CN105843704B - A kind of data guard method and system of the snapshot functions of combination distributed block storage - Google Patents
A kind of data guard method and system of the snapshot functions of combination distributed block storage Download PDFInfo
- Publication number
- CN105843704B CN105843704B CN201610146907.8A CN201610146907A CN105843704B CN 105843704 B CN105843704 B CN 105843704B CN 201610146907 A CN201610146907 A CN 201610146907A CN 105843704 B CN105843704 B CN 105843704B
- Authority
- CN
- China
- Prior art keywords
- snapshot
- server
- backup
- production server
- data
- 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
- G06F11/1415—Saving, restoring, recovering or retrying at system level
- G06F11/1435—Saving, restoring, recovering or retrying at system level using file system or storage system metadata
-
- 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
- G06F11/1458—Management of the backup or restore process
- G06F11/1469—Backup restoration techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Retry When Errors Occur (AREA)
Abstract
The present invention provides a kind of data guard method and system of the snapshot functions of combination distributed block storage, and ceph block devices are directly or indirectly mounted in production server；Backup client is disposed in production server, and the I/O drivers for realizing back-up processing between production server book and ceph block devices are installed；The mirror of production server book and ceph block devices is established by backup server；After starting snapshot task, the generated data of the application of production server are written in the mirrored volume on production server book and ceph block devices simultaneously, and a snapshot is executed to the mirrored volume on ceph block devices, Snapshot time point and SNAPSHOT INFO are recorded to backup server, subsequently to restore.The data guard method and system of the snapshot functions of the combination distributed block storage of the present invention are capable of providing the Backup Data of 99.9999% or more degree of reliability, and ensure that these data can be resumed out at any time, provide data safety guarantee to the user.
Description
Technical field
The present invention relates to the technical fields that data backup and computerized information store, more particularly to a kind of combination distribution
The data guard method and system of the snapshot functions of block storage.
Background technology
With the arriving of cloud computing era, IT (Information Technology, Internet technology) service shows
Scale, clustering, and supporting the subsequent storage device of cloud framework is provided by each storage manufacturer, data capacity was from the past
It is even higher that TB (1TB=1024GB) grade rises to PB (1PB=1024PB) grade.Therefore, come ensure system availability and can
Gradually being suffered from a problem that by traditional data backup mode of property exactly can not be within the effective time to such flood tide
Data are protected.Then, a kind of new number based on cloud computing framework facility is being thought deeply by each Data Protection Scheme provider
The timeliness of data protection is improved according to protection, such as in conjunction with storage snapshot.
Storage networking industry association (StorageNetworking Industry Association, SNIA) is to snapshot
(Snapshot) definition is：About a completely available copy of specified data acquisition system, which includes corresponding data at certain
The image of a time point (time point that copy starts).Snapshot can be a copy of the data represented by it, can also be
One duplicate of data.The effect of snapshot can mainly carry out online data backup and restore.When storage device is answered
With quick data recovery can be carried out when failure or file corruption, data are restored to the state at some available time point.
Another effect of snapshot is to provide another data access channel for storage user.When former data carry out at application on site
When reason, user can access snapshot data, can also carry out the work such as testing using snapshot.Therefore, storage snapping technique makes
The protection of big data is more efficient, high practicability.
However, above-mentioned backup scenario has the following disadvantages：
1) the problem of the availability and reliability and retractility of storage itself
The storage volume or storage itself that snapshot is relied on have failed, and snapshot cannot also recover.For this purpose,
Certain backup manufacturers provide the function of being backed up to snapshot data, but this seems and walks back the mode of conventional backup recovery.
2) the dilatation problem that storage itself is faced
If snapshot restores data by real clone's (the entire image of snapshot all copies in a new storage volume), restore
The memory space relied on is also to be doubled and redoubled.
3) problem of various complexity can be also faced in the computer room of complicated pluralism
It relies solely on and stores the reliabilty and availability of itself and can not meet high request to production server safety.
Invention content
In view of the foregoing deficiencies of prior art, the purpose of the present invention is to provide a kind of storages of combination distributed block
The data guard method and system of snapshot functions intercept and capture the snapshot functions for rewriteeing and being stored with distributed block in conjunction with the I/O of driving layer
Come carry out large-scale computer room and cloud computing platform mass data backup and restore problem, to provide almost fault-free, support
High concurrent and the data storage for simply increasing very much clustered node, provide data safety guarantee to the user.
In order to achieve the above objects and other related objects, the present invention provides a kind of snapshot functions of combination distributed block storage
Data guard method, include the following steps：Ceph block devices are directly mounted in production server；In production server
Backup client is disposed, and the I/O drivers for realizing back-up processing between production server book and ceph block devices are installed；
The mirror of production server book and ceph block devices is established by backup server；After starting snapshot task, lead to
Crossing backup client after the mirrored volume in the data copy of production server book to ceph block devices, will start I/O drivings
The generated data of the application of subsequent production server are written on production server book and ceph block devices program simultaneously
Mirrored volume in, and a snapshot is executed to the mirrored volume on ceph block devices, records Snapshot time point and SNAPSHOT INFO to standby
Part server；When carrying out recovery data, backup server obtains Snapshot time point, reinforms the backup client of new production server
End executes the corresponding snapshot of the Snapshot time point and snapshot is cloned a new volume after primary protection operates, and finally will newly be wrapping with
The data that new production server is loaded onto to be restored.
According to the data guard method of the snapshot functions of above combination distributed block storage, wherein：Start snapshot task
Afterwards, the application of production server is in silent status；After snapshot task, the application release silent status of production server.
Correspondingly, the present invention also provides a kind of data protection systems of the snapshot functions of combination distributed block storage, including
Carry module, deployment module, backup module and recovery module；
The carry module is used to ceph block devices being directly mounted to production server；
The deployment module is installed for disposing backup client in production server and realizes production server data
The I/O drivers of back-up processing between volume and ceph block devices；
The backup module is used to establish the mirror image of production server book and ceph block devices by backup server
Relationship；After starting snapshot task, by backup client by the data copy of production server book to ceph block devices
On mirrored volume after, start I/O drivers, by the generated data of the application of subsequent production server simultaneously be written production clothes
In mirrored volume in device book of being engaged in and ceph block devices, and a snapshot, record are executed to the mirrored volume on ceph block devices
Snapshot time point and SNAPSHOT INFO are to backup server；
The recovery module is used for when carrying out recovery data, is obtained Snapshot time point by backup server, is reinformed
The backup client of new production server executes the corresponding snapshot of the Snapshot time point and clones snapshot after primary protection operates
Go out a new volume, finally will newly be wrapping with the data for being loaded onto new production server to be restored.
According to the data protection system of the snapshot functions of above combination distributed block storage, wherein：The backup module
In, after starting snapshot task, the application of production server is in silent status；After snapshot task, production server is answered
With release silent status.
In addition, the present invention also provides a kind of data guard methods of the snapshot functions of combination distributed block storage, including with
Lower step：
By ceph block device carries to backup server, then by backup server by the ceph block device carries after the carry
To production server；
Backup client is disposed in production server, and is installed and realized between production server book and ceph block devices
The I/O drivers of back-up processing；
The mirror of production server book and ceph block devices is established by backup server；Appoint when starting snapshot
After business, by backup client by after the mirrored volume in the data copy of production server book to ceph block devices, start
Production server book and ceph is written simultaneously in the generated data of the application of subsequent production server by I/O drivers
In mirrored volume on block device, and a snapshot is executed to the mirrored volume on ceph block devices, records Snapshot time point and snapshot
Information is to backup server；
When carrying out recovery data, backup server notifies the backup client of new production server according to Snapshot time point
The corresponding snapshot of the Snapshot time point is executed, snapshot is cloned into a new volume after primary protection operates, finally newly will be wrapping with load
Data to new production server to be restored.
According to the data guard method of the snapshot functions of above combination distributed block storage, wherein：Start snapshot task
Afterwards, the application of production server is in silent status；After snapshot task, the application release silent status of production server.
According to the data guard method of the snapshot functions of above combination distributed block storage, wherein：Backup server will
Ceph block devices after the carry pass through iSCSI protocol or FC carries to production server；When carrying out data recovery, backup clothes
New be wrapping with is loaded onto new production server by business device by iSCSI protocol or FC.
Correspondingly, the present invention also provides a kind of data protection systems of the snapshot functions of combination distributed block storage, including
Carry module, deployment module, backup module and recovery module；
The carry module is used for ceph block device carries to backup server, then by the ceph block devices after the carry
Pass through iSCSI protocol or FC carries to production server；
The deployment module is installed for disposing backup client in production server and realizes production server data
The I/O drivers of back-up processing between volume and ceph block devices；
The backup module is used to establish the mirror image of production server book and ceph block devices by backup server
Relationship；After starting snapshot task, by backup client by the data copy of production server book to ceph block devices
On mirrored volume after, start I/O drivers, by the generated data of the application of subsequent production server simultaneously be written production clothes
In mirrored volume in device book of being engaged in and ceph block devices, and a snapshot, record are executed to the mirrored volume on ceph block devices
Snapshot time point and SNAPSHOT INFO are to backup server；
The recovery module is used for when carrying out recovery data, and backup server notifies new production according to Snapshot time point
The backup client of server executes the corresponding snapshot of the Snapshot time point and snapshot is cloned one after primary protection operates
New volume, the data for being loaded onto new production server to be restored will be newly wrapping with finally by iSCSI protocol or FC.
According to the data protection system of the snapshot functions of above combination distributed block storage, wherein：The backup module
In, after starting snapshot task, the application of production server is in silent status；After snapshot task, production server is answered
With release silent status.
According to the data protection system of the snapshot functions of above combination distributed block storage, wherein：The carry module
In, the ceph block devices after the carry are passed through iSCSI protocol or FC carries to production server by backup server；It is described extensive
In multiple module, new be wrapping with is loaded onto new production server by backup server by iSCSI protocol or FC.
As described above, the data guard method and system of the snapshot functions of the combination distributed block storage of the present invention, have
Following advantageous effect：
(1) the advantages of combining High Availabitity, the high reliability of distributed block storage, is capable of providing 99.9999% or more nothing
The Backup Data of failure rate, and ensure that these data can be resumed out at any time, and can restore in arbitrary Snapshot time point, from
And provide data safety guarantee to the user；
(2) distributed block storage easily property extending transversely is utilized, clustered node can be easily added, possess for backup
The computer room or cloud platform of mass data provide unlimited expansible memory space；
(3) combine the snapshot functions of distributed storage that can provide 2 for data in the case where ensuring using consistency
RPO and 1 minute or so the RTO times of or so minute；In conjunction with continuous data protection (Continuous Data
Protection, CDP) technology, the RPO times can be down to 20 seconds or so.
Description of the drawings
Fig. 1 is shown as the structural schematic diagram of ceph framework levels；
Fig. 2 is shown as the first embodiment of the data guard method of the snapshot functions of the combination distributed block storage of the present invention
Flow chart；
Fig. 3 is shown as the first embodiment of the data protection system of the snapshot functions of the combination distributed block storage of the present invention
Structural schematic diagram；
Fig. 4 is shown as the second embodiment of the data guard method of the snapshot functions of the combination distributed block storage of the present invention
Flow chart；
Fig. 5 is shown as the second embodiment of the data protection system of the snapshot functions of the combination distributed block storage of the present invention
Structural schematic diagram.
Component label instructions
11 carry modules
12 deployment modules
13 backup modules
14 recovery modules
21 carry modules
22 deployment modules
23 backup modules
24 recovery modules
Specific implementation mode
Illustrate that embodiments of the present invention, those skilled in the art can be by this specification below by way of specific specific example
Disclosed content understands other advantages and effect of the present invention easily.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also be based on different viewpoints with application, without departing from
Various modifications or alterations are carried out under the spirit of the present invention.
It should be noted that the diagram provided in the present embodiment only illustrates the basic conception of the present invention in a schematic way,
Then component count, shape and size when only display is with related component in the present invention rather than according to actual implementation in schema are painted
System, when actual implementation kenel, quantity and the ratio of each component can be a kind of random change, and its assembly layout kenel also may be used
It can be increasingly complex.
In the prior art, mainstream distributed block of increasing income is stored as sheepdog and ceph.
Sheepdog is a distributed objects storage system, aims at virtual machine and provides block storage, matches with no single-point, zero
Set, can linear expansion the advantages of.The R＆D team of main exploitation and guardian from group of Alibaba of sheepdog projects.
It is rich that ceph is that the Sage Weil (the joint founder of DreamHost) in the University of California branch schooles Santa Cruz are aimed at
The free software distributed file system of new generation of scholar's paper design, can easy expansion to number PB capacity, support a variety of work
The high-performance of load has high reliability, scalability, retractility.Ceph projects possess more exploitations, safeguard and use
Person has extensive use in every field, and obtains the support of a large amount of tissue of increasing income.
Recovery time target (Recovery Time Objective, RTO) refers to after disaster occurs, from IT system delay machine
Start when service pause, to IT system restore to can support each department running, restore operation when, between this 2 points when
Between section.
Recovery point objectives (Recovery Point Objective, RPO) refer to wanting real for system and application data
It is now able to restore to each department's business can be supported to operate, system and creation data should be restored to required for certain update degree
Time.This update degree can be upper one week Backup Data, can also be the real time data of last transaction.
In conjunction with the snapshot functions of every excellent properties and itself that distributed block stores, combination distributed block of the invention is deposited
The data guard method and system of the snapshot functions of storage can in the case where meeting ideal RTO and RPO easily backup and
Restore the computer room or cloud platform that possess mass data and complex environment.Available data shows that ceph can provide the fast of second grade
According to and cloning function.
The data protection of the snapshot functions of the combination distributed block storage of the present invention is mainly illustrated by taking ceph as an example below
Method and system.It should be noted that the storage of other distributed blocks is still suitable for the invention and stores in conjunction with distributed block
The data guard method and system of snapshot functions, belong to the scope of protection of the present invention within.
Fig. 1 show the structure of ceph framework levels, wherein librbd only supports linux kernel.Due to ceph officials
Client only supports linux kernel, usually supports win32 by following two approach：
(1) ceph clients are realized using the win32 that third party increases income
For example, realizing ceph clients using ceph-dokan, win2008, win2012 and win7 are supported at present.
The advantages of which is disk read-write speed block, no longer needs to do layer data forwarding again by backup cabinet, directly
It can write data into ceph clusters.But the system environments that a set of ceph clients are supported is very limited.It only supports at present
Linux system supports win32 systems reluctantly, and other systems do not support substantially, to influence ceph clients realization stabilization
Property and compatibility.
(2) the backup cabinet for combining Linux environment becomes backup by ceph blocks storage carry to backup cabinet
One disk of cabinet is utilizing Internet disc driver performance tests (Internet Small Computer
System Interface, iSCSI) or optical-fibre channel (Fibre Channel, FC) by the Disk Mapping to arbitrary client
End.
The advantages of which is to support all systems for supporting iSCSI and FC environment, convenience good.However, readwrite performance
It is relatively poor.
Illustrate the number of the snapshot functions of the combination distributed block storage of the present invention respectively below in conjunction with above two mode
According to guard method and system.
In first embodiment with reference to the data guard method of Fig. 2, the present invention combination distributed block snapshot functions stored,
Include the following steps：
Step S11, ceph block devices are directly mounted in production server.
Step S12, backup client is disposed in production server, and is installed and realized production server book and ceph
The I/O drivers of back-up processing between block device.
Step S13, the mirror of production server book and ceph block devices is established by backup server；When opening
After dynamic snapshot task, by backup client by the mirrored volume in the data copy of production server book to ceph block devices
Afterwards, start I/O drivers, production server book is written into simultaneously in the generated data of the application of subsequent production server
With in the mirrored volume on ceph block devices, and on ceph block devices mirrored volume execute a snapshot, record Snapshot time point
With SNAPSHOT INFO to backup server.
Wherein, backup server establishes the mirror of production server book and ceph block devices, and configures meter
The task of drawing puts initiation snapshot task at a reasonable time.After starting snapshot task monitoring, backup server is notified that backup visitor
Family end is formatted the mirrored volume on ceph block devices, then will be given birth to by way of disk verification or image copying again
It produces in the mirrored volume in the data copy to ceph block devices of server data volume, starts I/O drivers so that subsequent production
The generated data of application of server are written simultaneously in the mirrored volume on production server book and ceph block devices, and right
Mirrored volume on ceph block devices executes a snapshot, records Snapshot time point and SNAPSHOT INFO to backup server.Snapshot is believed
Breath includes snapshot contents, snapshot size, snapshot position etc..
Specifically, after starting snapshot task, the application of backup client notice production server is in silent status, then
Mirrored volume on I/O buffering layer data write-in ceph block devices is notified backup clothes by notice driving layer after the completion of operation to be written
Business device calls ceph client interfaces or order line to execute a snapshot to the mirrored volume.After snapshot task, production clothes
The application release silent status of business device.
Specifically, backup client calls ceph client-side interfaces or order line to the mirrored volume on the ceph block devices
Execute a snapshot.Ceph block devices execute a Snapshot time at 20 seconds or so, and I/O buffers layer data and ceph block devices are written
On mirrored volume the time required to determined by the load and buffer layer configuration strategy of production server.
In addition, Linux driving layers and windows driving layer I/O write operations distribution realization principle difference are as follows：
(1), Linux drives layer I/O write operations to distribute realization principle
A, Linux drivings are realized is driven to handle User space application monitoring programme with kernel state using netlink communication modes
Dynamic communication；
B, after Linux drive loads, initialization netlink carrys out the message of monitoring users state；
C, after backup tasks start, the claimed book of user configuration and mirror image volume information are passed into driving
Module simultaneously saves it in memory, then will be pointed by the make_request_fn of the block I/O request queue of book equipment
The function pointer of kernel acquiescence retains to memory, and the I/O processing functions that make_request_fn is directed toward to our realizations again refer to
Needle.At the same time, the write operation of I/O is cached into a queue, and will be in the data image to mirrored volume of book.
D, after mirrored volume replicates, in the write operation to above-mentioned queue for continuing to filter out I/O, team is then handled one by one
I/O write operations in row, it is distributed in book and mirrored volume.
(2), windows drives layer I/O write operations to distribute realization principle
In windows kernels, input output request packet (I/O request package, IRQ) be it is a kind of it is important with
The related data structure of input and output.When User space application program is communicated with bottom layer driving, application program can send out I/O and ask
It asks, operating system can ask I/O to be converted to corresponding IRQ data, and different IRQ can be sent according to type transfer to different
Function.Communication between client-side program and driving layer can be transmitted using DeviceIoControl methods, extend self-defined control
Code processed.
Windows drives layer I/O write operations distribution realization principle substantially identical with above-mentioned Linux drivings, only in I/O mistakes
Filter and the upper details of distribution are distinguished a little.When backup tasks start Backup Data volume, by changing the corresponding note of book equipment
The equipment is loaded filtration drive by volume table information.The corresponding distribution processor functions of IRQ_MJ_SCSI are directed toward at Driver Entry
The function pointer for the filter function that oneself is realized.
When step S14, carrying out recovery data, backup server obtains Snapshot time point, reinforms new production server
Backup client executes the corresponding snapshot of the Snapshot time point and snapshot is cloned a new volume after primary protection operates, finally
The data for being loaded onto new production server to be restored will be newly wrapping with.
Specifically, before ceph requires clone's snapshot, protection operation is executed to ensure that snapshot is not deleted, to ensure to clone
Volume is available.
When it is new be wrapping be loaded onto new production server environment when, recovered by the application on newly-generated server to take over
Data.
In first embodiment with reference to the data protection system of Fig. 3, the present invention combination distributed block snapshot functions stored,
Including carry module 11, deployment module 12, backup module 13 and recovery module 14.
Carry module 11 is used to ceph block devices being directly mounted to production server.
Deployment module 12 is installed for disposing backup client in production server and realizes production server book
The I/O drivers of back-up processing between ceph block devices.
Backup module 13 is connected with carry module 11 and deployment module 12, for establishing production service by backup server
The mirror of device book and ceph block devices；After starting snapshot task, by backup client by production server number
After the mirrored volume in the data copy to ceph block devices of volume, start I/O drivers, by the application of subsequent production server
Generated data are written simultaneously in the mirrored volume on production server book and ceph block devices, and on ceph block devices
Mirrored volume execute a snapshot, record Snapshot time point and SNAPSHOT INFO to backup server.
Wherein, backup server establishes the mirror of production server book and ceph block devices, and configures meter
The task of drawing puts initiation snapshot task at a reasonable time.After starting snapshot task monitoring, backup server is notified that backup visitor
Family end is formatted the mirrored volume on ceph block devices, then will be given birth to by way of disk verification or image copying again
It produces in the mirrored volume in the data copy to ceph block devices of server data volume, starts I/O drivers so that subsequent production
The generated data of application of server are written simultaneously in the mirrored volume on production server book and ceph block devices, and right
Mirrored volume on ceph block devices executes a snapshot, records Snapshot time point and SNAPSHOT INFO to backup server.
Specifically, after starting snapshot task, the application of backup client notice production server is in silent status, then
Mirrored volume on I/O buffering layer data write-in ceph block devices is notified backup clothes by notice driving layer after the completion of operation to be written
Business device calls ceph client interfaces or order line to execute a snapshot to the mirrored volume.After snapshot task, production clothes
The application release silent status of business device.
Specifically, backup client calls ceph client-side interfaces or order line to the mirrored volume on the ceph block devices
Execute a snapshot.Ceph block devices execute a Snapshot time at 20 seconds or so, and I/O buffers layer data and ceph block devices are written
On mirrored volume the time required to determined by the load and buffer layer configuration strategy of production server.
Recovery module 14 is connected with backup module 13, for when carrying out recovery data, being obtained by backup server fast
According to time point, the backup client for reinforming new production server executes primary protection behaviour to the corresponding snapshot of the Snapshot time point
Snapshot is cloned into a new volume after work, finally will newly be wrapping with the data for being loaded onto new production server to be restored.
Specifically, before ceph requires clone's snapshot, protection operation is executed to ensure that snapshot is not deleted, to ensure to clone
Volume is available.
When it is new be wrapping be loaded onto new production server environment when, recovered by the application on newly-generated server to take over
Data.
With reference to Fig. 4, the second embodiment of the data guard method of the snapshot functions of combination distributed block of the invention storage
In, include the following steps：
Step S21, by ceph block device carries to backup server, then by backup server by the ceph blocks after the carry
Hanging equipment is loaded onto production server.
Specifically, by ceph block device carries to backup server, then by backup server by the ceph blocks after the carry
Equipment passes through iSCSI protocol or FC carries to production server.
Step S22, backup client is disposed in production server, and is installed and realized production server book and ceph
The I/O drivers of back-up processing between block device.
Step S23, the mirror of production server book and ceph block devices is established by backup server；When opening
After dynamic snapshot task, by backup client by the mirrored volume in the data copy of production server book to ceph block devices
Afterwards, start I/O drivers, production server book is written into simultaneously in the generated data of the application of subsequent production server
With in the mirrored volume on ceph block devices, and on ceph block devices mirrored volume execute a snapshot, record Snapshot time point
With SNAPSHOT INFO to backup server.
Wherein, backup server establishes the mirror of production server book and ceph block devices, and configures meter
The task of drawing puts initiation snapshot task at a reasonable time.After starting snapshot task monitoring, backup server is notified that backup visitor
Family end is formatted the mirrored volume on ceph block devices, then will be given birth to by way of disk verification or image copying again
It produces in the mirrored volume in the data copy to ceph block devices of server data volume, starts I/O drivers so that subsequent production
The generated data of application of server are written simultaneously in the mirrored volume on production server book and ceph block devices, and right
Mirrored volume on ceph block devices executes a snapshot, records Snapshot time point and SNAPSHOT INFO to backup server.
Specifically, after starting snapshot task, the application of backup client notice production server is in silent status, then
Mirrored volume on I/O buffering layer data write-in ceph block devices is notified backup clothes by notice driving layer after the completion of operation to be written
Business device calls ceph client interfaces or order line to execute a snapshot to the mirrored volume.After snapshot task, production clothes
The application release silent status of business device.
Specifically, backup client calls ceph client-side interfaces or order line to the mirrored volume on the ceph block devices
Execute a snapshot.Ceph block devices execute a Snapshot time at 20 seconds or so, and I/O buffers layer data and ceph block devices are written
On mirrored volume the time required to by the load of production server, buffer layer configuration strategy and backup server forwarding data needed for
Time determines.
When step S24, carrying out recovery data, backup server notifies the standby of new production server according to Snapshot time point
Part client executes the corresponding snapshot of the Snapshot time point and snapshot is cloned a newly volume after primary protection operates, and finally will
Newly it is wrapping with the data for being loaded onto new production server to be restored.
Wherein, the data for being loaded onto new production server to be restored will be newly wrapping with by iSCSI protocol or FC.
Specifically, before ceph requires clone's snapshot, protection operation is executed to ensure that snapshot is not deleted, to ensure to clone
Volume is available.
When it is new be wrapping be loaded onto new production server environment when, recovered by the application on newly-generated server to take over
Data.
With reference to Fig. 5, the second embodiment of the data protection system of the snapshot functions of combination distributed block of the invention storage
In, including carry module 21, deployment module 22, backup module 23 and recovery module 24.
Carry module 21 is used for ceph block device carries to backup server, then will be after the carry by backup server
Ceph block devices pass through iSCSI protocol or FC carries to production server.
Specifically, carry module 21 by ceph block device carries to backup server, then by backup server by the carry
Ceph block devices afterwards pass through iSCSI protocol or FC carries to production server.
Deployment module 22 is installed for disposing backup client in production server and realizes production server book
The I/O drivers of back-up processing between ceph block devices.
Backup module 23 is connected with 21 deployment module 22 of carry module, and production server is established for passing through backup server
The mirror of book and ceph block devices；After starting snapshot task, by backup client by production server data
After mirrored volume in the data copy of volume to ceph block devices, start I/O drivers, subsequent production server is applied into institute
The data of generation are written simultaneously in the mirrored volume on production server book and ceph block devices, and on ceph block devices
Mirrored volume executes a snapshot, records Snapshot time point and SNAPSHOT INFO to backup server.
Wherein, backup server establishes the mirror of production server book and ceph block devices, and configures meter
The task of drawing puts initiation snapshot task at a reasonable time.After starting snapshot task monitoring, backup server is notified that backup visitor
Family end is formatted the mirrored volume on ceph block devices, then will be given birth to by way of disk verification or image copying again
It produces in the mirrored volume in the data copy to ceph block devices of server data volume, starts I/O drivers so that subsequent production
The generated data of application of server are written simultaneously in the mirrored volume on production server book and ceph block devices, and right
Mirrored volume on ceph block devices executes a snapshot, records Snapshot time point and SNAPSHOT INFO to backup server.
Specifically, after starting snapshot task, the application of backup client notice production server is in silent status, then
Mirrored volume on I/O buffering layer data write-in ceph block devices is notified backup clothes by notice driving layer after the completion of operation to be written
Business device calls ceph client interfaces or order line to execute a snapshot to the mirrored volume.After snapshot task, production clothes
The application release silent status of business device.
Specifically, backup client calls ceph client-side interfaces or order line to the mirrored volume on the ceph block devices
Execute a snapshot.Ceph block devices execute a Snapshot time at 20 seconds or so, and I/O buffers layer data and ceph block devices are written
On mirrored volume the time required to by the load of production server, buffer layer configuration strategy and backup server forwarding data needed for
Time determines.
Recovery module 24 is connected with backup module 23, for when carrying out recovery data, when backup server is according to snapshot
Between point, will after notifying the backup client of new production server to execute primary protection operation to the corresponding snapshot of the Snapshot time point
Snapshot clones a new volume, finally will newly be wrapping with the data for being loaded onto new production server to be restored.
Wherein, the data for being loaded onto new production server to be restored will be newly wrapping with by iSCSI protocol or FC
Specifically, before ceph requires clone's snapshot, protection operation is executed to ensure that snapshot is not deleted, to ensure to clone
Volume is available.
When it is new be wrapping be loaded onto new production server environment when, recovered by the application on newly-generated server to take over
Data.
It following is a brief introduction of the operational order of ceph clients：
(1) the mirrored volume information storage of the entitled foo of a 1GB is created in the entitled ponds swimmingpool, held
Row is to issue orders：
rbd create bar--size 1024--pool swimmingpool
(2) snapshot is created to the mirrored volume of entitled foo, executed to issue orders：
rbd snap create swimmingpool/foo@snapname
(3) snapshot is deleted, is executed to issue orders
rbd snap rm swimmingpool/foo@snapname
(4) snapshot is arranged and protects, then clone snapshot, execute to issue orders
rbd snap protect swimmingpool/foo@snapname
rbd clone swimmingpool/foo@snapname newpool/sub-foo
In conclusion the data guard method and system of the snapshot functions of the combination distributed block storage of the present invention combine and divide
Cloth block storage High Availabitity, high reliability the advantages of, be capable of providing the Backup Data of 99.9999% or more degree of reliability,
And ensure that these data can be resumed out at any time, and can restore in arbitrary Snapshot time point, to provide data to the user
Safety guarantee；Easily property extending transversely is stored using distributed block, can easily add clustered node, possesses sea for backup
It measures the computer room of data or cloud platform provides unlimited expansible memory space；In conjunction with the snapshot functions of distributed storage, true
It protects using in the case of consistency, RPO and 1 minute of 2 minutes or so or so RTO times can be provided for data；In conjunction with
CDP technologies, RPO times can be down to 20 seconds or so.So the present invention effectively overcomes various shortcoming in the prior art and has
High industrial utilization.
The above-described embodiments merely illustrate the principles and effects of the present invention, and is not intended to limit the present invention.It is any ripe
The personage for knowing this technology can all carry out modifications and changes to above-described embodiment without violating the spirit and scope of the present invention.Cause
This, institute is complete without departing from the spirit and technical ideas disclosed in the present invention by those of ordinary skill in the art such as
At all equivalent modifications or change, should by the present invention claim be covered.
Claims (10)
1. a kind of data guard method of the snapshot functions of combination distributed block storage, it is characterised in that：Include the following steps：
Ceph block devices are directly mounted in production server；
Backup client is disposed in production server, and is installed and backed up between realization production server book and ceph block devices
The I/O drivers of processing；
The mirror of production server book and ceph block devices is established by backup server；When startup snapshot task
Afterwards, I/O is started by after the mirrored volume in the data copy of production server book to ceph block devices by backup client
The generated data of the application of subsequent production server are written production server book simultaneously and are set with ceph blocks by driver
In standby upper mirrored volume, and a snapshot is executed to the mirrored volume on ceph block devices, records Snapshot time point and SNAPSHOT INFO
To backup server；
When carrying out recovery data, backup server obtains Snapshot time point, reinforms the backup client pair of new production server
The corresponding snapshot of the Snapshot time point executes and snapshot is cloned a new volume after primary protection operates, and is finally loaded onto new be wrapping with
Data of the new production server to be restored.
2. the data guard method of the snapshot functions of combination distributed block storage according to claim 1, it is characterised in that：
After starting snapshot task, the application of production server is in silent status；After snapshot task, the application of production server is released
Put silent status.
3. a kind of data protection system of the snapshot functions of combination distributed block storage, it is characterised in that：Including carry module, portion
Affix one's name to module, backup module and recovery module；
The carry module is used to ceph block devices being directly mounted to production server；
The deployment module for disposing backup client in production server, and install realize production server book with
The I/O drivers of back-up processing between ceph block devices；
The backup module is used to establish the mirror of production server book and ceph block devices by backup server；
After starting snapshot task, by backup client by the mirror in the data copy of production server book to ceph block devices
After volume, start I/O drivers, production server number is written into simultaneously in the generated data of the application of subsequent production server
According to volume with ceph block devices on mirrored volume in, and on ceph block devices mirrored volume execute a snapshot, record snapshot when
Between point and SNAPSHOT INFO to backup server；
The recovery module is used for when carrying out recovery data, is obtained Snapshot time point by backup server, is reinformed new life
The backup client of server is produced to snapshot is cloned one after the primary protection operation of the Snapshot time point corresponding snapshot execution
A new volume finally will newly be wrapping with the data for being loaded onto new production server to be restored.
4. the data protection system of the snapshot functions of combination distributed block storage according to claim 3, it is characterised in that：
In the backup module, after starting snapshot task, the application of production server is in silent status；It is raw after snapshot task
Produce the application release silent status of server.
5. a kind of data guard method of the snapshot functions of combination distributed block storage, it is characterised in that：Include the following steps：
By ceph block device carries to backup server, then by backup server by the ceph block devices carry after the carry to life
Produce server；
Backup client is disposed in production server, and is installed and backed up between realization production server book and ceph block devices
The I/O drivers of processing；
The mirror of production server book and ceph block devices is established by backup server；When startup snapshot task
Afterwards, I/O is started by after the mirrored volume in the data copy of production server book to ceph block devices by backup client
The generated data of the application of subsequent production server are written production server book simultaneously and are set with ceph blocks by driver
In standby upper mirrored volume, and a snapshot is executed to the mirrored volume on ceph block devices, records Snapshot time point and SNAPSHOT INFO
To backup server；
When carrying out recovery data, backup server notifies the backup client of new production server to this according to Snapshot time point
The corresponding snapshot of Snapshot time point executes and snapshot is cloned a new volume after primary protection operation, new will finally be wrapping be loaded onto it is new
Data of the production server to be restored.
6. the data guard method of the snapshot functions of combination distributed block storage according to claim 5, it is characterised in that：
After starting snapshot task, the application of production server is in silent status；After snapshot task, the application of production server is released
Put silent status.
7. the data guard method of the snapshot functions of combination distributed block storage according to claim 5, it is characterised in that：
Ceph block devices after the carry are passed through iSCSI protocol or FC carries to production server by backup server；Carry out data
When recovery, new be wrapping with is loaded onto new production server by backup server by iSCSI protocol or FC.
8. a kind of data protection system of the snapshot functions of combination distributed block storage, it is characterised in that：Including carry module, portion
Affix one's name to module, backup module and recovery module；
The carry module is for passing through ceph block device carries to backup server, then by the ceph block devices after the carry
ISCSI protocol or FC carries are to production server；
The deployment module for disposing backup client in production server, and install realize production server book with
The I/O drivers of back-up processing between ceph block devices；
The backup module is used to establish the mirror of production server book and ceph block devices by backup server；
After starting snapshot task, by backup client by the mirror in the data copy of production server book to ceph block devices
After volume, start I/O drivers, production server number is written into simultaneously in the generated data of the application of subsequent production server
According to volume with ceph block devices on mirrored volume in, and on ceph block devices mirrored volume execute a snapshot, record snapshot when
Between point and SNAPSHOT INFO to backup server；
The recovery module is used for when carrying out recovery data, and backup server notifies new production service according to Snapshot time point
The backup client of device executes the corresponding snapshot of the Snapshot time point and snapshot is cloned a new volume after primary protection operates,
It will newly be wrapping with the data for being loaded onto new production server to be restored finally by iSCSI protocol or FC.
9. the data protection system of the snapshot functions of combination distributed block storage according to claim 8, it is characterised in that：
In the backup module, after starting snapshot task, the application of production server is in silent status；It is raw after snapshot task
Produce the application release silent status of server.
10. the data protection system of the snapshot functions of combination distributed block storage according to claim 8, feature exist
In：In the carry module, the ceph block devices after the carry are passed through iSCSI protocol or FC carries to life by backup server
Produce server；In the recovery module, new be wrapping with is loaded onto new production service by backup server by iSCSI protocol or FC
Device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610146907.8A CN105843704B (en) | 2016-03-15 | 2016-03-15 | A kind of data guard method and system of the snapshot functions of combination distributed block storage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610146907.8A CN105843704B (en) | 2016-03-15 | 2016-03-15 | A kind of data guard method and system of the snapshot functions of combination distributed block storage |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105843704A CN105843704A (en) | 2016-08-10 |
| CN105843704B true CN105843704B (en) | 2018-10-19 |
Family
ID=56588071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610146907.8A Active CN105843704B (en) | 2016-03-15 | 2016-03-15 | A kind of data guard method and system of the snapshot functions of combination distributed block storage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105843704B (en) |
Families Citing this family (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106354582B (en) * | 2016-08-18 | 2019-01-18 | 无锡华云数据技术服务有限公司 | A kind of continuous data protection method |
| CN106445741B (en) * | 2016-09-28 | 2019-08-02 | 郑州云海信息技术有限公司 | One kind realizing oracle database disaster-tolerant backup method based on ceph |
| CN106445746A (en) * | 2016-10-12 | 2017-02-22 | 北京智网科技股份有限公司 | Method and device for disaster recovery backup facing emergency replacement |
| CN106445563A (en) * | 2016-11-24 | 2017-02-22 | 合肥中科云巢科技有限公司 | Operation system mirror image manufacturing method and device |
| CN106708662A (en) * | 2016-12-13 | 2017-05-24 | 上海上讯信息技术股份有限公司 | Method for restoring database and equipment |
| CN110704249A (en) * | 2016-12-30 | 2020-01-17 | 成都华为技术有限公司 | Method, device and system for ensuring application consistency |
| CN107291883B (en) * | 2017-06-20 | 2020-09-18 | 苏州浪潮智能科技有限公司 | Snapshot method and device of distributed storage system |
| CN107291391B (en) * | 2017-06-21 | 2021-03-09 | 郑州云海信息技术有限公司 | Data processing method and system for distributed storage system |
| CN107885616A (en) * | 2017-09-29 | 2018-04-06 | 上海爱数信息技术股份有限公司 | A kind of mass small documents back-up restoring method based on file system parsing |
| CN107729020B (en) * | 2017-10-11 | 2020-08-28 | 北京航空航天大学 | Method for realizing rapid deployment of large-scale container |
| CN107888675A (en) * | 2017-11-08 | 2018-04-06 | 长沙曙通信息科技有限公司 | A kind of distributed memory system data recovery implementation method |
| CN108363642A (en) * | 2018-01-10 | 2018-08-03 | 长沙开雅电子科技有限公司 | A kind of distributed memory system data recovery implementation method |
| CN110399248B (en) * | 2018-04-25 | 2023-08-15 | 阿里巴巴集团控股有限公司 | Image file creation and acquisition method, device and server |
| US10802926B2 (en) * | 2018-09-21 | 2020-10-13 | Microsoft Technology Licensing, Llc | Accessing prior storage segment server state |
| CN109324760B (en) * | 2018-09-21 | 2022-03-22 | 郑州云海信息技术有限公司 | Disk data cloud method, system, equipment and computer storage medium |
| CN109408291A (en) * | 2018-10-31 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of distributed memory system disk snapshot restoration methods, device and electronic equipment |
| CN109766220A (en) * | 2019-01-04 | 2019-05-17 | 平安科技（深圳）有限公司 | Back-up restoring method, device and the computer readable storage medium of application system |
| CN110750325A (en) * | 2019-02-25 | 2020-02-04 | 北京嘀嘀无限科技发展有限公司 | Mirror image backup method and device of virtual machine, fault recovery method and device and electronic equipment |
| CN110008227B (en) * | 2019-04-11 | 2021-08-10 | 苏州浪潮智能科技有限公司 | Consistency group reliability verification method and related device |
| CN110187998A (en) * | 2019-05-09 | 2019-08-30 | 上海爱数信息技术股份有限公司 | The application disaster tolerance system and method stored based on virtual platform and ceph |
| CN110188000B (en) * | 2019-05-16 | 2023-11-17 | 上海爱数信息技术股份有限公司 | Application disaster recovery method and system based on virtualization and iSCSI or FC |
| CN110333823B (en) * | 2019-05-31 | 2023-07-07 | 平安科技（深圳）有限公司 | Data migration method, device, equipment and storage medium based on nbd equipment |
| CN110321075B (en) * | 2019-05-31 | 2023-05-23 | 平安科技（深圳）有限公司 | Data migration method, device, equipment and storage medium based on nbd equipment |
| CN110321251B (en) * | 2019-06-03 | 2023-06-20 | 平安科技（深圳）有限公司 | Data backup method, device, equipment and storage medium based on network block equipment |
| CN110262928A (en) * | 2019-07-02 | 2019-09-20 | 南京云信达科技有限公司 | A method of the fast quick-recovery Backup Data based on emergency backup system |
| CN110287063A (en) * | 2019-07-02 | 2019-09-27 | 南京云信达科技有限公司 | A kind of application method of Backup Data |
| CN110633260B (en) * | 2019-08-07 | 2023-06-06 | 上海爱数信息技术股份有限公司 | NAS storage data continuous protection method and protection system |
| CN111143128B (en) * | 2019-12-24 | 2023-08-18 | 南京壹进制信息科技有限公司 | Openstack cloud host data protection method and system |
| 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 |
| CN112130959B (en) * | 2020-09-29 | 2021-12-07 | 上海英方软件股份有限公司 | Virtual machine protection system and method |
| CN112650447B (en) * | 2020-12-18 | 2024-02-13 | 北京浪潮数据技术有限公司 | Backup method, system and device for ceph distributed block storage |
| CN112486735A (en) * | 2020-12-21 | 2021-03-12 | 上海英方软件股份有限公司 | Data replication system and method for guaranteeing data consistency of application layer |
| CN114584626B (en) * | 2022-03-07 | 2024-08-09 | 浪潮云信息技术股份公司 | Method and device for disaster recovery of hybrid cloud storage |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1873622A (en) * | 2006-04-20 | 2006-12-06 | 北京艾德斯科技有限公司 | Method and equipment for backing up, replicating and recovering data under environment |
| CN102402471A (en) * | 2011-12-23 | 2012-04-04 | 创新科软件技术（深圳）有限公司 | Method and system for backing up data in real time based on snapshot function of memory array |
| CN104714858A (en) * | 2013-12-13 | 2015-06-17 | 中国移动通信集团公司 | Data backup method, data recovery method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2632935C (en) * | 2005-12-19 | 2014-02-04 | Commvault Systems, Inc. | Systems and methods for performing data replication |
-
2016
- 2016-03-15 CN CN201610146907.8A patent/CN105843704B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1873622A (en) * | 2006-04-20 | 2006-12-06 | 北京艾德斯科技有限公司 | Method and equipment for backing up, replicating and recovering data under environment |
| CN102402471A (en) * | 2011-12-23 | 2012-04-04 | 创新科软件技术（深圳）有限公司 | Method and system for backing up data in real time based on snapshot function of memory array |
| CN104714858A (en) * | 2013-12-13 | 2015-06-17 | 中国移动通信集团公司 | Data backup method, data recovery method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105843704A (en) | 2016-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105843704B (en) | A kind of data guard method and system of the snapshot functions of combination distributed block storage | |
| US11099956B1 (en) | Snapshot-based disaster recovery orchestration of virtual machine failover and failback operations | |
| US11237919B2 (en) | Data transfer to a distributed storage environment | |
| US10877856B2 (en) | System for redirecting requests after a secondary storage computing device failure | |
| US11016696B2 (en) | Redundant distributed data storage system | |
| US20200364194A1 (en) | Migration of a database management system to cloud storage | |
| US11010011B2 (en) | File manager integration with virtualization in an information management system with an enhanced storage manager, including user control and storage management of virtual machines | |
| US20200177507A1 (en) | Dynamic management of effective bandwidth of data storage operations | |
| US10698632B2 (en) | Integrated snapshot interface for a data storage system | |
| US11218450B2 (en) | Data protection component scaling in a cloud-based data storage system | |
| US9769260B2 (en) | Cross-system storage management for transferring data across autonomous information management systems | |
| US20170091221A1 (en) | System and method for providing a virtualized replication and high availability environment | |
| US9201736B1 (en) | Methods and apparatus for recovery of complex assets in distributed information processing systems | |
| US20180095855A1 (en) | Heartbeat monitoring of virtual machines for initiating failover operations in a data storage management system, using ping monitoring of target virtual machines | |
| US11074138B2 (en) | Multi-streaming backup operations for mailboxes | |
| US11256529B2 (en) | Methods and apparatus for providing hypervisor level data services for server virtualization | |
| US12019525B2 (en) | Cloud-based recovery of backed up data using auxiliary copy replication and on-demand failover resources | |
| CN116457760A (en) | Asynchronous cross-region block volume replication | |
| US9602341B1 (en) | Secure multi-tenant virtual control server operation in a cloud environment using API provider | |
| US11656947B2 (en) | Data set recovery from a point-in-time logical corruption protection copy | |
| CN110413370B (en) | Method and system for implementing data backup of virtual machine depending on original equipment mapping disk | |
| Anthoniraj et al. | Disaster recovery planning using fault tolerant mechanism in virtualization environment |
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 |