CN111857944A

CN111857944A - Method, device, system and computer readable storage medium for virtual machine live migration

Info

Publication number: CN111857944A
Application number: CN201910365413.2A
Authority: CN
Inventors: 姚旭; 马昊骢; 吴纯赟
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-04-30
Filing date: 2019-04-30
Publication date: 2020-10-30

Abstract

The application provides a method, a device, a system and a computer readable storage medium for virtual machine live migration, wherein the method comprises the following steps: backing up the mirror image of the virtual machine in the source host machine to a distributed storage system; and under the condition that the virtual machine meets the preset condition of live migration, migrating the virtual machine from the source host machine to the target host machine, wherein the target image of the migrated virtual machine in the target host machine comprises a backup image copied from the distributed storage system. Compared with the existing scheme of migrating all the mirror images of the virtual machines from the source host machine, most data or all data of the mirror images of the virtual machines in the target host machine come from the backup mirror images, so that the mirror image data volume directly transmitted from the source host machine to the target host machine is reduced, and the total time consumption in the live migration process is reduced.

Description

Method, device, system and computer readable storage medium for virtual machine live migration

Technical Field

The present application relates to the field of virtual machine technologies, and in particular, to a method, an apparatus, a system, and a computer-readable storage medium for live migration of a virtual machine.

Background

Live migration, also called live migration, refers to moving a virtual machine from one physical server to another without interrupting the service. In the hot migration process, the virtual machine can be normally used, so that a user does not perceive the migration process.

Most of the existing hot migration technologies migrate the mirror image and the memory of the virtual machine stored in the source host to the target host. However, the live migration data volume of the virtual machine is large, so that the time consumption of the whole live migration process is long.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

In view of the above, an object of the present application is to provide a method, an apparatus, a system and a computer-readable storage medium for virtual machine live migration, so as to reduce the time consumption of the overall live migration process.

In view of the above, in a first aspect, the present application provides a method for virtual machine live migration, where the method is applied to a virtual machine system, where the virtual machine system includes a source host of a virtual machine, a distributed storage system, and a target host of the virtual machine, and the method includes: backing up the mirror image of the virtual machine in the source host machine to a distributed storage system; and under the condition that the virtual machine meets the preset condition of live migration, migrating the virtual machine from the source host machine to the target host machine, wherein the target image of the migrated virtual machine in the target host machine comprises a backup image copied from the distributed storage system.

Therefore, compared with the existing scheme of migrating all the mirror images of the virtual machines from the source host, most or all data of the mirror images of the virtual machines in the target host come from the backup mirror images, so that the mirror image data volume directly transmitted from the source host to the target host is reduced, and the total time consumption in the live migration process is reduced.

In one possible embodiment, backing up an image of a virtual machine in a source host to a distributed storage system includes: when a virtual machine is initially established in a source host machine, backing up all mirror image data of the virtual machine to a distributed storage system; after the virtual machine is established, periodically recording incremental data of a mirror image of the virtual machine, and periodically backing up the incremental data to the distributed storage system, wherein the incremental data of the mirror image of the virtual machine is difference data between a current mirror image and a mirror image backed up last time.

Therefore, the method and the device improve the backup efficiency by combining the full backup mode and the incremental backup mode.

In one possible implementation, a first network block device NBD is created in the source host, a reliable autonomous distributed object storage block device RBD is created in the distributed storage system, and the first NBD and the RBD have a first mapping relation; the method for backing up all mirror image data of the virtual machine to the distributed storage system comprises the following steps: backing up all mirror image data to an RBD in the distributed storage system through the first NBD; periodically backing up the incremental data to the distributed storage system, including backing up the incremental data to an RBD in the distributed storage system via the first NBD.

In one possible embodiment, periodically recording incremental data of an image of a virtual machine includes: incremental data of the image of the virtual machine is periodically recorded through a bitmap in the image of the virtual machine.

In one possible embodiment, migrating a virtual machine from a source host to a target host includes: determining incremental data to be migrated of the mirror image of the virtual machine, wherein the incremental data to be migrated is difference data between the current mirror image and the mirror image backed up last time, and the incremental data to be migrated is mirror image data which is not uploaded to a distributed storage system; and migrating the incremental data to be migrated and the backup mirror image in the distributed storage system to a target mirror image in a target host.

Therefore, most of mirror image data or all mirror image data of the virtual machine in the application are copied from the distributed storage system, and do not need to be migrated from the source host, so that the read-write operation of the source host can be reduced, and the bandwidth consumption of the source host in the live migration process can be saved.

In a possible implementation manner, an RBD is created in the distributed storage system, a second network block device NBD is created in the target host, and the second NBD and the RBD have a second mapping relationship; the method for migrating the incremental data to be migrated and the backup mirror image in the distributed storage system to the target mirror image in the target host comprises the following steps: migrating the incremental data to be migrated to a target mirror image in the target host machine through a network, and migrating the backup mirror image in the distributed storage system to the target mirror image in the target host machine through a second NBD and an RBD.

In a second aspect, the present application provides a management apparatus for virtual machine live migration, which is applied to a virtual machine system, where the virtual machine system includes a source host of a virtual machine, a distributed storage system, and a target host of the virtual machine, and includes: the backup module is used for backing up the mirror image of the virtual machine in the source host machine to the distributed storage system; and the migration module is used for migrating the virtual machine from the source host machine to the target host machine under the condition that the virtual machine meets the preset condition of live migration, wherein the target image of the migrated virtual machine in the target host machine comprises a backup image copied from the distributed storage system.

In one possible embodiment, the backup module includes: the backup unit is used for backing up all mirror image data of the virtual machine to the distributed storage system when the virtual machine is initially established in the source host; and the recording backup unit is used for periodically recording the incremental data of the mirror image of the virtual machine after the virtual machine is established, and periodically backing up the incremental data to the distributed storage system, wherein the incremental data of the mirror image of the virtual machine is the difference data between the current mirror image and the mirror image backed up last time.

In one possible implementation, a first NBD is created in the source host, an RBD is created in the distributed storage system, and the first NBD and the RBD have a first mapping relation; the backup unit is further used for backing up all mirror image data to an RBD in the distributed storage system through the first NBD; and the recording backup unit is also used for backing up the incremental data to the RBD in the distributed storage system through the first NBD.

In a possible embodiment, the recording backup unit is further configured to periodically record incremental data of the image of the virtual machine through a bitmap in the image of the virtual machine.

In one possible embodiment, the migration module includes: the device comprises a determining unit, a storage unit and a processing unit, wherein the determining unit is used for determining incremental data to be migrated of a mirror image of a virtual machine, the incremental data to be migrated is difference data between a current mirror image and a mirror image which is backed up last time, and the incremental data to be migrated is mirror image data which is not uploaded to a distributed storage system; and the migration unit is used for migrating the incremental data to be migrated and the backup mirror image in the distributed storage system to the target mirror image in the target host.

In one possible implementation, an RBD is created in the distributed storage system, a second NBD is created in the target host, and the second NBD and the RBD have a second mapping relationship; the migration unit is further configured to migrate the incremental data to be migrated to a target image in the target host through a network, and migrate the backup image in the distributed storage system to the target image in the target host through the second NBD and the RBD.

In a third aspect, the present application provides a system for implementing virtual machine live migration, where the system includes: a source host, a target host, a distributed storage system, and a management apparatus for live migration of a virtual machine according to any one of the second aspect; the source host is used for backing up the mirror image of the virtual machine to the distributed storage system according to the backup instruction sent by the management device; and the target host is used for creating a target image of the virtual machine according to the migration instruction sent by the management device, wherein the target image comprises a backup image copied from the distributed storage system.

In a fourth aspect, the present application provides an electronic device, which may include: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operated, the processor executing the machine-readable instructions to perform the steps of the method according to any one of the first aspect.

In a fifth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method according to any one of the first aspect.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

FIG. 1 is a schematic diagram illustrating a scenario of a method for live migration of a virtual machine in the prior art;

fig. 2 is a schematic view of a scenario illustrating an optional virtual machine live migration method according to an embodiment of the present application;

FIG. 3 is a flowchart illustrating a method for virtual machine live migration according to an embodiment of the present disclosure;

fig. 4 shows a flowchart of step S310 in a method for live migration of a virtual machine according to an embodiment of the present application;

fig. 5 shows a flowchart of step S320 in a method for live migration of a virtual machine according to an embodiment of the present application;

FIGS. 6 and 7 are diagrams illustrating a full backup of an image in a source host according to an embodiment of the present application;

FIG. 8 is a diagram illustrating incremental backup of an image in a source host according to an embodiment of the present disclosure;

FIG. 9 is a schematic diagram of a first NBD and RBD disconnect map and a second NBD and RBD setup map provided by an embodiment of the present application;

FIG. 10 is a diagram illustrating creation of a target image on a target host according to an embodiment of the present application;

FIG. 11 is a schematic diagram illustrating migrating mirrored incremental data and memory from a source host according to an embodiment of the present application;

FIG. 12 is a diagram illustrating a method for copying a backup image from a ceph distributed file system according to an embodiment of the present application;

fig. 13 is a schematic structural diagram illustrating a management apparatus for virtual machine live migration according to an embodiment of the present application;

fig. 14 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.

Detailed Description

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In order to facilitate understanding of the technical solution of the present application, the following describes the implementation steps of the existing thermomigration solution.

Referring to fig. 1, fig. 1 is a schematic view illustrating a scenario of a virtual machine live migration method in the prior art. As shown in fig. 1, source host 110 and target host 120 are network-connected, wherein an image of virtual machine VM1 and VM1 and an image of virtual machine VM2 and VM2 are created in source host 110, and an image of virtual machine VM3 and VM3 are created on target host 120. When the virtual machine in the source host 110 meets the preset condition of live migration, the virtual machine 1 in the source host 110 may be migrated to the target host 120. It is of course understood that multiple virtual machines in source host 110 may also be migrated in parallel.

In addition, the step of migrating the virtual machine VM1 in the source host 110 to the target host 120 includes:

step 1, copying the mirror image of the virtual machine by using a block-mirror technology.

And 2, after the first mirror image copying is finished, starting to migrate the memory of the virtual machine, and continuously and iteratively copying the mirror image.

Step 3, when the remaining virtual machine memory and the mirror data are sufficiently small, the virtual machine VM1 in the source host 110 is suspended, and the remaining memory and the mirror data are migrated to the target host 120.

And 4, pulling up the virtual machine in the target host machine 120 and continuously running, wherein the user operating the virtual machine is not aware because the virtual machine is suspended for a short time.

In the existing live migration scheme, step 1 is to copy the image of the virtual machine VM1 by using a block-mirror technology, and the first image copy time depends on the size of the image. For example, in the case where the image size of the virtual machine VM1 of the first copy is 480GB, the bandwidth for the live migration at this time is 2.4GB (300MB), so that the time taken for the first copy is about 1600 seconds. However, during the live migration of a virtual machine, if the bandwidth usage of the source host 110 has reached a bottleneck, then the bandwidth resources of the source host 110 may be strained by 2.4GB for the live migration compared to the available bandwidth resources. Additionally, if the bandwidth available for live migration is smaller, this situation may result in a longer first copy time, and thus more data written by the user during the first copy, and thus a longer second copy time, and so on, which is a vicious circle, and thus the existing live migration solution may make it very time consuming to migrate virtual machine VM1 from source host 110. In addition, the resources occupied by the virtual machine for migration at the source host 110 may be released long, and thus, it may also reduce the resource scheduling efficiency of the source host 110.

Therefore, in order to solve some of the above problems or issues, the present application performs a full backup of an image of a virtual machine initially created in the source host 110, and backs up all image data of the image to the cloud distributed storage system 130. Subsequently, after the virtual machine runs, the image of the virtual machine is periodically subjected to incremental backup, and the incremental data is stored in the distributed storage system 130. When the source host 110 fails or the resource utilization rate reaches a bottleneck, the live migration is triggered, and when the specified virtual machine in the source host 110 is live migrated, only incremental data of the virtual machine in the source host 110, which is not uploaded to the distributed storage system 130, may be migrated to the target host 120, since the incremental data at this time is very small, the above step 1 may be completed quickly, so that the live migration process may quickly enter step 2, and subsequently after the memory migration of the virtual machine is completed, other mirror image data of the specified virtual machine may be copied from the distributed storage system 130, so as to ensure that the specified virtual machine can normally operate. Therefore, compared to step 1 in the existing live migration scheme, the present application can at least reduce the data amount of the image migrated from the source host 110, thereby reducing the bandwidth consumption of the source host 110 during the live migration.

Referring to fig. 2, fig. 2 is a schematic view of a scenario of an optional virtual machine live migration method according to an embodiment of the present disclosure. As shown in fig. 2, the source host 110 is connected to the target host 120 through the network 140, and the source host 110 and the target host 120 include but are not limited to: a supercomputer, mainframe, minicomputer, microcomputer, or various other physical devices capable of running a virtual machine.

In addition, a plurality of virtual machines and images of a plurality of virtual machines may be created in the source host 110, and accordingly, a plurality of virtual machines and images of a plurality of virtual machines may also be created in the target host 120. It should be noted that the present application does not limit the type of virtual machine. For example, the virtual machine may be a Xen virtual machine, a KVM (kernel virtual machine), a hyperv virtual machine (microsoft virtualization technology), and the like, and all of them should be within the protection scope of the present application.

In addition, the source host 110 is also connected to the distributed storage system 130 through the network 140, and correspondingly, the target host 120 is also connected to the distributed storage system 130 through the network 140. It should be noted that the present application is not limited as to the type of distributed storage system 130. For example, the distributed storage system 130 may be a cloud storage system such as a hundredth cloud or a huayao cloud, besides a server cluster, and all shall be within the protection scope of the present application.

In addition, the source host 110, the target host 120 and the distributed storage system 130 are all connected to a control node 150 through a network 140, and the control node 150 may be one of management devices for virtual machine live migration, which is capable of controlling the live migration process of a virtual machine.

It should be noted that, although fig. 1 illustrates a management apparatus for virtual machine live migration, it should be understood by those skilled in the art that the management apparatus for virtual machine live migration may be integrated in the source host 110 or the target host 120, in addition to being the separate control node 150, and the present application is not limited thereto.

In addition, in the implementation of the present application, the network 140 includes, but is not limited to: a wide area network, a metropolitan area network, or a local area network, and each device in the system is connected through a corresponding interface through a network 140.

The following describes in detail the virtual machine live migration method provided in the embodiment of the present application with reference to fig. 3 according to the content described in the scenario of the virtual machine live migration method shown in fig. 2.

To facilitate understanding of the embodiments of the present application, several elements introduced in the description of the embodiments of the present application are first introduced here:

VM (virtual machine): it means that one or more virtual computers can be simulated on one physical computer through virtual machine software, and these virtual machines work just like real computers, and the operating system and application program can be installed on the virtual machines, and the virtual machines can also access the resources of the network 140, and for the application program running in the virtual machine, the virtual machine works just like real computers.

Host machine: it is the physical machine that runs the virtual machine.

qemu-kvm (virtual machine process): it is located on hosts (including source host 110 and target host 120). Wherein each virtual machine corresponds to one qemu-kvm process.

image (mirror): the storage file is a storage file of the virtual machine, and the data of the virtual machine system and the virtual machine can be stored in the storage file, and the storage file can be stored locally on the host machine or in the distributed storage system 130. In addition, a virtual machine may have multiple images, each of which may be stored in a local or distributed storage system 130, respectively.

block-backup (online backup): it includes full block backup and incrementalmlock backup modes. The full backup mode refers to a complete copy of all data or applications at a certain time point; the incremental backup is directed to the last backup (no matter which mode of backup), all data which are changed after the last backup are backed up, and meanwhile, the incremental mode is not affected by the startup and shutdown and the hot migration of the virtual machine.

bitmap (bitmap): it is used for recording data written since the existence of a bitmap in the case where the bitmap is added to a certain file. And after each backup of the file is finished, the bitmap can record the written data in the file again, so that only the changed data or the difference data compared with the previous backup is needed to be backed up in each backup.

block-mirror (on-line copy): it is used for on-line copying, because on-line copying is an iterative process, and there is new data written in the copying process, copying all data for the first time, and then copying the difference data in the last copying process each time.

RBD (RADOS (reliable autonomous Distributed Object store) BLOCK DEVICE, which may also be referred to as RADOS BLOCK DEVICE): it may be provided in plurality. Wherein, a plurality of RBDs can be mapped with a plurality of corresponding NBDs one by one.

NBD (Network Block Device ): it is used to map the disk space of a remote host or the RADOS block devices on the distributed storage system 130 to a local NBD via network protocols. In the present scenario, the backup image in each distributed storage system 130 has a corresponding one of the NBDs in the hosts.

In the present application, the NBDs (including the first NBD and the second NBD) establish network connections with RBDs in the distributed system 130 via network addresses, forming a logically one-to-one mapping.

map (build map): it means that a mapping is established between two devices.

unmap (break map): it refers to unmapping between two devices.

qcow 2: it is a format of mirroring. If a backing file (bottom image) is specified when the qcow2 file is created, all reads will penetrate to the bottom image and all writes will be applied to the new image. In addition, the new mirror image can be backed up to the bottom mirror image, and the bottom mirror image can be pulled to the new mirror image. In the present application, the image of the virtual machine in the subsequent embodiments of fig. 6-12 may be qcow 2.

Referring to fig. 3, fig. 3 is a flowchart illustrating a method for virtual machine live migration according to an embodiment of the present application, where the method includes the following steps:

step S310, the image of the virtual machine in the source host 110 is backed up to the distributed storage system 130.

It should be understood that the source host 110 may also be referred to as a source end and also as a source physical machine, and correspondingly, the target host 120 may also be referred to as a target end and also as a target source host, which is not limited in this application.

This step S310 may include a step S311 of backing up all the mirrored data of the virtual machine to the distributed storage system 130 when the virtual machine is initially created in the source host 110.

In step S311, when the virtual machine and the image of the virtual machine are initially created in the source host 110, one or more first devices may be further created in the source host 110, and the first devices may be mapped to the image of the virtual machine one by one, where the first devices may be used to send backup data (including all image data and incremental data) of the image of the virtual machine in the source host 110 to the distributed storage system 130.

The management apparatus for virtual machine live migration may configure a mapping relationship between an initially created mirror image of a virtual machine and a first device, and issue the configured mapping relationship to the source host 110 in a manner of issuing an instruction, where the source host 110 configures the mapping between the mirror image of the virtual machine and the first device according to the received instruction.

In addition, it is understood that, of course, the mapping relationship between the image and the first device may also be configured by the source host 110 itself, and then the source host 110 uploads the mapping relationship between the image and the first device configured for the virtual machine to the management apparatus for virtual machine live migration, so that the live migration management apparatus for the virtual machine can control the subsequent process of live migration.

In addition, target devices having a one-to-one mapping relationship with the first device may also be provided in the distributed storage system 130, that is, a first device and a target device are associated. The target device may store an index path of the mirror image data backed up by the first device, so as to quickly search for the relevant mirror image data. It is understood that the configuration process of the first mapping relationship between the first device and the target device (the first mapping relationship may be a one-to-one mapping) is similar to the process of the mapping relationship between the mirror image and the first device, and the description is not repeated here.

In addition, the management apparatus for virtual machine live migration may also control the mirror image of the virtual machine to perform full backup, so as to backup all mirror image data of the mirror image of the virtual machine to the distributed storage system 130, which includes the following specific processes: all image data of the image of the virtual machine is backed up to the corresponding target device in the distributed storage system 130 through the first mapping relationship between the first device and the target device.

It should also be understood that the specific device types of the first device and the target device may be set according to actual requirements. For example, the first device may be a first NBD and the target device may be an RBD, as implementations of the application are not limited in this respect.

The step S310 may further include a step S312 of periodically recording incremental data of the image of the virtual machine after the virtual machine is created, and backing up the incremental data to the distributed storage system 130.

Optionally, a bitmap is set in the mirror image of the virtual machine in the source host 110, and the bitmap can record difference data between the mirror image at the current time and the mirror image after the last backup, where the difference data includes added data and deleted data.

It should also be understood that, although the embodiment of the present application exemplifies a recording manner of the incremental data, a person skilled in the art should understand that the incremental data of the virtual machine may also be recorded in other manners as long as it is ensured that the incremental data can be determined, and the implementation of the present application is not limited to this.

In step S312, after determining difference data between the mirror image of the current backup and the mirror image after the last backup of the virtual machine, the determined incremental data may be backed up to the corresponding block device in the distributed storage system 130 through the first mapping relationship between the first device and the target device, so that the incremental backup of the mirror image of the virtual machine may be implemented. Further, it should be understood that for the same image, the target device storing all the image data of the full backup and the target device storing the incremental data of the incremental backup are the same target device.

In addition, the backup process of the incremental data described above may also be understood as a process of updating the full amount of mirrored data in the distributed storage system 130. Correspondingly, the process of subsequently copying the backup image from the distributed storage system 130 may also be understood as the process of copying the full amount of mirrored data in the distributed storage system 130.

Optionally, in a case where the predetermined time period is a preset cycle time, the incremental data recording the image of the virtual machine may be periodically acquired, and the incremental data may be periodically backed up to the distributed storage system 130.

For example, in the case of a cycle time of 5 minutes, incremental data of the mirror image after the virtual machine is created for 5 minutes is recorded by a bitmap, and when the cycle time is 5 minutes after the virtual machine is created, the incremental data is backed up to the RBD in the distributed storage system 130 by the first mapping relationship between the first NBD and the RBD, and the above steps are repeated subsequently, and the description is not repeated here.

It should also be understood that, although the above is made for the cycle time, a person skilled in the art may set the cycle time according to actual needs as long as it is ensured that the incremental data can be uploaded, and the embodiment of the present application is not limited thereto.

Step S320, migrating the virtual machine from the source host 110 to the target host 120 when the virtual machine meets the preset condition of live migration, where a target image of the migrated virtual machine in the target host 120 includes a backup image copied from the distributed storage system 130, and the backup image is image backup data stored in a block device in the distributed storage system 130.

It should be understood that the preset condition of the above-mentioned live migration may be that the source host 110 fails, that the read/write performance of the virtual machine is drastically reduced due to the failure of the RAID card of the source host 110, and that the load of the source host 110 reaches a bottleneck, which is not limited in this application.

Referring to fig. 4, fig. 4 illustrates a specific step of step S320 in a method for live migration of a virtual machine. This step S320 may include a step S321 of disconnecting the first mapping relationship between the first device and the target device and establishing a second mapping relationship between the target device and the second device in a case where the second device is created in the target host 120.

In step S321, the management apparatus for virtual machine live migration determines whether the virtual machine in the source host 110 satisfies the preset condition for live migration, and if the management apparatus for virtual machine live migration determines that the virtual machine satisfies the preset condition for live migration, the management apparatus for virtual machine live migration may migrate at least one virtual machine in the source host 110 according to the running condition of the virtual machine.

In addition, the management apparatus for virtual machine live migration may further send, to the target host 120, an instruction including a second mapping relationship between the second device and the target device according to the determined information of the at least one designated virtual machine. The target host 120 creates a second device having a second mapping relationship with the target device in the target host 120 according to the above instruction. In addition, the second device may be mapped one-to-one with the target device.

In addition, after the creation of the second device in the target host 120 is completed, the target host 120 may feed back the creation completion to the thermal migration management apparatus of the virtual machine. After determining that the second device is created, the thermal migration management apparatus of the virtual machine may send an instruction to disconnect or unbind the first mapping relationship between the first device and the target device to the source host 110, so that the source host 110 disconnects the mapping between the first device and the target device according to the instruction.

In addition, the live migration management apparatus of the virtual machine may further send an instruction for establishing a second mapping relationship or binding between the second device and the target device to the target host 120, so that the target host 120 establishes a mapping association between the second device and the target device according to the instruction.

In addition, it is understood that the setting of the second device may also be set by the target host 120 itself, and then the source host 110 uploads the condition of the second device to the management apparatus of the virtual machine live migration, so that the live migration management apparatus of the virtual machine can control the process of the subsequent live migration.

This step S320 may include a step S322 of creating an empty target image on the target host 120, where a mapping relationship exists between the empty target image and the second device.

In step S322, the live migration management apparatus of the virtual machine creates an empty target image on the target host 120, where the empty target image is an empty target image and has the same specification as the image of the source virtual machine, where the empty target image is used to receive the backup image copied from the distributed storage system 130 and the incremental data copied from the source host 110 and not uploaded to the distributed storage system 130 or the incremental data to be migrated.

Furthermore, the target image may have a one-to-one correspondence relationship with the second device, and may also have a one-to-one correspondence relationship with the image of the virtual machine designated for migration in the source host 110 (i.e., the structure of the virtual machine designated for migration in the source host 110 is the same as the structure of the target image of the virtual machine in the target host 120).

In addition, the image of the virtual machine in the source host 110 and the target image of the virtual machine in the target host 120 have the same specification. For example, the size of the image of the virtual machine in source host 110 may coincide with the target image size of the virtual machine in target host 120. For another example, the name of the image of the virtual machine in the source host 110 may be the same as the name of the target image of the virtual machine in the target host 120, which is not limited in this application.

This step S320 may further include a step S323 of migrating the incremental data of the image of the virtual machine on the source host 110 and the memory of the virtual machine into the target host 120.

In step S323, optionally, when the time point of the disconnection of the mapping between the first device and the target device corresponding to the virtual machine is the time point when the image of the virtual machine has just uploaded the incremental data (i.e., the image of the virtual machine does not have the incremental data), the memory of the virtual machine may be migrated to the target host 120 through the network 140.

Optionally, when the mapping between the first device and the target device corresponding to the virtual machine is disconnected at a time point that the image of the virtual machine in the source host 110 has incremental data, the thermal migration management apparatus of the virtual machine controls the source host 110 to migrate the incremental data of the image of the virtual machine to the corresponding target image in the target host 120 through the network 140. Subsequently, after the incremental data migration of the mirror image of the virtual machine is completed, the live migration management device of the virtual machine controls the source host 110 to migrate the memory of the virtual machine to the target host 120.

The step S320 may further include a step S324 of migrating the backup image stored in the target device in the distributed storage system 130 to the target host 120.

In step S320, in a case that the target device in the distributed storage system 130 and the second device in the target host 120 have a second mapping relationship (the second mapping relationship may be a one-to-one mapping), the backup image stored in the target device in the distributed storage system 130 may be migrated to the target image in the target host 120 through the second device, so that the virtual machine can be restored before live migration.

According to the method provided by the embodiment of the application, by backing up the image of the virtual machine in the source host 110 to the distributed storage system 130 and subsequently migrating the backup image in the distributed storage system 130 to the target image in the target host 120, compared with a scheme that the data volume of the first-time copied image is larger when the image is copied from the source host 110 in the existing live migration scheme, since most or all data of the image of the virtual machine on the target host 120 in the application comes from the backup image in the distributed storage system 130, the data volume of the image directly transmitted from the source host 110 to the target host 120 is reduced, and thus the time consumption of the live migration process is reduced.

In addition, for most of the mirror image data copied for the first time in the conventional live migration process, in the scenario of the present application, reading from the source host 110 is not required, and the read-write operation of the source host 110 is reduced.

In addition, since most of the mirrored data is copied from the distributed storage system 130, and only a small part of the mirrored data is copied from the source host 110 locally, compared with the existing scheme in which all the mirrored data is migrated from the source host 110, the method can also quickly complete the live migration process of the virtual machine on the basis of reducing the total time consumption of the live migration. In addition, resources occupied by the virtual machine migration on source host 110 may also be released more quickly.

In order to facilitate understanding of the technical solution of the present application, the technical solution of the present application is described below by taking the first device as the first NBD, the target device as the RBD, the second device as the second NBD, and the distributed storage system 130 as the ceph distributed file system 131 as an example. In addition, it should be understood that other first devices, target devices, second devices, and other distributed storage systems are also applicable, and the embodiments of the present application are not limited thereto.

1. When the virtual machine is created, the mirror image of the virtual machine is subjected to one-time full backup.

A) As shown in fig. 6, an RBD is created in ceph distributed file system 131, which has a one-to-one mapping relationship with a first NBD in source host 110.

The first NBD is connected to the RBD through the network 140, then the virtual machine process qemu-kvm writes data into the first NBD, and the first NBD stores the written data into the RBD. In addition, when the virtual machine corresponding to the virtual machine process qemu-kvm reads data from the first NBD, the RBD sends the corresponding data to the first NBD. In addition, qemu-kvm can also perform read and write operations (r/w) on the image.

B) As shown in fig. 7, based on the block-backup mechanism, all mirrored data of the mirror in the source host 110 is stored in the RBDs in the ceph distributed file system 131 through the first mapping relationship between the first NBD and the RBDs.

2. After the virtual machine is created, the image of the virtual machine is periodically subjected to incremental backup.

As shown in fig. 8, the management apparatus for virtual machine live migration drives the virtual machine process qemu-kvm to perform backup increment every 5 minutes. In the backup increment, a bitmap is set in the mirror image, the bitmap is a disk state bitmap created in the incremental backup process, and records updated mirror image data between two incremental backup operations, wherein the mirror image data can be a disk mirror block.

In addition, based on the block-backup mechanism, incremental data of the image of the virtual machine in the source host 110 is stored in the corresponding RBD in the ceph distributed file system 131 through the first mapping relationship between the first NBD and the RBD.

3. Thermomigration process

A) And in the case that the virtual machine in the source host 110 meets the preset live migration condition, disconnecting the mapping between the RBD in the ceph distributed file system 131 and the first NBD in the source host 110, i.e., executing unmap operation.

In addition, a mapping is established between the RBD in ceph distributed file system 131 and the second NBD in target host 120, i.e., a map operation is performed.

B) As shown in fig. 10, the management apparatus for virtual machine live migration creates a target image in the target host 120, and creates the target image on the target host 120 with the second NBD as the bottom image, and the newly created empty target image has the same specification as the image in the source host 110, wherein the specification includes name, size, and the like, and the bottom image is a part of the image in the target image. Wherein, the target image is used for receiving the image data sent by the source host 110 in the disk migration process of step C).

Furthermore, it should be understood that, in fig. 10, in order to distinguish the mirror in the source host 110 from the mirror in the target host 120, the mirror in the source host 110 is represented as a mirror, and the mirror in the target host 120 is represented as a target mirror, and specifications such as names and sizes of the two mirrors are the same, and this is only for distinguishing the description.

C) As shown in fig. 11, the incremental data to be migrated (which is not yet stored in the RBD) recorded in the bitmap used by the virtual machine process qemu-kvm through incremental backup, the incremental data to be migrated marked by the bitmap, or the data of the disk mirror block corresponding to the incremental data to be migrated. After the migration of the incremental data to be migrated is completed, the memory data of the virtual machine in the source host 110 is migrated to the target host 120.

D) As shown in fig. 12, on the target host 120, the backup image in the corresponding RBD in the ceph distributed file system 131 is merged into the target image through the second mapping relationship between the RBD and the second NBD, so that the image structure of the virtual machine is restored before live migration.

E) And continuing to perform incremental backup on the image of the virtual machine in the target host machine 120.

It should be understood that the above method of virtual machine live migration is merely exemplary, and those skilled in the art may make various modifications according to the above method.

The method for virtual machine live migration according to the embodiment of the present application is described in the foregoing with reference to fig. 2 to 12, and the management apparatus for virtual machine live migration according to the embodiment of the present application is described below with reference to fig. 13.

Referring to fig. 13, fig. 13 illustrates a management apparatus 1300 for virtual machine live migration according to an embodiment of the present application, and it should be understood that the management apparatus 1300 corresponds to the method embodiments of fig. 3 to fig. 5, and is capable of performing the steps related to the method embodiments, and specific functions of the management apparatus 1300 may be referred to the description above, and detailed descriptions are appropriately omitted herein to avoid redundancy. Specifically, the management apparatus 1300 is applied to a virtual machine system including a source host 110 of a virtual machine, a distributed storage system 130, and a target host 120 of the virtual machine, and the management apparatus 1300 includes: a backup module 1301, configured to backup an image of a virtual machine in the source host 110 to the distributed storage system 130; a migration module 1302, configured to migrate the virtual machine from the source host 110 to the target host 120 when the virtual machine meets a preset condition of live migration, where a target image of the migrated virtual machine in the target host 120 includes a backup image copied from the distributed storage system 130.

In one possible implementation, the backup module 1301 includes: a backup unit (not shown) for backing up all image data of the virtual machine into the distributed storage system 130 when the virtual machine is initially created in the source host 110; and a recording backup unit (not shown) configured to, after the virtual machine is created, periodically record incremental data of an image of the virtual machine, where the incremental data of the image of the virtual machine is difference data between a current image and an image backed up last time, and periodically backup the incremental data into the distributed storage system 130.

In one possible embodiment, a first NBD is created in the source host 110, an RBD is created in the distributed storage system 130, and the first NBD has a first mapping relationship with the RBD; the backup unit is further configured to backup all mirrored data to the RBD in the distributed storage system 130 through the first NBD; and the recording backup unit is also used for backing up the incremental data to the RBD in the distributed storage system 130 through the first NBD.

In one possible implementation, the migration module 1302 includes: a determining unit (not shown) configured to determine incremental data to be migrated of the mirror image of the virtual machine, where the incremental data to be migrated is difference data between a current mirror image and a mirror image that is backed up last time, and the incremental data to be migrated is mirror image data that is not uploaded to the distributed storage system 130; a migration unit (not shown) configured to migrate the incremental data to be migrated and the backup image in the distributed storage system 130 to the target image in the target host 120.

In one possible embodiment, an RBD is created in the distributed storage system 130, a second NBD is created in the target host 120, and the second NBD has a second mapping relationship with the RBD; the migration unit is further configured to migrate the incremental data to be migrated to the target image in the target host 120 through the network 140, and migrate the backup image in the distributed storage system 130 to the target image in the target host 120 through the second NBD and the RBD.

In the embodiment of the present application, by backing up the image of the virtual machine in the source host 110 to the distributed storage system 130, and subsequently migrating the backup image in the distributed storage system 130 to the target image in the target host 120, compared with a scheme that the data amount of the first-time copied image is larger when the image is copied from the source host 110 in the existing live migration scheme, since most or all data of the image of the virtual machine on the target host 120 in the present application comes from the backup image in the distributed storage system 130, the data amount of the image directly transmitted from the source host 110 to the target host 120 is reduced, thereby reducing the time consumption of the live migration process.

The embodiment of the present application further provides a system for implementing virtual machine live migration, where the system includes a virtual machine source host 110, a distributed storage system 130, a virtual machine target host 120, and any one of the foregoing virtual machine live migration management devices; a source host 110, configured to backup the image of the virtual machine in the distributed storage system 130 according to the backup instruction sent by the management apparatus; and the target host 120 is configured to create a target image of the virtual machine according to the migration instruction sent by the management device, where the target image includes a backup image copied from the distributed storage system 130.

As shown in fig. 14, an embodiment of the present application further provides an electronic device 1400, including: a processor 1401, a storage medium 1402 and a bus 1403, wherein the storage medium 1402 stores machine readable instructions executable by the processor 1401, and when the electronic device 1400 is operated, the processor 1401 and the storage medium 1402 communicate with each other via the bus, and the processor 1401 executes the machine readable instructions to perform the steps of any one of the methods in fig. 3 to 5.

The processor 1401 may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component.

In addition, the storage medium 1402 described above may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The non-volatile memory may be a Read-only memory (ROM), a programmable Read-only memory (PROM), an erasable programmable Read-only memory (erasabprom, EPROM), an electrically erasable programmable Read-only memory (EEPROM), or a flash memory. The volatile memory may be a Random Access Memory (RAM) which functions as an external cache. By way of example, and not limitation, many forms of RAM are available, such as static random access memory (StaticRAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), and direct memory bus random access memory (DRRAM). The storage medium 1402 described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

The present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the methods in fig. 3 to 5.

Specifically, the storage medium can be a general-purpose storage medium, such as a removable disk, a hard disk, or the like, and when the computer program on the storage medium is executed, the method for performing the virtual machine live migration can be executed.

For convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this application. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for virtual machine live migration, which is applied to a virtual machine system, wherein the virtual machine system comprises a source host of a virtual machine, a distributed storage system and a target host of the virtual machine, and the method comprises the following steps:

backing up an image of a virtual machine in the source host machine to the distributed storage system;

and under the condition that the virtual machine meets the preset condition of live migration, migrating the virtual machine from the source host machine to the target host machine, wherein the target image of the migrated virtual machine in the target host machine comprises a backup image copied from the distributed storage system.

2. The method of claim 1, wherein backing up the image of the virtual machine in the source host to the distributed storage system comprises:

When the virtual machine is initially created in the source host machine, backing up all mirror image data of the virtual machine to the distributed storage system;

after the virtual machine is created, periodically recording incremental data of a mirror image of the virtual machine, and periodically backing up the incremental data to the distributed storage system, wherein the incremental data of the mirror image of the virtual machine is difference data between a current mirror image and a mirror image backed up last time.

3. The method of claim 2, wherein a first Network Block Device (NBD) is created in the source host, a reliable autonomous distributed object storage block device (RBD) is created in the distributed storage system, and the first NBD has a first mapping relationship with the RBD;

wherein the backing up all the mirror image data of the virtual machine to the distributed storage system includes:

backing up the full mirrored data to the RBD in the distributed storage system through the first NBD;

the periodically backing up the incremental data to the distributed storage system includes,

backing up the incremental data to the RBD in the distributed storage system through the first NBD.

4. The method of claim 2 or 3, wherein the periodically recording incremental data of the image of the virtual machine comprises:

and periodically recording incremental data of the image of the virtual machine through a bitmap in the image of the virtual machine.

5. The method of claim 1, wherein migrating the virtual machine from the source host to the target host comprises:

determining incremental data to be migrated of the mirror image of the virtual machine, wherein the incremental data to be migrated is difference data between a current mirror image and a mirror image which is backed up last time, and the incremental data to be migrated is mirror image data which is not uploaded to the distributed storage system;

and migrating the incremental data to be migrated and the backup image in the distributed storage system to a target image in the target host machine.

6. The method of claim 5, wherein an RBD is created in the distributed storage system, wherein a second Network Block Device (NBD) is created in the target host, and wherein the second NBD has a second mapping relationship with the RBD;

wherein the migrating the incremental data to be migrated and the backup image in the distributed storage system to the target image in the target host comprises:

Migrating the incremental data to be migrated to a target image in the target host machine through a network, and migrating a backup image in the distributed storage system to the target image in the target host machine through the second NBD and the RBD.

7. A management device for virtual machine live migration is applied to a virtual machine system, wherein the virtual machine system comprises a source host machine of a virtual machine, a distributed storage system and a target host machine of the virtual machine, and the management device comprises:

the backup module is used for backing up the mirror image of the virtual machine in the source host machine to the distributed storage system;

and the migration module is used for migrating the virtual machine from the source host machine to the target host machine under the condition that the virtual machine meets the preset condition of live migration, wherein the target image of the migrated virtual machine in the target host machine comprises a backup image copied from the distributed storage system.

8. A system for implementing virtual machine live migration, comprising: a source host, a target host, a distributed storage system, and the virtual machine live migration management apparatus of claim 7;

The source host is used for backing up the mirror image of the virtual machine to the distributed storage system according to the backup instruction sent by the management device;

the target host is configured to create a target image of the virtual machine according to the migration instruction sent by the management device, where the target image includes a backup image copied from the distributed storage system.

9. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the method according to any one of claims 1 to 6.