub-module and the virtual machine management sub-module. The virtual machine management agent sub-module communicates with the virtual machine management sub-module through the connection channel, and the virtual machine management sub-module is used to implement the management function of the virtual machine.
The method according to claim 7, wherein the virtual machine simulator further comprises a virtual machine hardware simulation module, the virtual machine hardware simulation module is provided in the server, and the virtual machine hardware simulation module is used for The hardware of the virtual machine is simulated according to the hardware of the server.
The method according to claim 7 or 8, wherein the virtual machine management submodule is further configured to access the data of the virtual machine from the memory of the server in a direct memory access DMA manner through the connection channel .
The method according to any one of claims 8 to 9, wherein the hardware of the virtual machine includes one or more of the following: a virtual processor, a virtual memory, a virtual motherboard chip, a virtual input and output device, and a virtual bus.
The method according to any one of claims 7 to 10, wherein the management functions include: a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function, and a parameter parsing function. one or any combination.
The method according to any one of claims 7 to 11, wherein the server and the offload card are connected through a virtual high-speed peripheral component interconnect PCIe bus or a computing express link CXL bus.
An apparatus for creating a virtual machine, wherein the apparatus is applied to a server, an uninstall card is inserted on the server, and a connection channel is established between the server and the uninstall card, and the apparatus includes:

The receiving module is used to receive the virtual machine creation request sent by the cloud management platform;

a processing module, configured to create a virtual machine in the server according to the virtual machine creation request, start a virtual machine management agent sub-module set in the server, and start and set in the uninstall card through the connection channel The virtual machine management submodule;

Wherein, the virtual machine is implemented by a virtual machine simulator, and the virtual machine simulator includes a virtual machine management module, and the virtual machine management module includes the virtual machine management agent sub-module and the virtual machine management sub-module. The virtual machine management agent sub-module communicates with the virtual machine management sub-module through the connection channel, and the virtual machine management sub-module is used to implement the management function of the virtual machine.
The apparatus according to claim 13, wherein the virtual machine simulator further comprises a virtual machine hardware simulation module, the virtual machine hardware simulation module is provided in the server, and the virtual machine hardware simulation module is used for The hardware of the virtual machine is simulated according to the hardware of the server.
The apparatus according to claim 13 or 14, wherein the virtual machine management sub-module is further configured to access the data of the virtual machine from the memory of the server in a direct memory access DMA manner through the connection channel .
The apparatus according to any one of claims 14 to 15, wherein the hardware of the virtual machine includes one or more of the following: a virtual processor, a virtual memory, a virtual motherboard chip, a virtual input and output device, and a virtual bus.
The device according to any one of claims 13 to 16, wherein the management functions include: a virtual machine life cycle management function, a virtual machine migration function, a virtual remote control function, an asynchronous information processing function, and a parameter parsing function. one or any combination.
The apparatus according to any one of claims 13 to 17, wherein the server and the offload card are connected through a virtual high-speed peripheral component interconnect PCIe bus or a computing express link CXL bus.
A computer device, characterized in that the computer device comprises: a processor and a memory, wherein a computer program is stored in the memory; when the processor executes the computer program, the computer device implements any one of claims 7 to 12 the method described.
A storage medium, characterized in that, when an instruction in the storage medium is executed by a processor, the method of any one of claims 7 to 12 is implemented.