CN101609419B - Continuous on-line transferring data backup method of virtual machine and device thereof - Google Patents
Continuous on-line transferring data backup method of virtual machine and device thereof Download PDFInfo
- Publication number
- CN101609419B CN101609419B CN2009100880161A CN200910088016A CN101609419B CN 101609419 B CN101609419 B CN 101609419B CN 2009100880161 A CN2009100880161 A CN 2009100880161A CN 200910088016 A CN200910088016 A CN 200910088016A CN 101609419 B CN101609419 B CN 101609419B
- Authority
- CN
- China
- Prior art keywords
- state
- image file
- host
- data block
- disk image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a continuous on-line transferring data backup method of a virtual machine and a device thereof. The method comprises the steps of updating the virtual machine memory data in a primary host machine and the data block in a virtual disk image file in a network storage according to the running of an embodiment of a primary virtual machine, and sending the updated virtual machine memory data to a backup host machine; temporarily stopping running the embodiment of the primary virtual machine, locking the state of the data block in the virtual disk image file through a locking operation to obtain the light-weight snapshot of the virtual disk image file, and recovering the running of the embodiment of the primary virtual machine, wherein the virtual disk image file is used for the backup host machine to perform a rollback operation. Through the above processing, the light-weight snapshot of the embodiment of the invention only occupies a little memory space. In addition, a technical scheme of the embodiment of the invention does not need to transmit lots of data, and is capable of quickly establishing the light-weight snapshot and improving the performance of the virtual machine.
Description
Technical field
The present invention relates to the online migrating technology of virtual machine field, relate in particular to a kind of data back up method of virtual machine consistently online migration, main host, backup host and virtual machine consistently online migratory system.
Background technology
In the prior art, virtual machine is on computer operating system, to make up a virtualization layer, can good encapsulation property, isolation, the transparency and manageability be provided for application program.The migration of virtual machine is meant that operating operation system example can shift between different physical hosts, and in transition process operating system and on the application of operation can continuous service with service and be not interrupted.The details of numerous application programs that operating personnel need not to move in the relational operation system is very easy to bookkeeping.At present, virtual machine (vm) migration is divided into two kinds, is respectively general migration and online migration, and wherein, online migration is meant under the prerequisite of the application program operation that encapsulates in the maintenance virtual machine, simultaneously system is moved the hardware platform of switching virtual machine operation.Through the online migrate attribute of virtual machine, can reduce of the influence of bottom hardware fault to virtual environment.This makes virtual machine progressively become the important tool that makes up high-availability system.
At present, the high-availability system based on virtual machine technique mainly comprises following several kinds:
(1) consistently online based on diskcopy moves: this mode has further expanded the mode of online migration; Consider when fault takes place; Need the state of virtual machine be rolled back to a consistance version a little earlier, and network memory (Network Attached Storage; Hereinafter to be referred as: the virtual disk image file NAS) can't be with the internal storage state rollback of virtual machine, and this mode has proposed a kind of lasting migration scheme based on diskcopy.In this scheme, main host has independently virtual disk image file separately with the backup host; When virtual machine when the virtual disk image file writes data; The disk driver of virtual machine is when writing data virtual disk image file itself; With data through Network Synchronization be sent to the backup host; And after once online migration is accomplished; The Disk State increments of change that is produced during with this online migration by the backup host writes the virtual disk image file of backup host, thereby guarantees when hardware fault taking place cause the virtual machine rollback, and the virutal machine memory state and the state of virtual disk image file that back up on the host satisfy consistance.But because the virtual disk image file is long-pending bigger, to the having relatively high expectations of the network bandwidth, and the transmission time is longer during transmission.
(2) consistently online based on snapshot (Snapshot) moves: existing snapshot mechanism can be created one or more snapshots in the virtual disk image file of virtual machine.At a time for after the virtual disk image file establishment snapshot, the Disk State (that is, the state of data block in the virtual disk image file) before this moment will be retained, and follow-up disk write operation can not cause losing of this moment Disk State before.And the virtual machine disk is returned to a certain snapshot, then expression uses the Disk State of preserving in this historical snapshot to replace existing Disk State, thereby data in magnetic disk is rolled back to the historical juncture of snapshot representative.This scheme has promptly been utilized the snapshot mechanism of virtual disk, after accomplishing an online migration of virtual machine, for virtual disk is created a snapshot; When main host breaks down when causing backing up host rollback virtual machine state, virtual disk is synchronously rolled back to the corresponding state of corresponding snapshot.Because consistently online migration requires in the time interval of short (several seconds or tens of millisecond) continuously state transition with virtual machine to backup virtual machine; This just requires can create in the short period of time, delete snapshot; And when breaking down; In the time of need on the backup host, starting backup virtual machine, need in the short period of time disk to be reverted to snapshot state a little earlier.Yet; Because existing snapping technique is owing to need to follow the tracks of the number of times of quoting of each block; The time of therefore establishment, deletion and recovery snapshot is longer, and, create the fast volume that causes the virtual disk image file of presenting a note to constantly and continue increase; For the long playing computer system of needs, its space expense is difficult to meet the demands.
(3) consistently online based on incident moves: in this scheme; If trigger backup in the time of can taking place at accessing operation to virtual machine state to disk; And after backup is accomplished, carry out accessing operation again, then can guarantee when virtual machine breaks down rollback identical disk operating meeting playback in an identical manner disk; And obtain consistent result, thereby solve the consistency problem in the consistently online transition process.The problem of this scheme is the accessing operation of disk very frequent, and therefore, this mode can trigger the backup operation of virtual machine with higher frequency, and causes normal execution of virtual machine interruption to wait for the backup completion, and is therefore bigger to the virtual machine Effect on Performance.
As stated, the consistently online migrating technology is to utilize virtual machine to make up one of core technology of high-availability system, and guarantees virtual machine to break down and after the rollback, the consistance of virtual machine disk and internal storage state then is the basis of realizing that consistently online moves.At present, keep the conforming technology of virtual machine Disk State not perfect, still have problems: 1, need the virtual disk data of transmission of large capacity based on the consistently online migration mechanism of diskcopy, bigger to network bandwidth influence, and transmission speed is slower; 2, need consume great amount of time and storage space when establishment, deletion, the rollback disk snapshot based on the consistently online migration mechanism of snapshot; 3, the consistently online migration mechanism based on incident can block the visit of virtual machine to disk unit, and is bigger to the performance impact of virtual machine.
Summary of the invention
The embodiment of the invention provides a kind of data back up method and device of virtual machine consistently online migration; In order to solve the defective that disk snapshot in the prior art need take a large amount of storage spaces, be implemented in and use the lightweight snapshot that takies less storage space in the virtual machine consistently online transition process.
The embodiment of the invention provides a kind of data back up method of virtual machine consistently online migration; Comprise: the operation according to the host virtual machine instance is upgraded operation to virutal machine memory data on the main host and the data block in the virtual disk image file in the network memory, and the virutal machine memory data after will upgrading send to the backup host; Operation suspension host virtual machine instance; Lock through the state of lock operation data block in the virtual disk image file; Obtain the lightweight snapshot of virtual disk image file; And the host virtual machine instance that resumes operation; Wherein, The virtual disk image file is used to back up host and carries out rolling back action, and it is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled, said lightweight to impinge upon soon to preserve current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file.
The embodiment of the invention provides a kind of main host; Comprise host virtual machine, main host monitor of virtual machine and main host operating system and hardware; Wherein said main host monitor of virtual machine comprises that main host migration controller and main host lightweight snapshot drive said main host operating system and hardware bundle includes network storage drive; Said network memory drives and is used for according to the operation of host virtual machine instance the data block in the virtual disk image file of virutal machine memory data and network memory being upgraded operation; Said main host migration controller is used for the virutal machine memory data after upgrading are sent to the backup host; Said main host lightweight snapshot drives and is used for operation suspension host virtual machine instance; Lock through the state of lock operation data block in the virtual disk image file; Obtain the lightweight snapshot of virtual disk image file; And the host virtual machine instance that resumes operation; Wherein, The virtual disk image file is used to back up host and carries out rolling back action, and it is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled, said lightweight to impinge upon soon to preserve current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file; Said backup virtual machine is used for according to virtual disk image file after the rollback and virutal machine memory data run backup virtual machine instance.
The embodiment of the invention provides a kind of backup host; Comprise backup virtual machine, backup host monitor of virtual machine and backup host operating system and hardware; Wherein said backup host monitor of virtual machine comprises that backup host migration controller and backup host lightweight snapshot drive said backup host operating system and hardware bundle includes network storage drive; Said backup host migration controller is used to preserve the virutal machine memory data that main host sends; Said backup host lightweight snapshot drives and is used under the situation that the host virtual machine instance can not continue to move; The state of the virutal machine memory data consistent that the virtual disk image file is returned to and preserves through rolling back action according to the lightweight snapshot; It is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled; Said lightweight impinges upon soon and preserves current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file; Said backup virtual machine is used for according to virtual disk image file after the rollback and virutal machine memory data run backup virtual machine instance.
The embodiment of the invention provides a kind of virtual machine consistently online migratory system, comprises main host and the backup host in the foregoing description in network memory, the foregoing description.
The data back up method and the device of the virtual machine consistently online migration of the embodiment of the invention; State through data block in the lock operation locking virtual disk image file; Obtain the lightweight snapshot of virtual disk image file,, can avoid the waste of storage space because this lightweight snapshot only need take very little storage space; In addition; According to the technical scheme of the embodiment of the invention a large amount of virtual disk image file data of needs transmission not, and can create the lightweight snapshot fast, greatly improve the performance of virtual machine.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art; To do one to the accompanying drawing of required use in embodiment or the description of the Prior Art below introduces simply; Obviously; Accompanying drawing in describing below is some embodiments of the present invention, for those of ordinary skills,
Do not pay under the prerequisite of creative work property, can also obtain other accompanying drawing according to these accompanying drawings.
Embodiment
Fig. 1 is the process flow diagram of data back up method of the virtual machine consistently online migration of the embodiment of the invention;
Fig. 2 is the structural representation that the virtual disk of the embodiment of the invention is asked image file;
Fig. 3 is the synoptic diagram of the state exchange of the embodiment of the invention;
Fig. 4 is two signaling process figure that confirm operation of the embodiment of the invention;
Fig. 5 is the process flow diagram that the main host of the embodiment of the invention carries out the detailed process of consistently online migration;
Fig. 6 is the process flow diagram that the backup host of the embodiment of the invention carries out the detailed process of consistently online migration;
Fig. 7 is the structural representation of the main host of the embodiment of the invention;
Fig. 8 is the structural representation of the backup host of the embodiment of the invention;
Fig. 9 is the structural representation of the virtual machine consistently online migratory system of the embodiment of the invention.
Further specify the technical scheme of the embodiment of the invention below in conjunction with accompanying drawing and specific embodiment.
According to embodiments of the invention; A kind of data back up method of virtual machine consistently online migration is provided; Fig. 1 is the process flow diagram of data back up method of the virtual machine consistently online migration of the embodiment of the invention; As shown in Figure 1, may further comprise the steps according to the data back up method of the virtual machine consistently online of embodiment of the invention migration:
Step 101 according to the operation of host virtual machine instance virutal machine memory data on the main host and the data block in the virtual disk image file in the network memory are upgraded operation, and the virutal machine memory data after will upgrading sends to the backup host;
Particularly, in the process of host virtual machine instance operation, main host can upgrade operation to the data block of virutal machine memory data in the main host internal memory and the virtual disk image file in the network memory according to the operation of host virtual machine instance; For fear of when unexpected hardware fault takes place; Virtual machine can not be moved to non-fault machine in time and continued operation, and host virtual machine can be initiated online migration to the backup host constantly, for example; Host virtual machine is every initiates once online migration at a distance from the schedule time to the backup host; After initiating online migration, the virutal machine memory data after upgrading are sent to the backup host, can be in the process of operation to guarantee virtual machine; Running status with self is saved in the backup host continuously, makes the backup host preserve virtual machine state a little earlier.
Need to prove that in the virtual disk image file, the data of disk are unit composition data piece with specific size; Virtual machine can be reduced to the visit to the data block that comprises this address date to the visit of the data that comprise in a certain physical address of disk.In addition, include the mapping table of numbering virtual disk image file offset address by data block in the virtual disk image file.In embodiments of the present invention; As shown in Figure 2; Except preserving metadata and data block, the lightweight snapshot can also be preserved two mapping tables in the virtual disk image file in the virtual disk image file: current mapping table and snapshot mapping table, wherein; Current mapping table is used to identify disk under current state and snapshot state, the position of data block in the virtual disk image file; The snapshot mapping table is used to identify disk under snapshot state, the position of data block in the virtual disk image file.
In embodiments of the present invention, also define the state of data block, comprised following four kinds of states:
Unallocated state: be not the data block allocations storage space as yet in the virtual disk image file;
Distributed read-write state: be data block memory allocated space in the virtual disk image file, and allowed this data block is carried out read-write operation;
Distributed a read states: be data block memory allocated space in the virtual disk image file, but forbidden writing data to this storage space;
Distributed state to be recycled; Be data block memory allocated space in the virtual disk image file, but this data block is updated, when creating the lightweight snapshot, this data block will be recovered next time.
In the lightweight snapshot, any one data block can only be in the wherein a kind of state in above-mentioned four kinds of states.For the data block that is in unallocated state; Can in current mapping table, use and corresponding virtual disk image file offset address is designated unappropriated mode the state of data block is identified; For example, with corresponding virtual disk image file offset address sign " 0 "; For being in the data block of distributing read-write state, distributing a read states, perhaps having distributed state to be recycled, can use bitmap that the state of this data block is identified.
In step 101, the renewal operation that the data block in the right virtual disk image file of main host is carried out specifically comprises:
1, read operation: according to data block numbering read block data;
2, write operation: write data to given data block numbering;
3, batch operation: according to data block numbering distribute data piece in the virtual disk image file;
4, releasing operation: discharge given data block occupation space in the virtual disk image file.
Particularly, when carrying out read operation, handle below needing to carry out: 1, the data block of data block is numbered and be mapped as virtual disk image file offset address according to current mapping table; 2, the state of confirming this data block be in distribute read-write state, distribute a read states, or distributed under the situation of state to be recycled, read the data block of designated length and return according to virtual disk image file offset address; 3, be under the situation of unallocated state at the specified data bulk state, do not operate, directly return.
When carrying out write operation, handle below needing to carry out: 1, the data block of data block is numbered and be mapped as virtual disk image file offset address according to current mapping table; 2, be under the situation of unallocated state at the state of confirming this data block, carry out batch operation, write data according to virtual disk image file offset address then and return; 3, be under the situation of distributing read-write state at the state of confirming this data block, directly write data and return according to virtual disk image file offset address; 4, be under the situation of distributing a read states at the state of confirming this data block; New data block is carried out batch operation; With the status modifier of former data block for distributing state to be recycled, 5, the virtual disk image file offset address that distributes according to batch operation writes data and return.
As above, when carrying out write operation, might need to carry out batch operation, carrying out batch operation needs following the processing: confirm that 1, minimum unallocated continuation address is an address space to be allocated; 2, with the status modifier of the data block in the address space to be allocated for distributing read-write state; 3, obtain the virtual disk image file offset address of address space to be allocated; 4, use the virtual disk image file offset address of address space to be allocated to upgrade current mapping table; 5, the virtual disk image file offset address that returns address space to be allocated is given write operation.
When carrying out releasing operation, handle below needing to carry out: the status modifier of the data block that 1, will discharge is unallocated; The station location marker of the data block that 2, will discharge in current mapping table is unallocated, and for example, the station location marker of data block in current mapping table that needs are discharged is " 0 ".
By on can know; In step 101; Main host also can upgrade the current mapping table of data block numbering under the preservation current state with virtual disk image file offset address corresponding relation when the data block in the virtual disk image file in the network memory being upgraded operation synchronously.
Step 102; Operation suspension host virtual machine instance; Lock through the state of lock operation, obtain the lightweight snapshot of virtual disk image file data block in the virtual disk image file, and the host virtual machine instance that resumes operation; Wherein, the virtual disk image file is used to back up host and carries out rolling back action.
Particularly; When the online migration of main host reaches a predetermined critical point; Main host needs operation suspension host virtual machine instance, locks through the state of lock operation to data block in the virtual disk image file, obtains the lightweight snapshot of this virtual disk image file; Wherein, above-mentioned virtual disk image file is used to back up host and carries out rolling back action; After obtaining the lightweight snapshot, main host just can recover the operation of host virtual machine instance.
State through data block in the lock operation locking virtual disk image file; Obtain the lightweight snapshot of virtual disk image file,, can avoid the waste of storage space because this lightweight snapshot only need take very little storage space; In addition; According to the technical scheme of the embodiment of the invention a large amount of virtual disk image file data of needs transmission not, and can create the lightweight snapshot fast, greatly improve the performance of virtual machine.
As stated; Lock operation can be created the lightweight snapshot for the virtual disk image file; When carrying out lock operation, comprise following processing: 1, for being in the data block of distributing read-write state in the current mapping table, with the status modifier of this data block for distributing a read states; 2, for being in the data block of distributing state to be recycled in the current mapping table, this data block is carried out releasing operation; 3, use current mapping table to cover the snapshot mapping table, wherein, above-mentioned snapshot mapping table is used to represent that disk data block under snapshot state is numbered and the corresponding relation of virtual disk image file offset address.
In addition; In step 102; Behind the lightweight snapshot that obtains this virtual disk image file, main host can be kept at this lightweight snapshot in the virtual disk image file, also can the lightweight snapshot be kept in the internal memory of main host; If the lightweight snapshot is kept in the internal memory of main host, main host need send to subsequent use host with this lightweight snapshot as the part of virutal machine memory data through network.Because the speed of read/write data is much larger than the speed of read/write data from disk from internal memory; The lightweight snapshot is kept in the internal memory of main host; Can reduce read/write operation, thereby shorten the running time, can in the shorter time, carry out online migration disk.
After step 102, under the situation that the host virtual machine instance can not continue to move, the backup host returns to the virtual disk image file and back up the state of the virutal machine memory data consistent of host preservation through rolling back action according to this lightweight snapshot; Subsequently, the backup host just can be according to virtual disk image file after the rollback and virutal machine memory data run backup virtual machine instance.
Wherein, Rolling back action can return to the state when setting up the lightweight snapshot with the virtual disk image file; When the execution rolling back action comprises, comprise following processing: 1,, this data block is carried out releasing operation for being in the data block of distributing read-write state in the current mapping table; 2, for being in the data block of distributing state to be recycled in the current mapping table, with this data block status modifier for distributing a read states; 3, use the snapshot mapping table to cover current mapping table.
Below, the conversion of each operation to the data bulk state is elaborated.Fig. 3 is the synoptic diagram of the state exchange of the embodiment of the invention, and is as shown in Figure 3, if data block is in unallocated state, after the execution batch operation, the state transformation of data block is for distribute read-write state; If the data block to distributing read-write state conducts the locking operations, the state transformation of this data block is for distribute a read states; If the data block to distributing a read states is carried out write operation, then the state transformation of former data block is for distribute state to be recycled, and the state transformation of new data block is for distribute read-write state; If the data block to distributing state to be recycled is carried out lock operation and releasing operation, the state transformation of this data block is unallocated state; If carry out rolling back action to being in the data block of distributing state to be recycled, then the state exchange of this data block is for distribute a read states; If carry out rolling back action to being in the data of distributing read-write state, the state exchange of this data block is unallocated state.The data block of distributing read-write state is carried out the read/write operation attitude of will remaining stationary; The data block of distributing a read states is carried out the read operation attitude of also will remaining stationary.
Continuing based on virtual machine in the high-availability system of migration, needing to guarantee state consistency between disk internal state consistance and disk and internal memory.Wherein, Network memory internal state consistance is meant at any time; Be in conforming state between each data block that comprises in the virtual disk image file, thereby make the normal data in the reading disk and can not read and write mistake of operating system and application program of virtual disk driver and virtual machine; State consistency is meant that the data that copy in the virutal machine memory data that virtual machine had and the data block of disk from the virtual disk image file on the backup host are consistent in time between disk and internal memory; That is the final modification time of the data that, copy in final time of revising of virutal machine memory data and the data block of disk from the virtual disk image file is identical.
In embodiments of the present invention; The lightweight snapshot can be realized " lock operation " and " rolling back action " for network memory; When accomplishing once online migration; Can " lock " in the network memory state of data block in the virtual disk image file, guarantee that subsequent write operation can not influence the historic state of data block in the virtual disk image file, the consistance of promptly protecting the disk internal state; When virtual machine breaks down when needing rollback; The state of data block in the virtual disk image file in can " rollback " network memory; Abandon and the inconsistent modification of virutal machine memory data, the data block in the virtual disk image file is rolled back to one and the corresponding to version of virutal machine memory data.Can guarantee the consistance of state between disk and internal memory.
In addition, the embodiment of the invention can also further guarantee the consistance of state between disk and internal memory through two affirmation mechanisms.Main host can obtain to back up the affirmation of host to the virutal machine memory data after being saved on the backup host through online migration mechanism the virutal machine memory data; Main host can obtain disk snapshot and confirm after disk is carried out the operation of lightweight snapshot creation.Have only after main host obtains two affirmations, can recover the virtual machine operation, thereby guarantee the consistance of state between disk and internal memory.Fig. 4 is two signaling process figure that confirm operation of the embodiment of the invention, and is as shown in Figure 4, comprises following processing:
Through above-mentioned processing, can guarantee the consistance of state between disk and internal memory.Those skilled in the art can understand; When realizing the above-mentioned processing procedure of the embodiment of the invention; Main host also can receive only Memory Backup and confirm or receive only the lightweight snapshot to set up affirmation; Another is confirmed that main host can be given tacit consent to and receives, can reach the conforming purpose that guarantees state between disk and internal memory equally.
Below, from the angle of main host, processing procedure shown in Figure 1 is carried out complete explanation respectively with the backup host.Fig. 5 is the process flow diagram that the main host of the embodiment of the invention carries out the detailed process of consistently online migration, and is as shown in Figure 5, comprises following processing:
Step 504 when the online migration of main host arrives predetermined critical point, is suspended the operation of host virtual machine instance;
Through above-mentioned processing, main host can be created the lightweight snapshot for the virtual disk image file through the lock operation that the lightweight snapshot provides, and guarantees that subsequent write operation can not influence the historic state of virtual disk image file.
Fig. 6 is the process flow diagram that the backup host of the embodiment of the invention carries out the detailed process of online migration, and is as shown in Figure 6, comprises following processing:
Through above-mentioned processing; When virtual machine breaks down rollback; Can abandon virtual disk image file and the inconsistent modification of virutal machine memory state through rolling back action; The virtual disk image file is rolled back to one and the corresponding to version of internal storage state, make that virtual machine can be at backup host relaying reforwarding row.
According to another embodiment of the present invention; A kind of main host is provided; Fig. 7 is the structural representation of the main host of the embodiment of the invention, and is as shown in Figure 7, comprises according to the main host of the embodiment of the invention: update module 70, sending module 72 and locking module 74; Below, the main host according to the embodiment of the invention is carried out detailed explanation.
Particularly, in the process of host virtual machine instance operation, update module 70 is upgraded operation according to the operation of host virtual machine instance to the data block in the virtual disk image file in virutal machine memory data and the network memory.
Need to prove that in the virtual disk image file, the data of disk are unit composition data piece with specific size; Virtual machine can be reduced to the visit to the data block that comprises this address date to the visit of the data that comprise in a certain physical address of disk.In addition, as shown in Figure 2 in embodiments of the present invention, except preserving metadata and data block, the lightweight snapshot can also be preserved two mapping tables in the virtual disk image file in the virtual disk image file.
In embodiments of the present invention, also defined the state of data block, comprised following four kinds of states: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled.In the lightweight snapshot, any one data block can only be in the wherein a kind of state in above-mentioned four kinds of states.For the data block that is in unallocated state; Can in current mapping table, use and corresponding virtual disk image file offset address is designated unappropriated mode the state of data block is identified; For example, with corresponding virtual disk image file offset address sign " 0 "; For being in the data block of distributing read-write state, distributing a read states, perhaps having distributed state to be recycled, can use bitmap that the state of this data block is identified.
The renewal operation that data block in the virtual disk image file that update module is 70 pairs is carried out specifically comprises: 1, read operation, 2, write operation, 3, batch operation, 4, releasing operation.The concrete processing procedure that update module 70 is carried out aforesaid operations can be understood with reference to the related content in the foregoing description, repeats no more inferior.
In addition; Update module 70 also can be upgraded the current mapping table of data block numbering under the preservation current state with virtual disk image file offset address corresponding relation when the data block in the virtual disk image file in the network memory being upgraded operation synchronously.
For fear of when unexpected hardware fault takes place; Virtual machine can not be moved to non-fault machine in time and continued operation, and host virtual machine can be initiated online migration to the backup host constantly, for example; Host virtual machine is every initiates once online migration at a distance from the schedule time to the backup host; After initiating online migration, the virutal machine memory data after sending module 72 will upgrade send to the backup host, can be in the process of operation to guarantee virtual machine; Running status with self is saved in the backup host continuously, makes the backup host preserve virtual machine state a little earlier.
When the online migration of main host reaches a predetermined critical point; Locking module 74 needs operation suspension host virtual machine instance; Lock through the state of lock operation data block in the virtual disk image file; Obtain the lightweight snapshot of this virtual disk image file, wherein, above-mentioned virtual disk image file is used to back up host and carries out rolling back action; After obtaining the lightweight snapshot, recover the operation of host virtual machine instance.
As stated; Lock operation can be created the lightweight snapshot for the virtual disk image file; When carrying out lock operation, comprise following processing: 1, for being in the data block of distributing read-write state in the current mapping table, with the status modifier of this data block for distributing a read states; 2, for being in the data block of distributing state to be recycled in the current mapping table, this data block is carried out releasing operation; 3, use current mapping table to cover the snapshot mapping table.
After locking module 74 obtains the lightweight snapshot of this virtual disk image file; Main host can be kept at this lightweight snapshot in the virtual disk image file; Also the lightweight snapshot is kept in the internal memory of main host; If the lightweight snapshot is kept in the internal memory of main host, main host need send to subsequent use host with this lightweight snapshot as the part of virutal machine memory data through network.Because the speed of read/write data is much larger than the speed of read/write data from disk from internal memory; The lightweight snapshot is kept in the internal memory of main host; Can reduce read/write operation, thereby shorten the running time, can in the shorter time, carry out online migration disk.
Through above-mentioned processing, main host can be created the lightweight snapshot for the virtual disk image file through the lock operation that the lightweight snapshot provides, and guarantees that subsequent write operation can not influence the historic state of virtual disk image file.
According to an embodiment more of the present invention; A kind of backup host is provided; Fig. 8 is the structural representation of the backup host of the embodiment of the invention, and is as shown in Figure 8, comprises according to the backup host of the embodiment of the invention: preserve module 80, roll-back module 82, operation module 84.Below, the backup host according to the embodiment of the invention is elaborated.
Particularly, the preservation module 80 of backup host is preserved the virutal machine memory data that main host sends; Under the situation that the host virtual machine instance can not continue to move, the roll-back module 82 of backup host returns to the virtual disk image file and back up the state of the virutal machine memory data consistent of host preservation through rolling back action according to this lightweight snapshot; Subsequently, operation module 84 just can be according to virtual disk image file after the rollback and virutal machine memory data run backup virtual machine instance.
Wherein, The rolling back action of roll-back module 82 can return to the state when setting up the lightweight snapshot with the virtual disk image file; When roll-back module 82 execution rolling back actions comprise; Comprise following processing: 1,, this data block is carried out releasing operation for being in the data block of distributing read-write state in the current mapping table; 2, for being in the data block of distributing state to be recycled in the current mapping table, with this data block status modifier for distributing a read states; 3, use the snapshot mapping table to cover current mapping table.
Through above-mentioned processing; When virtual machine breaks down rollback; The backup host can abandon virtual disk image file and the inconsistent modification of virutal machine memory state through rolling back action; The virtual disk image file is rolled back to one and the corresponding to version of internal storage state, make that virtual machine can be at backup host relaying reforwarding row.
According to an embodiment more of the present invention; A kind of virtual machine consistently online migratory system is provided, and Fig. 9 is the structural representation of the virtual machine consistently online migratory system of the embodiment of the invention, and is as shown in Figure 9; Virtual machine consistently online migratory system according to the embodiment of the invention comprises: main host 1, network memory 3, backup host 5; Wherein, comprise in the main host 1: host virtual machine 10, promptly above-mentioned host virtual machine instance; All use host virtual machine 10, main host monitor of virtual machine 12, main host operating system and hardware 14 in the following description; Comprise in the main host monitor of virtual machine 12: main host migration controller 120, main host lightweight snapshot drives 122, comprises in main host operating system and the hardware 14: network memory drives 140; Comprise in the backup host 5: backup virtual machine 50; It is above-mentioned backup virtual machine instance; All use backup virtual machine 50, backup host monitor of virtual machine 52, backup host operating system and hardware 54 in the following description; Comprise in the backup host monitor of virtual machine 52: backup host migration controller 520, backup host lightweight snapshot drives 522; Comprise in backup host operating system/hardware 54: network memory drives 540; Wherein, main host monitor of virtual machine 12 is connected with backup host monitor of virtual machine 52 through network, and the backup virtual machine 50 on the host virtual machine 10 of main host 1 and the backup host 5 is shared the virtual disk image file through network memory 3.Below, the virtual machine consistently online migratory system according to the embodiment of the invention is elaborated.
Particularly, be example at first with main host 1, the structure of host is described.Comprise internal memory and disk in the host virtual machine 10, wherein, the main host migration controller 120 in internal memory and the main host monitor of virtual machine 12 links to each other; Under the control of main host migration controller 120; Can the virutal machine memory data be sent to backup host migration controller 520, the disk that main host lightweight snapshot driving 122 is connected in the host virtual machine 10, and the virtual disk image file that drives in 140 pairs of network memories 3 through network memory carries out the lightweight snapshot; Wherein, Network memory drives 140 and is used for network memory 3 is driven, and, network memory 3 is carried out read/write operation that is.
Below, main host 1 and the process that backup host 5 carries out the consistently online migration are elaborated.
At first; Main host 1 loads the virtual disk image file of host virtual machine 10 from network memory 3 through network memory driving 140; And start host virtual machine 10, subsequently, main host 1 is confirmed suitable backup host 5 according to start-up parameter or predefined strategy.
Main host migration controller 120 is initiated online migration to backup host migration controller 520; After initiating online migration; Virutal machine memory data after main host migration controller 120 will upgrade send to the backup host; When the online migration of main host 1 arrived predetermined critical point, main host 1 suspended the operation of host virtual machine 10, and the lock operation that main host lightweight snapshot driving 122 provides through the lightweight snapshot is that the virtual disk image file of host virtual machine 10 is created the lightweight snapshot; After setting up the lightweight snapshot, the operation that main host 1 recovers host virtual machine 10.
Main host 1 judges whether to arrive new backup point, if do not reach new backup point, then main host 1 is not done any operation, continues to keep the operation of host virtual machine 10; If reach new backup point, main host migration controller 120 is initiated online migration once more to backup host 5; If host virtual machine 10 breaks down, 1 operation that stops host virtual machine 10 of main host, and the state of cleaning host virtual machine 10.
When backup host 5 is waited for online migration; Backup host 5 needs to detect the health status of the host virtual machine instance on the main host 1; If host virtual machine 10 breaks down and definite host virtual machine 10 can not normally move; Backup host 5 loads the virutal machine memory data that host virtual machine 10 transmits through online migration pattern, otherwise backup host 5 continues to wait for online migration.
After backup host 5 loads the virutal machine memory data of host virtual machine 10 through online migration pattern transmission; Backup host lightweight snapshot drives the lightweight snapshot of preserving in the 522 tool loads virtual disk image files that provide through the lightweight snapshot, and the rolling back action that uses the lightweight snapshot to provide the corresponding to state of virutal machine memory data that the virtual disk image file is rolled back to and preserves.
Carrying out behind the rolling back action, backup host 5 operation backup virtual machine 50, backup host 5 is as main host 1 operation, and backup virtual machine 50 backs up host migration controller 520 subsequently and initiates consistently online migration next time as host virtual machine 10 operations.
In addition, the embodiment of the invention can also further guarantee the consistance of state between disk and internal memory through two affirmation mechanisms.Main host 1 can obtain to back up the affirmation of 5 pairs of virutal machine memory data of host after being saved on the backup host 5 through online migration mechanism the virutal machine memory data; After 1 pair of disk of main host carries out the operation of lightweight snapshot creation, can obtain disk snapshot and confirm.Have only after main host 1 obtains two affirmations, can recover the virtual machine operation, thereby guarantee the consistance of state between disk and internal memory.Its detailed process can be understood with reference to Fig. 4, repeats no more at this.
In sum, main host can be created the lightweight snapshot for the virtual disk image file through the lock operation that the lightweight snapshot provides, and guarantees that subsequent write operation can not influence the historic state of virtual disk image file; When virtual machine breaks down rollback; Can abandon virtual disk image file and the inconsistent modification of virutal machine memory state through rolling back action; The virtual disk image file is rolled back to one and the corresponding to version of internal storage state, make that virtual machine can be at backup host relaying reforwarding row.
And; Because the embodiment of the invention has been preserved current mapping table and snapshot mapping table; Make this lightweight snapshot only need take very little storage space; Can not increase the space of storage snapshot along with the increase of snapshot number of times, overcome the defective that disk snapshot in the prior art need take a large amount of storage spaces.In addition, the embodiment of the invention has guaranteed that also virtual machine can be in the process of operation, and the running status with self is saved in the backup host continuously, makes the backup host can have host virtual machine coherency state a little earlier; And, through setting up continuous storage space, and the state of data in magnetic disk piece is saved in the continuous memory block for the data block of disk, can reduce read/write operation to exterior storage, shorten the time of lock operation, rolling back action.According to the technical scheme of the embodiment of the invention a large amount of virtual disk image file data of needs transmission not, and can create the lightweight snapshot fast, greatly improve the performance of virtual machine.
Through the description of above embodiment, those skilled in the art can be well understood to each embodiment and can realize by the mode that software adds essential general hardware platform, can certainly pass through hardware.Based on such understanding; The part that technique scheme contributes to prior art in essence in other words can be come out with the embodied of software product; This computer software product can be stored in the computer-readable recording medium, like ROM/RAM, magnetic disc, CD etc., comprises that some instructions are with so that a computer equipment (can be a personal computer; Server, perhaps network equipment etc.) carry out the described method of some part of each embodiment or embodiment.
What should explain at last is: above embodiment is only in order to explaining technical scheme of the present invention, but not to its restriction; Although with reference to previous embodiment the present invention has been carried out detailed explanation, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these are revised or replacement, do not make the spirit and the scope of the essence disengaging various embodiments of the present invention technical scheme of relevant art scheme.
Claims (10)
1. the data back up method of a virtual machine consistently online migration is characterized in that, comprising:
Operation according to the host virtual machine instance is upgraded operation to virutal machine memory data on the main host and the data block in the virtual disk image file in the network memory, and the said virutal machine memory data after will upgrading send to the backup host;
The said host virtual machine instance of operation suspension; Lock through the state of lock operation data block in the said virtual disk image file; Obtain the lightweight snapshot of said virtual disk image file; And the said host virtual machine instance that resumes operation; Wherein, Said virtual disk image file is used for said backup host and carries out rolling back action, and it is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled, said lightweight to impinge upon soon to preserve current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file.
2. method according to claim 1 is characterized in that, also comprises:
Under the situation that said host virtual machine instance can not continue to move, said backup host returns to said virtual disk image file through said rolling back action the state of the said virutal machine memory data consistent of preserving with said backup host according to said lightweight snapshot;
Said virtual disk image file and the said virutal machine memory data run backup virtual machine instance of said backup host after according to rollback.
3. method according to claim 2 is characterized in that, after the lightweight snapshot of the said virtual disk image file of said acquisition, said method also comprises:
Said lightweight snapshot is kept in the internal memory of said main host; Perhaps
Said lightweight snapshot is kept in the said virtual disk image file.
4. method according to claim 3 is characterized in that,
It is one of following that said renewal operation comprises: read operation, write operation, batch operation, releasing operation.
5. method according to claim 4 is characterized in that, also comprises:
The current mapping table of data block numbering under the preservation current state with virtual disk image file offset address corresponding relation upgraded; Wherein, For the data block that is in unallocated state; Corresponding virtual disk image file offset address is designated unallocated, distributes read-write state for being in, distribute a read states, or the data block of having distributed state to be recycled, use bitmap that the state of this data block is identified.
6. method according to claim 5 is characterized in that,
Said read operation comprises: according to said current mapping table the data block of data block is numbered and be mapped as said virtual disk image file offset address; The state of confirming this data block be in said distributed read-write state, said distributed a read states, or said situation of having distributed state to be recycled under, read the data block of designated length according to said virtual disk image file offset address;
Said write operation comprises: according to said current mapping table the data block of data block is numbered and be mapped as said virtual disk image file offset address; State confirming this data block is under the situation of said unallocated state, carries out said batch operation, and writes data according to said virtual disk image file offset address; Be under the said situation of having distributed read-write state at the state of confirming this data block, directly write data according to said virtual disk image file offset address; Be under the said situation of having distributed a read states at the state of confirming this data block; New data block is carried out said batch operation; With the status modifier of former data block is the said state to be recycled that distributed, and the virtual disk image file offset address that distributes according to said batch operation writes data;
Said batch operation comprises: confirm that minimum unallocated continuation address is an address space to be allocated; With the status modifier of the data block in the said address space to be allocated is the said read-write state that distributed; Obtain the virtual disk image file offset address of said address space to be allocated; Use the virtual disk image file offset address of said address space to be allocated to upgrade said current mapping table, and return the virtual disk image file offset address of said address space to be allocated;
Said releasing operation comprises: the status modifier of the data block that will discharge is unallocated; With the said station location marker of data block in said current mapping table that needs to discharge is unallocated;
Said lock operation comprises: for being in said data block of having distributed read-write state in the said current mapping table, be a said read states that distributed with the status modifier of this data block; For being in said data block of having distributed state to be recycled in the said current mapping table, this data block is carried out said releasing operation; Use said current mapping table to cover the snapshot mapping table, wherein, said snapshot mapping table is used to represent that disk data block under snapshot state is numbered and the corresponding relation of virtual disk image file offset address;
Said rolling back action comprises: for being in said data block of having distributed read-write state in the said current mapping table, this data block is carried out said releasing operation; For being in said data block of having distributed state to be recycled in the said current mapping table, be a said read states that distributed with the status modifier of this data block; Use said snapshot mapping table to cover said current mapping table.
7. method according to claim 1 is characterized in that, also comprises:
After the said virutal machine memory data on the said main host are sent to said backup host, receive the Memory Backup that returns and confirm; And/or
After creating said lightweight snapshot, receive the lightweight snapshot that returns and set up affirmation for said virtual disk image file.
8. main host; It is characterized in that; Comprise host virtual machine, main host monitor of virtual machine and main host operating system and hardware; Wherein said main host monitor of virtual machine comprises that main host migration controller and main host lightweight snapshot drive said main host operating system and hardware bundle includes network storage drive;
Said network memory drives and is used for according to the operation of host virtual machine instance the data block in the virtual disk image file of virutal machine memory data and network memory being upgraded operation;
Said main host migration controller is used for the said virutal machine memory data after upgrading are sent to the backup host;
Said main host lightweight snapshot drives and is used for the said host virtual machine instance of operation suspension; Lock through the state of lock operation data block in the said virtual disk image file; Obtain the lightweight snapshot of said virtual disk image file; And the said host virtual machine instance that resumes operation; Wherein, Said virtual disk image file is used for said backup host and carries out rolling back action, and it is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled, said lightweight to impinge upon soon to preserve current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file.
9. backup host; It is characterized in that; Comprise backup virtual machine, backup host monitor of virtual machine and backup host operating system and hardware; Wherein said backup host monitor of virtual machine comprises that backup host migration controller and backup host lightweight snapshot drive said backup host operating system and hardware bundle includes network storage drive;
Said backup host migration controller is used to preserve the virutal machine memory data that main host sends;
Said backup host lightweight snapshot drives and is used under the situation that the host virtual machine instance can not continue to move; Through rolling back action the virtual disk image file is returned to the state with the said virutal machine memory data consistent of said preservation according to the lightweight snapshot; It is one of following that the state of data block described in the said lightweight snapshot comprises: unallocated state, distribute read-write state, distribute a read states, distribute state to be recycled; Said lightweight impinges upon soon and preserves current mapping table and snapshot mapping table in the said virtual disk image file; Said current mapping table be used for identifying disk under current state and snapshot state data block in the position of virtual disk image file, said snapshot mapping table be used for identifying disk under snapshot state data block in the position of virtual disk image file;
Said backup virtual machine is used for according to said virtual disk image file after the rollback and said virutal machine memory data run backup virtual machine instance.
10. a virtual machine consistently online migratory system comprises network memory, it is characterized in that, also comprises main host as claimed in claim 8 and backup host as claimed in claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100880161A CN101609419B (en) | 2009-06-29 | 2009-06-29 | Continuous on-line transferring data backup method of virtual machine and device thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100880161A CN101609419B (en) | 2009-06-29 | 2009-06-29 | Continuous on-line transferring data backup method of virtual machine and device thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101609419A CN101609419A (en) | 2009-12-23 |
| CN101609419B true CN101609419B (en) | 2012-05-30 |
Family
ID=41483179
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100880161A Expired - Fee Related CN101609419B (en) | 2009-06-29 | 2009-06-29 | Continuous on-line transferring data backup method of virtual machine and device thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101609419B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9135032B2 (en) | 2010-01-29 | 2015-09-15 | International Business Machines Corporation | System, method and computer program product for data processing and system deployment in a virtual environment |
Families Citing this family (69)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8538919B1 (en) | 2009-05-16 | 2013-09-17 | Eric H. Nielsen | System, method, and computer program for real time remote recovery of virtual computing machines |
| CN101741866B (en) * | 2010-02-01 | 2012-12-12 | 浪潮(北京)电子信息产业有限公司 | On-line storage system and method |
| CN101814043B (en) * | 2010-04-19 | 2014-07-16 | 浪潮电子信息产业股份有限公司 | Method for performing on-line recovery by using snapshot technology |
| CN101866307A (en) * | 2010-06-24 | 2010-10-20 | 杭州华三通信技术有限公司 | Data storage method and device based on mirror image technology |
| CN102521071B (en) * | 2011-11-24 | 2013-12-11 | 广州杰赛科技股份有限公司 | Private cloud-based virtual machine maintaining method |
| CN102541686B (en) * | 2011-11-29 | 2015-07-01 | 中标软件有限公司 | Method for achieving backup and disaster recovery of system by utilizing virtual machine |
| CN102520881B (en) * | 2011-12-02 | 2015-09-09 | 中标软件有限公司 | A kind of virtual machine snapshot management method of cloud computing platform and system |
| CN102436410B (en) * | 2011-12-12 | 2013-09-18 | 华中科技大学 | On-line storage method of virtual machine inspection point |
| CN102611745B (en) * | 2011-12-31 | 2015-02-25 | 华为数字技术（成都）有限公司 | On-line file moving method, device and system |
| WO2013105217A1 (en) * | 2012-01-10 | 2013-07-18 | 富士通株式会社 | Virtual machine management program, method and device |
| CN103116534A (en) * | 2012-06-27 | 2013-05-22 | 北京智网科技股份有限公司 | Novel information system emergency disaster recovery method |
| CN103677878B (en) | 2012-09-12 | 2017-07-14 | 国际商业机器公司 | A kind of method and apparatus of patch installing |
| CN104662522B (en) * | 2012-09-28 | 2018-11-02 | Emc 公司 | The system and method backed up using the full virtual machine of storage system functionality |
| CN103051685A (en) * | 2012-12-10 | 2013-04-17 | 广州杰赛科技股份有限公司 | Method for configuring office automation systems in cloud system |
| CN103885719B (en) * | 2012-12-21 | 2017-02-08 | 中国电信股份有限公司 | Device, system and method for online storage migration of virtual machine system |
| CN103019804B (en) * | 2012-12-28 | 2016-05-11 | 中国人民解放军国防科学技术大学 | The virtualized VPS quick migration method of OpenVZ |
| CN103218248B (en) * | 2013-03-25 | 2016-12-28 | 华为技术有限公司 | The update method of a kind of virtual machine image, server and desktop cloud system |
| CN103176868B (en) * | 2013-04-02 | 2017-07-28 | 浪潮电子信息产业股份有限公司 | A kind of file status backup method |
| CN104156369B (en) * | 2013-05-14 | 2019-04-02 | 上海盛大网络发展有限公司 | A kind of database mirroring production method and a kind of database |
| CN103399778B (en) * | 2013-07-01 | 2016-12-28 | 华为技术有限公司 | A kind of virtual machine online bulk migration method and apparatus |
| CN103442033A (en) * | 2013-08-06 | 2013-12-11 | 杭州华三通信技术有限公司 | Running state information synchronizing method and device |
| CN103440160B (en) * | 2013-08-15 | 2016-12-28 | 华为技术有限公司 | Virtual machine restoration methods and virtual machine migration method and device and system |
| CN104572656B (en) * | 2013-10-12 | 2019-02-19 | 华为技术有限公司 | A kind of method and device of disk mirroring space reclamation |
| CN103546556B (en) * | 2013-10-22 | 2017-04-05 | 同济大学 | One kind online moving method of virtual machine in future network XIA |
| CN104636125B (en) * | 2013-11-13 | 2018-03-23 | 联想(北京)有限公司 | A kind of information processing method and device |
| CN103605588B (en) * | 2013-11-21 | 2016-08-17 | 上海爱数信息技术股份有限公司 | A kind of disk backup method of virtue machine |
| CN103678045B (en) * | 2013-12-31 | 2017-11-21 | 曙光云计算集团有限公司 | The data back up method of virtual machine |
| CN103957229B (en) * | 2013-12-31 | 2017-07-14 | 电子科技大学 | Active update method, device and the server of physical machine in IaaS cloud system |
| JP5713138B1 (en) * | 2014-09-12 | 2015-05-07 | 富士ゼロックス株式会社 | Virtual computer system, printer control system, virtual computer program, and printer control program |
| CN104331318B (en) * | 2014-10-27 | 2018-04-06 | 上海斐讯数据通信技术有限公司 | A kind of Embedded Double system and its method of work |
| CN104407935B (en) * | 2014-11-07 | 2018-05-18 | 华为数字技术（成都）有限公司 | Snapshot rollback method and storage device |
| CN104461773A (en) * | 2014-11-18 | 2015-03-25 | 国云科技股份有限公司 | Backup deduplication method of virtual machine |
| CN104572248B (en) * | 2014-12-17 | 2018-01-23 | 中电科华云信息技术有限公司 | The method of virtual machine dynamical save snapshot |
| CN105989008B (en) * | 2015-01-27 | 2020-06-23 | 炬芯（珠海）科技有限公司 | Method and equipment for writing data into mirror image file |
| US9639395B2 (en) * | 2015-04-16 | 2017-05-02 | Google Inc. | Byte application migration |
| CN106484312B (en) * | 2015-09-01 | 2019-07-26 | 北京金山云网络技术有限公司 | A kind of magnetic disk of virtual machine data migration method and device |
| CN105224391B (en) * | 2015-10-12 | 2018-10-12 | 浪潮(北京)电子信息产业有限公司 | A kind of online backup method and system of virtual machine |
| CN105224676A (en) * | 2015-10-13 | 2016-01-06 | 北京奇虎科技有限公司 | The method of mobile terminal file migration and device |
| CN105446834B (en) | 2015-11-30 | 2018-10-19 | 华为技术有限公司 | The generation method and device of virtual machine snapshot |
| WO2017132872A1 (en) * | 2016-02-03 | 2017-08-10 | 华为技术有限公司 | Virtual machine backup method, backup device and host |
| CN105912428B (en) * | 2016-05-20 | 2019-01-08 | 上海数腾软件科技股份有限公司 | Realize that source data is converted into the system and method for virtual machine image in real time |
| CN106445730B (en) | 2016-07-22 | 2019-12-03 | 平安科技（深圳）有限公司 | A kind of method and terminal improving virtual machine performance |
| CN106250267A (en) * | 2016-07-22 | 2016-12-21 | 浪潮(北京)电子信息产业有限公司 | The data back up method of a kind of virtual machine and device |
| CN106293882A (en) * | 2016-08-12 | 2017-01-04 | 浪潮(北京)电子信息产业有限公司 | A kind of virtual disk snapshot implementing method and device |
| CN106293535B (en) * | 2016-08-12 | 2020-04-03 | 浪潮(北京)电子信息产业有限公司 | NAS performance optimization method and device |
| CN106250274B (en) * | 2016-08-16 | 2019-03-15 | 浪潮(北京)电子信息产业有限公司 | A kind of backup method and device of virtual machine |
| CN106603522A (en) * | 2016-12-09 | 2017-04-26 | 北京安天电子设备有限公司 | Cloud data virus detection method and cloud data virus detection system based on bypass listening |
| CN107153569B (en) * | 2017-03-23 | 2021-02-19 | 深信服科技股份有限公司 | Processing method and device for read-write failure of virtual machine |
| CN107168774B (en) * | 2017-06-14 | 2020-11-10 | 北京云联万维技术有限公司 | Virtual machine migration method and system based on local storage |
| CN109558205B (en) * | 2017-09-25 | 2022-11-29 | 南京中兴新软件有限责任公司 | Disk access method and device |
| CN107864051A (en) * | 2017-10-23 | 2018-03-30 | 许继电气股份有限公司 | The computer equipment fault of electric power monitoring system solves method and electric power monitoring system |
| CN108536477B (en) * | 2018-03-02 | 2021-06-15 | 烽火通信科技股份有限公司 | Method for distributing and setting centralized equipment port resources |
| CN110622478B (en) * | 2018-04-08 | 2020-11-06 | 华为技术有限公司 | Method and device for data synchronous processing |
| CN108845895B (en) * | 2018-07-06 | 2022-02-08 | 无锡地铁集团有限公司 | Streaming incremental backup method and device based on virtual disk file |
| CN109597671B (en) * | 2018-12-11 | 2022-06-10 | 国云科技股份有限公司 | Desktop cloud implementation method compatible with centralized and distributed operation |
| CN109814979B (en) * | 2018-12-17 | 2021-03-30 | 华为技术有限公司 | Data loading method and device, computer equipment and readable storage medium |
| CN109669818B (en) * | 2018-12-20 | 2019-11-12 | 广州鼎甲计算机科技有限公司 | Continuous data protection method and system without local cache |
| CN109828829B (en) * | 2019-01-22 | 2022-10-18 | 重庆邮电大学 | Virtual machine rapid evacuation method based on disaster early warning time |
| CN111913832B (en) * | 2019-05-10 | 2024-07-09 | 杭州信核数据科技股份有限公司 | Block-level CDP data storage disaster recovery method and device based on object storage |
| CN110442427B (en) * | 2019-07-29 | 2021-09-14 | 腾讯科技（深圳）有限公司 | Virtual machine snapshot management method, device, equipment and storage medium |
| CN110688064A (en) * | 2019-09-05 | 2020-01-14 | 浪潮电子信息产业股份有限公司 | Virtual disk migration method, device, equipment and readable storage medium |
| CN112559114A (en) * | 2019-09-24 | 2021-03-26 | 阿里巴巴集团控股有限公司 | Virtual machine generation method and device |
| CN111143025B (en) * | 2019-11-22 | 2023-03-24 | 中国船舶工业系统工程研究院 | Method for real-time virtual machine instance management |
| US11822950B2 (en) | 2020-07-23 | 2023-11-21 | Vmware, Inc. | Cloneless snapshot reversion |
| CN112395050B (en) * | 2020-11-30 | 2024-02-23 | 深信服科技股份有限公司 | Virtual machine backup method and device, electronic equipment and readable storage medium |
| CN112596950B (en) * | 2020-12-23 | 2024-07-16 | 深圳市科力锐科技有限公司 | Virtual machine data backup method, device, equipment and storage medium |
| CN114116135A (en) * | 2021-12-02 | 2022-03-01 | 北京天融信网络安全技术有限公司 | Virtual machine storage method and device, electronic equipment and storage medium |
| CN114968670B (en) * | 2022-07-27 | 2022-11-18 | 国网信息通信产业集团有限公司 | Virtual machine backup and recovery method and device |
| CN117093332B (en) * | 2023-10-20 | 2024-02-27 | 杭州优云科技有限公司 | Method and device for realizing cloning of virtual machine |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101101562A (en) * | 2007-07-10 | 2008-01-09 | 北京大学 | Dummy machine external storage on-line migration method |
| CN101464812A (en) * | 2009-01-06 | 2009-06-24 | 北京航空航天大学 | Virtual machine migration method |
-
2009
- 2009-06-29 CN CN2009100880161A patent/CN101609419B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101101562A (en) * | 2007-07-10 | 2008-01-09 | 北京大学 | Dummy machine external storage on-line migration method |
| CN101464812A (en) * | 2009-01-06 | 2009-06-24 | 北京航空航天大学 | Virtual machine migration method |
Non-Patent Citations (2)
| Title |
|---|
| 朱晓娜,李先贤,李沁.面向服务网格的虚拟环境部署运行管理系统.《计算机工程与应用》.2007,第43卷(第36期),第124-128页. * |
| 蔡炳,李沁,李先贤.虚拟运行环境描述管理工具的研究和实现.《计算机工程与设计》.2007,第28卷(第23期),第5628-5630,5645页. * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9135032B2 (en) | 2010-01-29 | 2015-09-15 | International Business Machines Corporation | System, method and computer program product for data processing and system deployment in a virtual environment |
| US9582311B2 (en) | 2010-01-29 | 2017-02-28 | International Business Machines Corporation | System, method and computer program product for data processing and system deployment in a virtual environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101609419A (en) | 2009-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101609419B (en) | Continuous on-line transferring data backup method of virtual machine and device thereof | |
| CN102981931B (en) | Backup method and device for virtual machine | |
| CN102073462B (en) | Virtual storage migration method and system and virtual machine monitor | |
| US7594137B2 (en) | Storage system and data restoration method thereof | |
| US8504765B2 (en) | Method of improving efficiency of capacity of volume used for copy function and apparatus thereof | |
| CN102299904B (en) | System and method for realizing service data backup | |
| US7127578B2 (en) | Storage device and information management system | |
| US7398363B2 (en) | Storage system and power source control method thereof, storage apparatus and data processing method | |
| CN101755257B (en) | Managing the copying of writes from primary storages to secondary storages across different networks | |
| CN101876926B (en) | Asymmetric software triple-computer hot backup fault-tolerant method | |
| CN106528327A (en) | Data processing method and backup server | |
| CN101441582A (en) | Backup and recovery method of computer data object and system and program product for implementing the same | |
| JP2001034426A (en) | Data processor storage system having dynamic re- synchronization of mirror logical data volume following storage system failure | |
| CN106603665B (en) | Cloud platform continuous data synchronous method and its device | |
| US11307789B2 (en) | Storage system and storage control method | |
| CN102065138A (en) | Method for realizing thermal backup control of servers in disk-less system | |
| CN115586950A (en) | Method and device for realizing non-stop cloud-loading migration based on memory system | |
| US10761767B2 (en) | Computer system and method for controlling storage apparatus that has replication direction from first logical device (in first storage) to second logical device (in second storage) and from said second logical device to third logical device (in said second storage), wherein said replication direction is reversed when second computer takes over for first computer | |
| CN102317921A (en) | High reliability computer system and constructive method thereof | |
| CN103036722A (en) | Method for achieving server hot backup in diskless system | |
| US20060212669A1 (en) | Control method for storage system, storage system, storage control apparatus, control program for storage system, and information processing system | |
| CN112380049A (en) | Application method of storage snapshot technology in mirror image | |
| JP7179947B2 (en) | Storage system and storage control method | |
| CN109614034A (en) | A method of for the complex data center bulk migration multi-platform across physical region | |
| CN112230859B (en) | Read-write migration method and device without interrupting service |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120530 Termination date: 20170629 |