CN106055431B - The method and device of backup and recovery based on the online rollback of snapshot - Google Patents
The method and device of backup and recovery based on the online rollback of snapshot Download PDFInfo
- Publication number
- CN106055431B CN106055431B CN201610368761.1A CN201610368761A CN106055431B CN 106055431 B CN106055431 B CN 106055431B CN 201610368761 A CN201610368761 A CN 201610368761A CN 106055431 B CN106055431 B CN 106055431B
- Authority
- CN
- China
- Prior art keywords
- snapshot
- data
- rollback
- source book
- data block
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of method and devices of the backup and recovery based on the online rollback of snapshot, when user creates snapshot, source book are carried out division addressing at a block level, and distribute the COW information that a COW metadata set is used to record source book；When the instruction for receiving snapshot rollback, first snapshot agent is notified to carry out brushing dirty processing, notify snapshot module after the completion, starts back-end data rollback thread；Rollback thread in backstage is according to the COW information of metadata set, by the data copy on snapped volume to source book；When source book there are read-write requests in rollback process, needs whether to be COW and current rollback progress according to the data block number of read-write source book, redirect reading and delay write-in.The data backup at some time point of source book may be implemented in the present invention, by the online rollback technology of offer, need not interrupt the data access of source book, so as to so that the data on source book are promptly restored to available Snapshot time point and give host application offer service.
Description
Technical field
The present invention relates to a kind of method and devices of the backup and recovery based on the online rollback of snapshot, belong to storage and hold
Calamity and redundancy technique field.
Background technology
With the fast development of computer technology, digital information gradually replaces paper media, and rapid growth is presented
Situation.Mass data has expedited the emergence of the development of cloud computing, big data and mass storage system (MSS), to the reliable of storage device and data
Property, the demand of safety are also higher and higher.There is the industry of higher requirement to Information Security and business continuance at some,
Such as government, bank securities, electric power, communication, must be requested that storage system provides online data protection.
Snapshot (Snapshot) is a kind of effective ways for preventing storage device data from losing and destroying, can be quickly real
The backup and recovery of existing online data.Almost all of senior middle school's low side storage system, snapshot are all an indispensable standard configurations
Function.Snapshot can provide online data on-demand backup and recovery, when application and trouble or file corruption occur for storage device
Fast data recovery can be carried out, to restore data to some available time dotted state.In addition to the function of Backup and Restore, soon
According to also a data access channel is provided for storage user, when former data carry out application on site processing, user can access
Snapshot data can test snapshot data, the work such as data mining.
Copy on write (Copy On Write, COW) is a kind of most common realization technology of snapshot, i.e., in first time to source
When volume is written data or updates the data, first check whether corresponding data block had been COW, if it has not, then first by source
Then data copy on volume on the data block address new data is written again to snapshot space.As shown in Figure 1, snapshot is empty
Between itself be also storage resource a part, former data to be modified are needed in source book for storing.
For some purposes, such as system is by poisoning intrusion, and file is destroyed, and needs source book reverting to some and can be used
Time point data when, it is necessary to use snapshot rollback function, i.e., by the write back data on snapped volume to source book.Usually exist
During rollback, source book can not externally provide service.And the time length of rollback depends on source book after snapshot creates
Data variation amount, if data variation is very big, the process of rollback may need to grow very much.RTO(Recovery Time
Objective, the tolerable time span interrupted of recovery time target, i.e. system) it is storage one very important finger of product
Mark, theoretically the shorter this time the better.It is an object of the invention to provide a kind of methods of online rollback, are returned when initiating snapshot
When rolling, source book do not need or just need the very short time can provide read-write service.
Invention content
Goal of the invention：In the data for needing for source book to be restored to some Snapshot time point, it is common practice to which host first stops
The only data access of source book waits all data copies on snapped volumes to complete then by the write back data on snapped volume to source book,
Just notice host decontrols the data access of source book, and the time that source book interrupts service arrives depending on source book at the time point for creating snapshot
Data variation amount between snapshot rollback time point, variable quantity is bigger, and the time that source book interrupts is longer.The present invention is directed to this
Defect provides a kind of data back up method and device based on the online rollback of snapshot, and source book is made to be remained able to during snapshot rollback
It is normally carried out reading and writing data and backup, is not necessarily to interrupting service.
Technical solution：A method of the backup and recovery based on the online rollback of snapshot, specific implementation are as follows：
When user creates snapshot, source book is subjected to division addressing at a block level, and a COW member is distributed to the snapshot created
Data set, each metadata represent a data block of source book, and metadata set is ranked up by source volume data block number, described
Data block be snapshot data access base unit.When first time some data block of source book being written, need pair
The data block executes COW operations, i.e., first by data copy old on source book to snapshot space, the snapshot space is that storage provides
The part in source is reserved when source book enables snapshot, and the size in space can arbitrarily extend, to work as source volume data variation
It can be expanded when measuring big.On source book after old data copy to snapshot space, a metadata is generated, by the data block of source book
Number, the data block of snapshot space number and data write time three parts information are constituted.Subsequently to the same data of source book
There is no need to execute COW operations again when block is written.All metadata for being COW constitute the metadata set of snapshot.
After receiving the instruction of snapshot rollback, first snapshot agent is notified to carry out brushing dirty processing, the snapshot agent category
It in a part for storage system, is mounted in host operating system, effect is when creating snapshot, snapshot rollback by host system
Disk is written in data in caching, this operating process is dirty brushing, because usually operating system can reserve certain spatial cache,
The partial data of user be written directly in system cache, and this partial data may be created still or rollback Snapshot time point it
The purpose of preceding data, snapshot agent is ensuring that the data that snapshot comes out are consistent with the data at user's desired time point.
It notifies snapshot module, backstage to start rollback thread after the completion of snapshot agent brush is dirty, that is, notifies that user's snapshot rollback success, user are
Can data access be carried out to source book immediately.
Rollback thread in backstage first obtains the COW metadata sets of snapshot, each first number is read by the sequence of source volume data block
According to for doing the source volume data block of COW, Data writing time and snapshot rollback time first being judged, if being later than snapshot rollback
Time need not then carry out data copy.Otherwise it is numbered according to the snapshot space data block preserved in metadata, from snapshot space
Data are read, and the corresponding source book space of the source volume data block number preserved in metadata is written.Current data block is executed
Write back data then takes out next metadata object from COW metadata sets, the institute until having traversed snapshot COW metadata sets
There is metadata object.The write back data of a data block is often executed, then using current data block as the progress of snapshot rollback.
It when source book has read data request in rollback process, is numbered according to the data block for reading source book, searches the COW members of snapshot
Data set is directly read from source book if corresponding data block was not COW or data block number is less than rollback progress
It takes.Otherwise it is numbered according to the snapshot space data block preserved in metadata, data is read from snapshot space.If in metadata set
There is no the corresponding metadata objects of the data block, then it represents that data are read also on source book directly from source book.If
It includes multiple data blocks to need the data read, then needs to judge each data block, waits all data blocks from snapshot sky
Between or source book read after the completion of merge.
When source book has write-in request of data in rollback process, is numbered according to the data block of write-in source book, search snapshot
COW metadata sets, if corresponding data block had been COW, and data block number to be written is less than rollback progress, then
Data can write direct source book；Otherwise data rewind first is carried out to the data block, the data write-in source being then written into again
Volume, and it is the current time in system to update Data writing time in metadata object.If the data block number of source book to be written is not
It was COW, then snapshot COW directly is executed to the data block, and generate a new COW metadata object, more new metadata pair
As middle source volume data block number and snapshot space data block number, the time of data write-in is the current time in system.
The device of backup and recovery based on the online rollback of snapshot, including：
Snapshot configuration processing module, is allocated snapshot space and manages, and handles the establishment snapshot of user, deletes snapshot
And the request of SNAPSHOT INFO change.When creating snapshot, data field is established for snapshot, distribution snapshot unique mark and record create
Snapshot time is established the COW metadata sets of the snapshot, is initially empty.When deleting snapshot, if appointing there is currently snapshot rollback
Business, then prompt the snapshots rollback tasks such as user needs to complete or stop snapshot rollback task, otherwise directly removes and creates snapshot
The relative data area of Shi Jianli.
Snapshot rollback processing module handles the snapshot rollback order of user, starts back-end data write-back thread, establishes rollback
Data field records the progress of rollback, rollback starts time and end time.User can be also handled simultaneously to stop, suspending rollback times
The order of business.
Source volume data reads and writes processing module, and receiving host asks the reading and writing data of source book.Root is needed for data write-in
According to the metadata set of snapshot to determine whether needing execution COW, if snapshot is carrying out data rewind, also need to basis time
Degree of rolling into decides whether to first carry out write back data.For reading source book request, if snapshot is carrying out data rewind,
Also need to according to whether the data block for reading source book was COW come determination data be read from snapshot space reading or source book.
If snapshot is not in rollback process, data are directly read from source book.
Snapshot reads processing module, read data request of the receiving host to snapshot.It is numbered, is searched according to the data block of read request
The COW metadata sets of snapshot, if corresponding data block had been COW, according to the snapshot space number preserved in metadata
According to block number, data are read from snapshot space.If the corresponding metadata object of the data block, table are not present in metadata set
Registration is directly read from source book according to also on source book.Data if necessary to reading include multiple data blocks, then need to every
A data block is judged, is merged after the completion of waiting all data blocks to be read from snapshot space or source book, is closed to host response
Data after and.
Description of the drawings
The memory system architecture figure that Fig. 1 is applicable in for the present invention；
Fig. 2 is snapshot rollback flow of the present invention；
Fig. 3 is the write back data flow of rollback thread in backstage of the present invention；
Source book during Fig. 4 is snapshot rollback of the embodiment of the present invention reads flow；
The data block source book for not being COW during Fig. 5 is snapshot rollback of the embodiment of the present invention writes flow；
The data block source book for being COW and being more than rollback progress during Fig. 6 is snapshot rollback of the embodiment of the present invention writes stream
Journey.
Specific implementation mode
With reference to specific embodiment, the present invention is furture elucidated, it should be understood that these embodiments are merely to illustrate the present invention
Rather than limit the scope of the invention, after having read the present invention, various equivalences of the those skilled in the art to the present invention
The modification of form falls within the application range as defined in the appended claims.
The memory system architecture schematic diagram that Fig. 1 is applicable in by the embodiment of the present invention, including host and storage device, host packet
The host of data service containing I/O and system management memory host.Wherein I/O data services host may include that multiple data services are answered
With and snapshot agent, data service apply such as database.Snapshot agent is used for and storage device is combined and creates snapshot and fast
According to rollback.Storage device includes storage control and memory space, Cache etc..Storage control is for controlling data
Storage and the snapshot functions for realizing data, memory space include source book memory space and snapshot storage space, the former is for storing
The current data of source book, the latter are that system is the memory space that snapshot is reserved, and the physical medium of memory space is usually hard disk, magnetic
Disk etc..Only a kind of structural schematic diagram of storage system, actual storage system can have more hosts or multiple shown in Fig. 1
Storage device, multiple main frames constitute application system.
When user creates snapshot, source book is subjected to division addressing at a block level, and a COW member is distributed to the snapshot created
Data set, each metadata represent a data block of source book, and metadata set is ranked up by source volume data block number, described
Data block be snapshot data access base unit.When first time some data block of source book being written, need pair
The data block executes COW operations, i.e., first by data copy old on source book to snapshot space, the snapshot space is that storage provides
The part in source is reserved when source book enables snapshot, and the size in space can arbitrarily extend, to work as source volume data variation
It can be expanded when measuring big.On source book after old data copy to snapshot space, a metadata is generated, by the data block of source book
Number, the data block of snapshot space number and data write time three parts information are constituted.Subsequently to the same data of source book
There is no need to execute COW operations again when block is written.All metadata for being COW constitute the metadata set of snapshot.
Existing snapshot rollback technology is required for suspending the data service on source book during rollback, and the time suspended
Length depends on the data variation amount of source book after snapshot creates, if data variation is very big, the process of rollback may need
It is very long.RTO (Recovery Time Objective, the tolerable time span interrupted of recovery time target, i.e. system) is
One very important index of product is stored, theoretically the shorter this time the better.In order to solve above-mentioned problem of the prior art,
Several embodiments of the present invention provide the technology of the online rollback of snapshot, and during snapshot rollback, source book need not suspend operation i.e.
The service of reading and writing data can be provided.Illustrate the flow of the online rollback of snapshot in conjunction with Fig. 2 and Fig. 3.
After storage device receives the instruction of snapshot rollback of managing main frame, first notify I/O hosts on snapshot agent into
Row brushes dirty processing, and by the data cached write-in disk in application system, the snapshot module of storage device is notified after the completion of brush is dirty, after
Platform starts rollback thread, initialization rollback data field, when the data field is mainly progress and the rollback startup for recording rollback
Between, wherein rollback progress indicates which data block has completed the write back data from snapshot space to source book；Rollback starts the time,
For judging whether the data block of source book needs to carry out write back data.After completing data field initialization, user can be immediately to source
Volume carries out data access.
Fig. 3 show the background process thread of snapshot rollback, and wherein progressDn represents rollback progress, indicates
Data block before ProgressDn has completed write back data.StartTime indicates the time of rollback task start.First obtain
The COW metadata sets of snapshot read each metadata by the sequence of source volume data block, the metadata by source book data
Block number, the data block number of snapshot space and data write time three parts information are constituted.For doing the source volume data of COW
Block, Dx represent source volume data block number, and dx represents data block number of the snapshot data of source book Dx data blocks in snapshot space,
After timex represents establishment Snapshot time point, the time of source book Dx data block datas change.Backstage rollback thread reads some
Metadata first judges data change time timex and snapshot rollback time startTime, if timex is more than startTime,
Write back data is first completed in other flow for expression or Dx data blocks change during rollback, then skips the number
According to block, next metadata object is taken from metadata set.Otherwise according to the snapshot space data block number dx preserved in metadata,
Data are read from snapshot space, and the corresponding source book spaces source volume data block number Dx are written.The number of current data block is executed
According to write-back, then next metadata object is taken out from COW metadata sets, until having traversed all of snapshot COW metadata sets
Metadata object.The write back data of a data block is often executed, then using current data block as the progress of snapshot rollback
progressDn。
During rollback, when the application on I/O data service hosts initiates read data request to source book, as shown in figure 4, needing
D2 and D3 data blocks on source book are read, the corresponding metadata objects of D2 are not present in metadata set, are indicated after creating snapshot, D2
Data block do not change, then data are directly read from source book；There are the corresponding metadata of D3 data blocks in metadata set
Object, and its snapshot data is stored in the corresponding snapshot storage spaces of d8.If time3 is less than snapshot rollback task start
Between startTime, then data read from snapshot space, it is shown in solid such as Fig. 4, otherwise directly read from source book, such as the void of Fig. 4
Shown in line.After the completion of two parts data are all read, after being merged according to the size of read data request, application is returned to, is completed
Source book read request during snapshot rollback.
During snapshot rollback, when the application on I/O data service hosts initiates write data requests to source book, such as Fig. 5 institutes
Show, step 501 host initiates write request D2` to source book D2 data blocks, and the corresponding metadata pair of D2 is not present in metadata set
As indicating after creating snapshot, the data block of D2 changes for the first time, needs the corresponding old data D2 of source book D2 data blocks is first
It copies snapshot space to, distributes a new data block d2 from snapshot space first, then execute step 502, D2 is copied to
Snapshot space, after the completion of copy, executes step 503, and source book D2 data blocks are written in new data, after the completion of write-in, generates one
A new metadata object, source volume data block number are D2, and snapshot space number is d2, Data writing time time2, is executed
New metadata object is inserted into the metadata set of snapshot by step 504, such as the corresponding metadata object of Fig. 5 dotted line frames.
Fig. 5 show the case where D2 data blocks were not COW, and also a kind of situation is that D2 had been COW, such as Fig. 6 institutes
Show, step 601, the data block D3 data changes of source book are D3` by tim3` moment, host application, and snapshot metadata is concentrated
Through there are the corresponding metadata objects of D3, but the rollback progress progressDn of snapshot is D2 at this time, is less than D3, it is therefore desirable to first
Step 602 is executed, by the D3 data blocks of the corresponding write back datas of snapshot space d8 to source book, step 603 is then executed, will apply
The D3 data blocks of new data D3` write-in source books repair the Data writing time of the corresponding metadata objects of D3 after completing write-in
It is changed to time3`.When rollback thread goes to the write-back of D3 data blocks, judge that time3` is more than the rollback task start time
StartTime then skips the data block.
The device of backup and recovery based on the online rollback of snapshot, including：
Snapshot configuration processing module, is allocated snapshot space and manages, and handles the establishment snapshot of user, deletes snapshot
And the request of SNAPSHOT INFO change.When creating snapshot, data field is established for snapshot, distribution snapshot unique mark and record create
Snapshot time is established the COW metadata sets of the snapshot, is initially empty.When deleting snapshot, if appointing there is currently snapshot rollback
Business, then prompt the snapshots rollback tasks such as user needs to complete or stop snapshot rollback task, otherwise directly removes and creates snapshot
The relative data area of Shi Jianli.
Snapshot rollback processing module handles the snapshot rollback order of user, starts back-end data write-back thread, establishes rollback
Data field records the progress of rollback, rollback starts time and end time.User can be also handled simultaneously to stop, suspending rollback times
The order of business.
Source volume data reads and writes processing module, and receiving host asks the reading and writing data of source book.Root is needed for data write-in
According to the metadata set of snapshot to determine whether needing execution COW, if snapshot is carrying out data rewind, also need to basis time
Degree of rolling into decides whether to first carry out write back data.For reading source book request, if snapshot is carrying out data rewind,
Also need to according to whether the data block for reading source book was COW come determination data be read from snapshot space reading or source book.
If snapshot is not in rollback process, data are directly read from source book.
Snapshot reads processing module, read data request of the receiving host to snapshot.It is numbered, is searched according to the data block of read request
The COW metadata sets of snapshot, if corresponding data block had been COW, according to the snapshot space number preserved in metadata
According to block number, data are read from snapshot space.If the corresponding metadata object of the data block, table are not present in metadata set
Registration is directly read from source book according to also on source book.Data if necessary to reading include multiple data blocks, then need to every
A data block is judged, is merged after the completion of waiting all data blocks to be read from snapshot space or source book, is closed to host response
Data after and.
According to the specific embodiment of the invention, the data backup at some time point of source book may be implemented, the online of offer is provided
Rollback technology need not interrupt the data access of source book, so as to so that source book be promptly restored to available Snapshot time point to
Host application offer service.
Claims (4)
1. a kind of method of the backup and recovery based on the online rollback of snapshot, which is characterized in that specific implementation is as follows：
Snapshot is created, source book is subjected to division addressing at a block level, and a COW metadata set is distributed to the snapshot created, each member
Data represent a data block of source book, and metadata set is ranked up by source volume data block number；
After receiving the instruction of snapshot rollback, snapshot agent is first notified to carry out brushing dirty processing, it is logical after the completion of snapshot agent brush is dirty
Know that snapshot module, backstage start rollback thread, initialization rollback data field, the data field is mainly the progress for recording rollback
Start the time with rollback, wherein rollback progress indicates which data block has completed the write back data from snapshot space to source book；
Rollback starts the time, for judging whether the data block of source book needs to carry out write back data；After completing data field initialization, user
Can data access be carried out to source book immediately；
Rollback thread in backstage first obtains the COW metadata sets of snapshot, each metadata is read by the sequence of source volume data block,
For doing the source volume data block of COW, Data writing time and snapshot rollback time are first judged, if Data writing time is later than
Snapshot rollback time need not then carry out data copy；Otherwise it is numbered according to the snapshot space data block preserved in metadata, from
Snapshot space reads data, and the corresponding source book space of the source volume data block number preserved in metadata is written；It has executed current
The write back data of data block then takes out next metadata object from COW metadata sets, until having traversed snapshot COW member numbers
According to all metadata objects of collection；The write back data of a data block is often executed, then using current data block as snapshot rollback
Progress.
2. the method for the backup and recovery as described in claim 1 based on the online rollback of snapshot, which is characterized in that when return
When source book has read data request during rolling, is numbered according to the data block for reading source book, search the COW metadata sets of snapshot, if
Corresponding data block was not COW or data block number is less than rollback progress, then was directly read from source book；Otherwise basis
The snapshot space data block number preserved in metadata, data are read from snapshot space；If the number is not present in metadata set
According to the corresponding metadata object of block, then it represents that data are read also on source book directly from source book；If necessary to reading
Data include multiple data blocks, then need to judge each data block, and all data blocks is waited to be read from snapshot space or source book
It is merged after taking into；
It when source book has write-in request of data in rollback process, is numbered according to the data block of write-in source book, searches the COW members of snapshot
Data set, if corresponding data block had been COW, and data block number to be written is less than rollback progress, then data
Source book can be write direct；Otherwise data rewind first is carried out to the data block, the data write-in source book being then written into again, and
It is the current time in system to update Data writing time in metadata object；If the data block number of source book to be written was not done
COW then directly executes snapshot COW to the data block, and generates a new COW metadata object, updates in metadata object
The time of source volume data block number and snapshot space data block number, data write-in is the current time in system.
3. the method for the backup and recovery as described in claim 1 based on the online rollback of snapshot, which is characterized in that when with
When family creates snapshot, source book is subjected to division addressing at a block level, and a COW metadata set is distributed to the snapshot created, it is each
A metadata represents a data block of source book, and metadata set is ranked up by source volume data block number；When first time is to source book
Some data block when being written, need to execute COW operations to the data block, i.e., first by data copy old on source book to soon
According to space；On source book after old data copy to snapshot space, a metadata is generated, by the data block number of source book, snapshot
The data block number and data write time three parts information in space are constituted；When subsequently the same data block of source book being written just
COW operations need not be executed again；All metadata for being COW constitute the metadata set of snapshot.
4. a kind of device of the backup and recovery based on the online rollback of snapshot, which is characterized in that including：
Snapshot configuration processing module, is allocated snapshot space and manages, handle user establishment snapshot, delete snapshot and
The request of SNAPSHOT INFO change；When creating snapshot, data field is established for snapshot, distribution snapshot unique mark and record create snapshot
Time establishes the COW metadata sets of the snapshot, is initially empty；When deleting snapshot, if there is currently the task of snapshot rollback,
It then prompts user that equal snapshots rollback task is needed to complete or stop snapshot rollback task, otherwise directly removes the when of creating snapshot and build
Vertical relative data area；
Snapshot rollback processing module handles the snapshot rollback order of user, starts back-end data write-back thread, establishes rollback data
Area records the progress of rollback, rollback starts time and end time；User can be also handled simultaneously to stop, suspending rollback task
Order；
Source volume data reads and writes processing module, and receiving host asks the reading and writing data of source book；Data are written and are needed according to fast
According to metadata set to determine whether need execute COW, if snapshot is carrying out data rewind, also need to according to rollback into
It spends to decide whether to first carry out write back data；For reading source book request, if snapshot is carrying out data rewind, also need
COW whether was according to the data block for reading source book come determination data is read from snapshot space reading or source book；If
Snapshot is not in rollback process, then data are directly read from source book；
Snapshot reads processing module, read data request of the receiving host to snapshot；It is numbered according to the data block of read request, searches snapshot
COW metadata sets, if corresponding data block had been COW, according to the snapshot space data block preserved in metadata
Number reads data from snapshot space；If the corresponding metadata object of the data block is not present in metadata set, then it represents that number
According to also on source book, directly being read from source book；Data if necessary to reading include multiple data blocks, then need to every number
Judged according to block, is merged after the completion of waiting all data blocks to be read from snapshot space or source book, after merging to host response
Data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610368761.1A CN106055431B (en) | 2016-05-30 | 2016-05-30 | The method and device of backup and recovery based on the online rollback of snapshot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610368761.1A CN106055431B (en) | 2016-05-30 | 2016-05-30 | The method and device of backup and recovery based on the online rollback of snapshot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106055431A CN106055431A (en) | 2016-10-26 |
| CN106055431B true CN106055431B (en) | 2018-09-21 |
Family
ID=57175053
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610368761.1A Active CN106055431B (en) | 2016-05-30 | 2016-05-30 | The method and device of backup and recovery based on the online rollback of snapshot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106055431B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106814972B (en) * | 2016-12-22 | 2018-04-17 | 北京华云网际科技有限公司 | The rollback method and device of the snapshot node of distributed block storage |
| CN108958968B (en) * | 2017-05-17 | 2023-12-22 | 中兴通讯股份有限公司 | File processing method and device |
| CN107291579B (en) * | 2017-07-05 | 2020-03-31 | 郑州云海信息技术有限公司 | Metadata cluster file snapshot method based on static subtree partition |
| CN107391307A (en) * | 2017-07-27 | 2017-11-24 | 郑州云海信息技术有限公司 | The method of testing and device of storage area network storage device snapshot functions |
| CN108984779A (en) * | 2018-07-25 | 2018-12-11 | 郑州云海信息技术有限公司 | Distributed file system snapshot rollback metadata processing method, device and equipment |
| CN109324926A (en) * | 2018-08-30 | 2019-02-12 | 黄疆 | A kind of junk data processing system and method based on rollback |
| CN109324929B (en) * | 2018-09-17 | 2021-11-09 | 郑州云海信息技术有限公司 | Snapshot creating method, device and equipment and readable storage medium |
| CN109189614A (en) * | 2018-10-19 | 2019-01-11 | 郑州云海信息技术有限公司 | A kind of snapshot rollback method and device |
| CN109271367A (en) * | 2018-12-07 | 2019-01-25 | 浪潮(北京)电子信息产业有限公司 | Distributed file system multinode snapshot rollback method and system |
| CN111324295B (en) * | 2018-12-17 | 2023-09-19 | 中国移动通信集团四川有限公司 | Data migration method, device, equipment and medium |
| CN109977092B (en) * | 2019-03-26 | 2021-06-08 | 武汉深之度科技有限公司 | Method and computing device for copying files |
| CN110502187B (en) * | 2019-07-09 | 2020-12-04 | 华为技术有限公司 | Snapshot rollback method and device |
| CN110795282B (en) * | 2019-10-18 | 2022-10-25 | 苏州浪潮智能科技有限公司 | Data rollback method, device, equipment and storage medium |
| CN111552437B (en) * | 2020-04-22 | 2024-03-15 | 上海天玑科技股份有限公司 | Snapshot method and snapshot device applied to distributed storage system |
| CN112416860B (en) * | 2020-11-16 | 2024-04-26 | 平安科技（深圳）有限公司 | Data rollback updating method, device, computer equipment and storage medium |
| CN112579358B (en) * | 2020-12-22 | 2024-03-22 | 深圳市科力锐科技有限公司 | Backup point detection method, device, equipment and storage medium |
| CN113342751B (en) * | 2021-07-30 | 2021-11-09 | 联想凌拓科技有限公司 | Metadata processing method, device, equipment and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704903A (en) * | 2004-06-03 | 2005-12-07 | 华为技术有限公司 | Snapshot backup method |
| CN102779080A (en) * | 2012-06-18 | 2012-11-14 | 上海爱数软件有限公司 | Method for generating snapshot, method and device for data recovery by using snapshot |
| CN104407935A (en) * | 2014-11-07 | 2015-03-11 | 华为数字技术（成都）有限公司 | Snapshot rollback method and storage equipment |
| CN104598293A (en) * | 2015-01-06 | 2015-05-06 | 杭州华三通信技术有限公司 | Online snapshot method and equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9842032B2 (en) * | 2013-08-27 | 2017-12-12 | Red Hat, Inc. | Memory first live snapshot |
-
2016
- 2016-05-30 CN CN201610368761.1A patent/CN106055431B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704903A (en) * | 2004-06-03 | 2005-12-07 | 华为技术有限公司 | Snapshot backup method |
| CN102779080A (en) * | 2012-06-18 | 2012-11-14 | 上海爱数软件有限公司 | Method for generating snapshot, method and device for data recovery by using snapshot |
| CN104407935A (en) * | 2014-11-07 | 2015-03-11 | 华为数字技术（成都）有限公司 | Snapshot rollback method and storage equipment |
| CN104598293A (en) * | 2015-01-06 | 2015-05-06 | 杭州华三通信技术有限公司 | Online snapshot method and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106055431A (en) | 2016-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106055431B (en) | The method and device of backup and recovery based on the online rollback of snapshot | |
| JP5346536B2 (en) | Information backup / restore processing device and information backup / restore processing system | |
| US7895394B2 (en) | Storage system | |
| JP4292882B2 (en) | Plural snapshot maintaining method, server apparatus and storage apparatus | |
| CN103761053B (en) | A kind of data processing method and device | |
| US20060047926A1 (en) | Managing multiple snapshot copies of data | |
| CN105718217B (en) | A kind of method and device of simplify configuration storage pool data sign processing | |
| EP2879040B1 (en) | Data storage method, data storage apparatus, and storage device | |
| KR101384438B1 (en) | Capturing and naming dynamic storage tiering configurations to support data pre-staging | |
| US9778860B2 (en) | Re-TRIM of free space within VHDX | |
| JP2008015769A (en) | Storage system and writing distribution method | |
| JP2005301497A (en) | Storage management system, restoration method and its program | |
| JP2008015768A (en) | Storage system and data management method using the same | |
| KR20120140652A (en) | Buffer disk in flashcopy cascade | |
| US20170302734A1 (en) | Cloud Computing Service Architecture | |
| CN103761190A (en) | Data processing method and apparatus | |
| JP4561168B2 (en) | Data processing system and method, and processing program therefor | |
| KR20000022716A (en) | Efficient volume copy using pre-configuration of log structured target storage | |
| JP2006011811A (en) | Storage control system and storage control method | |
| US20220091746A1 (en) | Storage device and storage control method | |
| CN109753222A (en) | Data-erasure method, server and computer-readable medium | |
| CN110019086A (en) | More copy read methods, equipment and storage medium based on distributed file system | |
| US20050223180A1 (en) | Accelerating the execution of I/O operations in a storage system | |
| JP7304215B2 (en) | Storage system and history information management method | |
| JP2010152781A (en) | Backup server device, backup/restore program, and backup/restore method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |