CN106775846A - For the method and device of the online migration of physical server - Google Patents
For the method and device of the online migration of physical server Download PDFInfo
- Publication number
- CN106775846A CN106775846A CN201611085506.2A CN201611085506A CN106775846A CN 106775846 A CN106775846 A CN 106775846A CN 201611085506 A CN201611085506 A CN 201611085506A CN 106775846 A CN106775846 A CN 106775846A
- Authority
- CN
- China
- Prior art keywords
- disk
- data
- file
- physical server
- address
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/60—Software deployment
- G06F8/61—Installation
- G06F8/63—Image based installation; Cloning; Build to order
Abstract
A kind of method and device of online migration for physical server is provided.Methods described includes：(A) partition information of disk is obtained, and the partition information is sent to another physical server, wherein, the partition information is the information of the subregion on disk；(B) will be stored in the address of the data on the disk and the data on the disk and be sent to another physical server；(C) control operation system enters resting state, so that internal storage data is saved in the disk；(D) address by the internal storage data and the internal storage data that are preserved on the disk on the disk is sent to another physical server, wherein, the physical server perform step (A) and (B) when, all the time in normal operating condition.According to methods described and device, the out of service time in transition process can be effectively reduced.
Description
Technical field
All things considered of the present invention is related to operating system migration field, more particularly, is related to a kind of for physical server
Online migration operation system method and device.
Background technology
Operating system migration, including offline migration and online two kinds of migration.
Offline migration (offline migration)：Refer in operation also known as static migrating (static migration)
In the case of system closedown, data in magnetic disk is migrated.
Online migration (online migration)：Also known as real-time migration, heat (hot) migration, dynamic migration (live
Migration), refer in the state of operating system, data in magnetic disk and internal storage data to be migrated.The technology is current
Have been widely used for virtual machine scene, i.e. virtual machine to online migration (the Virtual-to-Virtual live of virtual machine
Migration, abbreviation V2V Live Migration).But the operating system on physical server is migrated online
(Physical-to-Physical Live Migration, abbreviation P2P Live Migration) is not yet ripe.
Virtual machine is migrated online：The online Migration tools of virtual machine of current main flow, such as VMotion, XEN of VMware
XenMotion, requires between physical machine using SAN (storage area network), NAS (network-attached
) etc storage the shared external equipment of centralization, thus only need to consider that operating system memory performs state in migration
Migration, so as to obtain preferable migration performance.Migration mainly passes through pre-copy internal memory to virutal machine memory in virtualization layer online
(Pre-copy memory migration) technology is migrated.
Physical machine is migrated online：Currently without the scheme that the operating system of complete exercisable physical machine is migrated online,
The operating system migration of physical machine is mainly carried out by the way of offline.
The content of the invention
Exemplary embodiment of the invention is the method and device for providing a kind of online migration for physical server,
It can effectively reduce the out of service time in transition process.
Exemplary embodiment of the invention, there is provided a kind of method of online migration for physical server, wherein,
Methods described includes：(A) partition information of disk is obtained, and the partition information is sent to another physical server, wherein,
The partition information is the information of the subregion on disk；(B) data and the data on the disk be will be stored in institute
State the address on disk and be sent to another physical server；(C) control operation system enters resting state, so that interior poke
According to being saved in the disk；(D) ground of the internal storage data and the internal storage data that will be preserved on the disk on the disk
Location is sent to another physical server, wherein, the physical server perform step (A) and (B) when, all the time in just
Normal running status.
Alternatively, step (B) includes：(b1) file system metadata of each subregion of the disk is obtained；(b2) root
The data block being had been used in described each subregion is determined according to the file system metadata of each subregion；(b3) will storage
The address of data and the data on the disk on the disk is sent to another physical server, wherein,
The data include the file system metadata of the data and each subregion in the data block that is had been used in described each subregion.
Alternatively, step (B) also includes：(b4) while step (b3) is performed, the text stored on the disk is monitored
Part is pressed from both sides and whether file has change；(b5) whenever file is monitored and/or file has change, the file that will be modified
And/or the address of the data of file and the data of the file being modified and/or file on the disk is sent to
Another physical server.
Alternatively, whether monitor the file and file of storage on the disk includes the step of having change：Monitoring is described
Whether the file and file stored on disk are performed at least one among following operation：Create, change and delete.
In accordance with an alternative illustrative embodiment of the present invention, there is provided a kind of method of online migration for physical server,
Wherein, methods described includes：(A) partition information that another physical server sends is received, and according to the partition information to magnetic
Disk carries out subregion；(B) storage that reception another physical server sends is on the disk of another physical server
The address of data and the data, and write at the identical address of the disk data according to the address of the data；
(C) internal storage data and the ground of the internal storage data of described another physical server that another physical server sends are received
Location, and write at the identical address of the disk internal storage data according to the address of the internal storage data；(D) again from institute
State during the internal storage data that disk started and will be stored in the disk by the wake-up mechanism of operating system returns to internal memory.
Alternatively, data of the storage on the disk of another physical server include：Another physical server
Disk each subregion in data in the data block that has been used and each subregion file system metadata.
Alternatively, methods described also includes：(E) while step (B) is performed, another physical server hair is received
The file and/or the data of file and the text being modified being modified on the disk of the described another physical server for sending
The address of the data of part folder and/or file, and according to the address of the file being modified and/or the data of file by described in
The file and/or the data of file being modified are write at the identical address of the disk.
Alternatively, the file being modified and/or file are to be performed the text of at least one among following operation
Part is pressed from both sides and/or file：Create, change and delete.
In accordance with an alternative illustrative embodiment of the present invention, there is provided a kind of device of online migration for physical server,
Wherein, described device includes：Subregion synchronization module, is configured as obtaining the partition information of disk, and the partition information is sent out
Another physical server is sent to, wherein, the partition information is the information of the subregion on disk；Data simultaneous module, is matched somebody with somebody
It is set to the address of the data and the data that will be stored on the disk on the disk and is sent to another physics clothes
Business device；Dormancy control module, is configured as control operation system and enters resting state, so that internal storage data is saved in the magnetic
Disk；Internal storage data synchronization module, is configured as the internal storage data preserved on the disk and the internal storage data in the magnetic
Address on disk is sent to another physical server, wherein, perform operation in subregion synchronization module and data simultaneous module
While, the physical server is in normal operating condition all the time.
Alternatively, data simultaneous module includes：Metadata acquiring unit, is configured as obtaining each subregion of the disk
File system metadata；Data block determining unit, is configured as the file system metadata according to each subregion come really
The data block being had been used in fixed described each subregion；Valid data synchronization unit, is configured as will be stored on the disk
Address on the disk of data and the data be sent to another physical server, wherein, the data include
The data in data block and the file system metadata of each subregion being had been used in described each subregion.
Alternatively, data simultaneous module also includes：Monitoring unit, is configured as performing operation in valid data synchronization unit
While, whether the file and file for monitoring storage on the disk have change；Increment synchronization unit, is configured as whenever prison
Control unit monitors to file and/or file have change when, the file and/or the data of file and the quilt that will be modified
The address of the file of change and/or the data of file on the disk is sent to another physical server.
Alternatively, whether the file and file for being stored on the monitoring unit monitoring disk are performed among following operation
At least one：Create, change and delete.
In accordance with an alternative illustrative embodiment of the present invention, there is provided a kind of device of online migration for physical server,
Wherein, described device includes：Subregion synchronization module, is configured as receiving the partition information that another physical server sends, and root
Subregion is carried out to disk according to the partition information；Data simultaneous module, is configured as receiving another physical server transmission
Data and the data of the storage on the disk of another physical server address, and according to the address of the data
The data are write at the identical address of the disk；Internal storage data synchronization module, is configured as receiving another physics
The internal storage data of described another physical server that server sends and the address of the internal storage data, and according to the interior poke
According to address the internal storage data is write at the identical address of the disk；Internal memory recovery module, is configured as again from institute
State during the internal storage data that disk started and will be stored in the disk by the wake-up mechanism of operating system returns to internal memory.
Alternatively, data of the storage on the disk of another physical server include：Another physical server
Disk each subregion in data in the data block that has been used and each subregion file system metadata.
Alternatively, described device also includes：Increment synchronization module, is configured as performing the same of operation in data simultaneous module
When, the file that is modified on the disk for receiving described another physical server that another physical server sends and/or
The address of the data of the data of file and the file being modified and/or file, and according to the file being modified
The data of the file being modified and/or file are write the identical of the disk by the address of the data of folder and/or file
At address.
Alternatively, the file being modified and/or file are to be performed the text of at least one among following operation
Part is pressed from both sides and/or file：Create, change and delete.
In the method and device for the online migration of physical server of exemplary embodiment of the invention, thing
Reason server interrupts the time of service only since the physical server dormancy of source, and end is waken up to purpose physical server, has
Effect shortens the interrupt service time in transition process；Also, only the valid data on the disk of source physical server are carried out
Copy migration, effectively reduces transit time.
The aspect and/or advantage other by present general inventive concept is partly illustrated in following description, also one
Divide and be will be apparent by description, or can be learnt by the implementation of present general inventive concept.
Brief description of the drawings
By the description carried out with reference to the accompanying drawing for being exemplarily illustrated embodiment, exemplary embodiment of the present it is upper
State and be will become apparent with other purposes and feature, wherein：
Fig. 1 shows the flow of the method for the online migration for physical server according to an exemplary embodiment of the present invention
Figure；
Fig. 2 shows the data that will be stored on the disk according to an exemplary embodiment of the present invention and the data in institute
State the flow chart the step of address on disk is sent to purpose physical server；
Fig. 3 shows the flow of the method for the online migration for physical server according to an exemplary embodiment of the present invention
Figure；
Fig. 4 shows the block diagram of the device of the online migration for physical server according to an exemplary embodiment of the present invention；
Fig. 5 shows the block diagram of data simultaneous module according to an exemplary embodiment of the present invention；
Fig. 6 shows the block diagram of the device of the online migration for physical server according to an exemplary embodiment of the present invention.
Specific embodiment
Embodiments of the invention are reference will now be made in detail, the example of the embodiment is shown in the drawings, wherein, identical mark
Number identical part is referred to all the time.The embodiment will be below illustrated by referring to accompanying drawing, to explain the present invention.
Fig. 1 shows the flow of the method for the online migration for physical server according to an exemplary embodiment of the present invention
Figure.Here, the physical server (hereinafter referred to as source physical server) can be the physical servers such as physical machine, bare metal.
Methods described can be performed by source physical server, can also be realized by computer program.For example, methods described
Can be performed by the function program in the physical server of source.
As shown in figure 1, in step S10, obtaining the partition information of disk, and the partition information is sent to another physics
Server (hereinafter referred to as purpose physical server), wherein, the partition information is the information of the subregion on disk.Here,
The partition information may include to be divided into the information of which subregion, the initial address of each subregion and length etc. on disk.
In step S20, will be stored in the address of the data on the disk and the data on the disk and be sent to
Purpose physical server.Particularly, the data full dose on the disk of source physical server is synchronized to purpose physical server
Disk on, and migrated by address, it is ensured that address of the same data on the disk of source physical server (that is, stores position
Put) and address (that is, storage location) on the disk of purpose physical server it is identical.
It should be noted that source physical server is when step S10 and S20 is performed, all the time in normal operating condition, so as to ensure
Data full dose synchronous phase most long is taken in whole transition process, source physical server can normally provide service, will not
Interrupt service.
In step S30, control operation system enters resting state, so that internal storage data is saved in the disk.Specifically
Come, operating system enters after resting state, all of process is frozen, and internal storage data is saved to the swap file specified
Or in exchange partition.
In step S40, address of the internal storage data and the internal storage data that will be preserved on the disk on the disk
It is sent to purpose physical server.Particularly, the internal storage data preserved on the disk of source physical server is pressed into address transfer
Onto the disk of purpose physical server, it is ensured that address of the internal storage data on the disk of source physical server and in purpose physics
Address on the disk of server is identical, i.e. be maintained in swap file or exchange partition.
Internal storage data is saved in disk by using the dormancy mechanism of operating system, internal storage data is then moved into mesh
Physical server, so as to realize the migration of internal storage data.This process takes relatively fewer, reduces the clothes of source physical server
The business break period.Also, press address transfer due to being used in step S20, it is ensured that move to the interior of purpose physical server
The matching of the data stored on deposit data and disk.
For step S20, it is preferred that, described according to an exemplary embodiment of the present hereinafter with reference to Fig. 2
The method for the online migration of physical server in the data that will be stored on the disk and the data described
Address on disk is sent to the flow of purpose physical server.
As shown in Fig. 2 in step S201, obtaining the file system metadata of each subregion of the disk.
In step S202, file system metadata according to each subregion is made in each subregion described in determining
Data block.
As an example, each subregion can be directed to, the text of each subregion is obtained from the file system metadata of each subregion
Part metadata, then determines the data block (that is, the data block having been used) that file is used according to file metadata.
In step S203, will be stored in the address of the data on the disk and the data on the disk and be sent to
Purpose physical server, wherein, the data include data and each in the data block that is had been used in described each subregion
The file system metadata of subregion.
In view of disk Spatial General 6 R than larger (for example, hundreds of GB), and disk the slow (example of read or write speed
Such as, the read or write speed of ordinary magnetic disc is 130M/S or so, and the read or write speed of SSD is 500M/S or so), it is of the invention above-mentioned
Mode, the valid data that will only can be stored on disk are migrated, and effectively reduce transit time.
Preferably, the data that will be stored on the disk according to an exemplary embodiment of the present invention and the data are in institute
State the address on disk and be sent to the flow of purpose physical server and may also include：While step S203 is performed, institute is monitored
State whether the file and file stored on disk have change.Whenever file is monitored and/or file has change, will be by more
The file and/or the data of file and the data of the file being modified and/or file for changing are on the disk
Address is sent to purpose physical server.That is, to the file that is modified on the disk when step S203 is performed and/or file
It is subsynchronous that data carry out one again.
As an example, can monitor whether the file and file stored on the disk are performed among following operation extremely
One item missing：Create, change and delete.As an example, using the inotify mechanism of (SuSE) Linux OS to whole file system
File on disk and file in system are monitored.
Because the data full dose synchronizing process time is more long, period operating system is in running status again, and file system can be produced
Raw dirty data, increment synchronization is carried out by data full dose synchronizing process, be ensure that on the disk of source physical server
The uniformity of the data in data and purpose physical server.
Fig. 3 shows according to an exemplary embodiment of the present invention to exist for physical server (that is, purpose physical server)
The flow chart of the method for line migration.Here, purpose physical server can be the physical servers such as physical machine, bare metal.
Methods described can be performed by purpose physical server, can also be realized by computer program.For example, the side
Method can be performed by the function program in purpose physical server.
As shown in figure 3, in step S50, the partition information that source physical server sends is received, and according to the partition information
Subregion is carried out to disk.Here, the partition information may include to be divided into the starting of which subregion, each subregion on disk
The information of address and length etc..
Before step S50 is performed, purpose physical server can start from boot kernel, and boot kernel refer to one
Individual internal memory operation system can be Live CD or be loaded by PXE, iPXE；Additionally, can also be opened from common operating system
Move, but the disk is not used.
In step S60, data and institute of the storage of source physical server transmission on the disk of source physical server are received
The address of data is stated, and is write at the identical address of the disk data according to the address of the data.Particularly,
Data full dose on the disk of source physical server is synchronized on the disk of purpose physical server, and is moved by address
Move, it is ensured that address (that is, storage location) of the same data on the disk of source physical server and in purpose physical server
Address (that is, storage location) on disk is identical.
Preferably, data of the storage on the disk of source physical server may include：The disk of source physical server it is every
The data in data block and the file system metadata of each subregion being had been used in individual subregion.That is, only by source physical services
The valid data stored on the disk of device are migrated, and effectively reduce transit time.
In step S70, the internal storage data and the internal storage data of the source physical server that source physical server sends are received
Address, and the internal storage data is write at the identical address of the disk according to the address of the internal storage data.Specifically
Come, the internal storage data preserved on the disk of source physical server is pressed into address transfer to the disk of purpose physical server, protect
Address of the card internal storage data on the disk of source physical server is identical with the address on the disk of purpose physical server,
That is, it is maintained in swap file or exchange partition.
In step S80, start from the disk again and the wake-up mechanism by operating system will be stored in the disk
In internal storage data return to internal memory, so as to realize the migration (migration of process) of internal storage data.
As preferred exemplary, the side of the online migration for purpose physical server according to an exemplary embodiment of the present invention
Method may also include：While step S60 is performed, receive on the disk of the source physical server that source physical server sends by more
The file and/or the data of file that change and the address of the data of the file being modified and/or file, and according to institute
The address for stating the data of the file and/or file being modified writes the data of the file being modified and/or file
At the identical address of the disk.
As an example, the file being modified and/or file can be performed at least among following operation
The file and/or file of item：Create, change and delete.
Because the data full dose synchronizing process time is more long, the operating system of period source physical server is again in operation shape
State, file system can produce dirty data, and increment synchronization is carried out by data full dose synchronizing process, ensure that source physics clothes
The uniformity of the data being engaged in the data and purpose physical server on the disk of device.
Fig. 4 shows according to an exemplary embodiment of the present invention for the online of physical server (that is, source physical server)
The block diagram of the device of migration.
As shown in figure 4, the device bag of the online migration for source physical server according to an exemplary embodiment of the present invention
Include：Subregion synchronization module 10, data simultaneous module 20, dormancy control module 30 and internal storage data synchronization module 40.These modules
Can be realized by special device, as an example, the module can by digital signal processor, field programmable gate array,
The common hardware such as application processor, CPU processor realized, also can be realized by dedicated hardware processors such as special chips,
Can also be realized with software mode by computer program completely, for example, the function program being implemented as in the physical server of source.
Subregion synchronization module 10 is configured as obtaining the partition information of disk, and the partition information is sent into purpose thing
Reason server, wherein, the partition information is the information of the subregion on disk.Here, the partition information may include on
Disk is divided into the information of which subregion, the initial address of each subregion and length etc..
Data simultaneous module 20 is configured as will be stored in data and the data on the disk on the disk
Address be sent to purpose physical server.Particularly, data simultaneous module 20 is by the number on the disk of source physical server
It is synchronized on the disk of purpose physical server according to full dose, and is migrated by address, it is ensured that same data is in source physical services
Address (that is, storage location) on the disk of device and address (that is, the storage location) phase on the disk of purpose physical server
Together.
It should be noted that while subregion synchronization module 10 and data simultaneous module 20 perform operation, source physical server begins
Normal operating condition is in eventually, so as to ensure that data full dose synchronous phase most long, source physics are taken in whole transition process
Server can normally provide service, will not interrupt service.
Dormancy control module 30 is configured as control operation system and enters resting state so that internal storage data be saved in it is described
Disk.Particularly, operating system enters after resting state, and all of process is frozen, and internal storage data is saved to finger
In fixed swap file or exchange partition.
Internal storage data synchronization module 40 is configured as existing the internal storage data and the internal storage data that are preserved on the disk
Address on the disk is sent to purpose physical server.Particularly, internal storage data synchronization module 40 is by source physical services
The internal storage data preserved on the disk of device presses address transfer to the disk of purpose physical server, it is ensured that internal storage data is in source thing
The address managed on the disk of server is identical with the address on the disk of purpose physical server, i.e. be maintained in exchanging text
In part or exchange partition.
Internal storage data is saved in disk by using the dormancy mechanism of operating system, internal storage data is then moved into mesh
Physical server, so as to realize the migration of internal storage data.This process takes relatively fewer, reduces the clothes of source physical server
The business break period.Also, press address transfer because data simultaneous module 20 is used, it is ensured that move to purpose physical server
The matching of the data stored on internal storage data and disk.
It is preferred that, the example arrangement of data simultaneous module is described hereinafter with reference to Fig. 5.As shown in figure 5, root
Be may include according to the data simultaneous module 20 of exemplary embodiment of the present：Metadata acquiring unit 201, data block determining unit
202 and valid data synchronization unit 203.
Metadata acquiring unit 201 is configured as the file system metadata of each subregion for obtaining the disk.
The file system metadata that data block determining unit 202 is configured as each subregion according to is described every to determine
The data block being had been used in individual subregion.
As an example, data block determining unit 202 can be directed to each subregion, from the file system metadata of each subregion
The file metadata of each subregion is obtained, then determines that the data block that file is used (that is, has been used according to file metadata
Data block).
Valid data synchronization unit 203 is configured as will be stored in data and the data on the disk in the magnetic
Address on disk is sent to purpose physical server, wherein, the data include the data being had been used in described each subregion
The file system metadata of data and each subregion in block.
In view of disk Spatial General 6 R than larger (for example, hundreds of GB), and disk the slow (example of read or write speed
Such as, the read or write speed of ordinary magnetic disc is 130M/S or so, and the read or write speed of SSD is 500M/S or so), valid data synchronization unit
203 valid data that will only can be stored on disk are migrated, and effectively reduce transit time.
Preferably, data simultaneous module according to an exemplary embodiment of the present invention may also include：Monitoring unit (not shown)
With increment synchronization unit (not shown).
Monitoring unit is configured as while valid data synchronization unit is performed and operated, monitoring what is stored on the disk
Whether file and file have change.
As an example, monitoring unit can monitor whether the file and file stored on the disk are performed following operation
Among at least one：Create, change and delete.As an example, monitoring unit can utilize the inotify of (SuSE) Linux OS
Mechanism to whole file system in file on disk and file be monitored.
Increment synchronization unit is configured as when monitoring unit monitors file and/or file has change, will be by more
The file and/or the data of file and the data of the file being modified and/or file for changing are on the disk
Address is sent to purpose physical server.That is, the text to being modified on the disk when valid data synchronization unit 203 is performed and operated
It is subsynchronous that the data of part folder and/or file carry out one again.
Because the data full dose synchronizing process time is more long, period operating system is in running status again, and file system can be produced
Raw dirty data, increment synchronization is carried out by data full dose synchronizing process, be ensure that on the disk of source physical server
The uniformity of the data in data and purpose physical server.
Fig. 6 shows according to an exemplary embodiment of the present invention to exist for physical server (that is, purpose physical server)
The block diagram of the device of line migration.
As shown in fig. 6, the device of the online migration for purpose physical server according to an exemplary embodiment of the present invention
Including：Subregion synchronization module 50, data simultaneous module 60, internal storage data synchronization module 70 and internal memory recovery module 80.These moulds
Block can be realized by special device, as an example, the module can be by digital signal processor, field-programmable gate array
The common hardware such as row, application processor, CPU processor realizes, also can be by dedicated hardware processors such as special chips come real
It is existing, can also be realized with software mode by computer program completely, for example, the function being implemented as in purpose physical server
Program.
Subregion synchronization module 50 is configured as the partition information that reception source physical server sends, and is believed according to the subregion
Breath carries out subregion to disk.
Data simultaneous module 60 is configured as the disk of the storage in source physical server of reception source physical server transmission
On data and the data address, and the data are write the identical address of the disk according to the address of the data
Place.
Preferably, data of the storage on the disk of source physical server may include：The disk of source physical server it is every
The data in data block and the file system metadata of each subregion being had been used in individual subregion.
Internal storage data synchronization module 70 is configured as the interior poke of the source physical server that reception source physical server sends
According to the address with the internal storage data, and the internal storage data is write the phase of the disk according to the address of the internal storage data
At address.
Internal memory recovery module 80 is configured as starting from the disk again and the wake-up mechanism by operating system will be protected
Exist during the internal storage data in the disk returns to internal memory, so as to realize the migration (migration of process) of internal storage data.
As preferred exemplary, the dress of the online migration for purpose physical server according to an exemplary embodiment of the present invention
Put and may also include：Increment synchronization module (not shown).Increment synchronization module is configured as performing operation in data simultaneous module
Meanwhile, the file and/or the data of file being modified on the disk for receiving the source physical server that source physical server sends
And the address of the data of the file being modified and/or file, and according to the file and/or file being modified
The address of data the data of the file being modified and/or file are write at the identical address of the disk.
As an example, the file being modified and/or file can be performed at least among following operation
The file and/or file of item：Create, change and delete.
The method and device of the online migration for physical server of exemplary embodiment of the invention, physics clothes
Business device interrupts the time of service only since the physical server dormancy of source, and end is waken up to purpose physical server, effectively contracting
Interrupt service time in short transition process；Also, only the valid data on the disk of source physical server are copied
Migration, effectively reduces transit time.
Although having show and described some exemplary embodiments of the invention, it will be understood by those skilled in the art that
In the case where the principle of the invention that its scope is limited by claim and its equivalent and spirit is not departed from, can be to these
Embodiment is modified.
Claims (10)
1. a kind of method of online migration for physical server, wherein, methods described includes：
(A) partition information of disk is obtained, and the partition information is sent to another physical server, wherein, the subregion
Information is the information of the subregion on disk；
(B) will be stored in the address of the data on the disk and the data on the disk and be sent to another physics
Server；
(C) control operation system enters resting state, so that internal storage data is saved in the disk；
(D) address by the internal storage data and the internal storage data that are preserved on the disk on the disk is sent to described another
One physical server,
Wherein, the physical server perform step (A) and (B) when, all the time in normal operating condition.
2. method according to claim 1, wherein, step (B) includes：
(b1) file system metadata of each subregion of the disk is obtained；
(b2) according to the file system metadata of each subregion come the data block that is had been used in each subregion described in determining；
(b3) will be stored in the address of the data on the disk and the data on the disk and be sent to another thing
Reason server, wherein, the data include data in the data block that is had been used in described each subregion and each subregion
File system metadata.
3. method according to claim 2, wherein, step (B) also includes：
(b4) while step (b3) is performed, whether the file and file for monitoring storage on the disk have change；
(b5) whenever file is monitored and/or file has change, the file and/or the data of file that will be modified and
Address of the data of the file being modified and/or file on the disk is sent to another physical server.
4. method according to claim 3, wherein, monitor whether the file that is stored on the disk and file have change
The step of include：
Monitor at least one whether file and file stored on the disk is performed among following operation：Create, repair
Change and delete.
5. a kind of method of online migration for physical server, wherein, methods described includes：
(A) partition information that another physical server sends is received, and subregion is carried out to disk according to the partition information；
(B) data and institute of the storage of another physical server transmission on the disk of another physical server are received
The address of data is stated, and is write at the identical address of the disk data according to the address of the data；
(C) internal storage data and the internal storage data of described another physical server that another physical server sends are received
Address, and the internal storage data is write at the identical address of the disk according to the address of the internal storage data；
(D) start and the internal storage data in being stored in the disk by the wake-up mechanism of operating system from the disk again
In returning to internal memory.
6. method according to claim 5, wherein, store the packet on the disk of another physical server
Include：The data in data block and the text of each subregion being had been used in each subregion of the disk of another physical server
Part system metadata.
7. method according to claim 6, also includes：
(E) while step (B) is performed, the magnetic of described another physical server that another physical server sends is received
The file and/or the data of file that are modified on disk and the address of the data of the file being modified and/or file,
And according to the address of the file being modified and/or the data of file by the file being modified and/or file
Data are write at the identical address of the disk.
8. method according to claim 7, wherein, the file being modified and/or file are to be performed following behaviour
The file of at least one and/or file among work：Create, change and delete.
9. the device of a kind of online migration for physical server, wherein, described device includes：
Subregion synchronization module, is configured as obtaining the partition information of disk, and the partition information is sent into another physics clothes
Business device, wherein, the partition information is the information of the subregion on disk；
Data simultaneous module, is configured as will be stored in address of the data and the data on the disk on the disk
It is sent to another physical server；
Dormancy control module, is configured as control operation system and enters resting state, so that internal storage data is saved in the disk；
Internal storage data synchronization module, is configured as the internal storage data preserved on the disk and the internal storage data in the magnetic
Address on disk is sent to another physical server,
Wherein, while subregion synchronization module and data simultaneous module perform operation, the physical server is all the time in just
Normal running status.
10. the device of a kind of online migration for physical server, wherein, described device includes：
Subregion synchronization module, is configured as receiving the partition information that another physical server sends, and according to the partition information
Subregion is carried out to disk；
Data simultaneous module, is configured as receiving the storage of another physical server transmission in another physical server
Disk on data and the data address, and the data are write the phase of the disk according to the address of the data
At address；
Internal storage data synchronization module, is configured as receiving described another physical server that another physical server sends
The address of internal storage data and the internal storage data, and the internal storage data is write by the magnetic according to the address of the internal storage data
At the identical address of disk；
Internal memory recovery module, is configured as starting from the disk again and the wake-up mechanism by operating system will be stored in institute
State during the internal storage data in disk returns to internal memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611085506.2A CN106775846B (en) | 2016-11-30 | 2016-11-30 | Method and device for online migration of physical server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611085506.2A CN106775846B (en) | 2016-11-30 | 2016-11-30 | Method and device for online migration of physical server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106775846A true CN106775846A (en) | 2017-05-31 |
| CN106775846B CN106775846B (en) | 2021-02-23 |
Family
ID=58913910
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611085506.2A Active CN106775846B (en) | 2016-11-30 | 2016-11-30 | Method and device for online migration of physical server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106775846B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109308164A (en) * | 2017-07-27 | 2019-02-05 | 武汉海康存储技术有限公司 | A kind of method and apparatus of system migration |
| CN112817702A (en) * | 2021-02-25 | 2021-05-18 | 紫光云技术有限公司 | Virtual machine automation rapid cross-architecture migration method |
| CN117149095A (en) * | 2023-10-31 | 2023-12-01 | 苏州元脑智能科技有限公司 | NAS-based cluster management method, NAS-based cluster management device, computer equipment and media |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464812A (en) * | 2009-01-06 | 2009-06-24 | 北京航空航天大学 | Virtual machine migration method |
| US8239646B2 (en) * | 2007-07-31 | 2012-08-07 | Vmware, Inc. | Online virtual machine disk migration |
| CN103019804A (en) * | 2012-12-28 | 2013-04-03 | 中国人民解放军国防科学技术大学 | VPS (virtual private server) quick migration method for OpenVZ virtualization |
| CN103902233A (en) * | 2012-12-27 | 2014-07-02 | 中国电信股份有限公司 | Virtual machine system on-line storage migration method and device and server |
| CN105138394A (en) * | 2015-09-21 | 2015-12-09 | 上海爱数软件有限公司 | Method for thermally migrating K virtual machine (KVM) and system |
| CN105718377A (en) * | 2014-12-05 | 2016-06-29 | 华为技术有限公司 | Method and apparatus for copying magnetic disk data in virtual application |
-
2016
- 2016-11-30 CN CN201611085506.2A patent/CN106775846B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8239646B2 (en) * | 2007-07-31 | 2012-08-07 | Vmware, Inc. | Online virtual machine disk migration |
| CN101464812A (en) * | 2009-01-06 | 2009-06-24 | 北京航空航天大学 | Virtual machine migration method |
| CN103902233A (en) * | 2012-12-27 | 2014-07-02 | 中国电信股份有限公司 | Virtual machine system on-line storage migration method and device and server |
| CN103019804A (en) * | 2012-12-28 | 2013-04-03 | 中国人民解放军国防科学技术大学 | VPS (virtual private server) quick migration method for OpenVZ virtualization |
| CN105718377A (en) * | 2014-12-05 | 2016-06-29 | 华为技术有限公司 | Method and apparatus for copying magnetic disk data in virtual application |
| CN105138394A (en) * | 2015-09-21 | 2015-12-09 | 上海爱数软件有限公司 | Method for thermally migrating K virtual machine (KVM) and system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109308164A (en) * | 2017-07-27 | 2019-02-05 | 武汉海康存储技术有限公司 | A kind of method and apparatus of system migration |
| CN109308164B (en) * | 2017-07-27 | 2021-11-09 | 杭州海康存储科技有限公司 | System migration method and device |
| CN112817702A (en) * | 2021-02-25 | 2021-05-18 | 紫光云技术有限公司 | Virtual machine automation rapid cross-architecture migration method |
| CN117149095A (en) * | 2023-10-31 | 2023-12-01 | 苏州元脑智能科技有限公司 | NAS-based cluster management method, NAS-based cluster management device, computer equipment and media |
| CN117149095B (en) * | 2023-10-31 | 2024-02-06 | 苏州元脑智能科技有限公司 | NAS-based cluster management method, NAS-based cluster management device, computer equipment and media |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106775846B (en) | 2021-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9823877B2 (en) | Virtual machine backup from storage snapshot | |
| US10379967B2 (en) | Live rollback for a computing environment | |
| US9104645B2 (en) | System and method of replicating virtual machines for live migration between data centers | |
| Zhang et al. | VMThunder: Fast provisioning of large-scale virtual machine clusters | |
| US9547605B2 (en) | Method for data backup, device and system | |
| US9436571B2 (en) | Estimating data storage device lifespan | |
| US9424156B2 (en) | Identifying a potential failure event for a data storage device | |
| US20140032959A1 (en) | System and Method of Replicating Virtual Machines for Live Migration Between Data Centers | |
| WO2019028682A1 (en) | Multi-system shared memory management method and device | |
| WO2018090606A1 (en) | Data storage method and device | |
| CN105205143A (en) | File storage and processing method, device and system | |
| CN111026510A (en) | Cross-storage thermal migration method for virtual machine in cloud environment | |
| US10298709B1 (en) | Performance of Hadoop distributed file system operations in a non-native operating system | |
| US9645841B2 (en) | Dynamic virtual machine storage usage monitoring, provisioning, and migration | |
| CN106775846A (en) | For the method and device of the online migration of physical server | |
| CN112052230A (en) | Multi-machine room data synchronization method, computing equipment and storage medium | |
| EP3625683B1 (en) | System and method for load balancing backup data | |
| US20150212847A1 (en) | Apparatus and method for managing cache of virtual machine image file | |
| US20210042322A1 (en) | System and method of time-based snapshot synchronization | |
| US9766677B2 (en) | Cascading startup power draws of enclosures across a network | |
| US9430152B2 (en) | Data device grouping across data storage device enclosures for synchronized data maintenance | |
| US20130318102A1 (en) | Data Handling in a Cloud Computing Environment | |
| US20180365110A1 (en) | Application-aware database backups | |
| Sun et al. | An approach to minimizing downtime induced by taking live snapshot of virtual cluster | |
| US10409618B2 (en) | Implementing VM boot profiling for image download prioritization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |