CN109787808A - A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over - Google Patents
A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over Download PDFInfo
- Publication number
- CN109787808A CN109787808A CN201811453737.3A CN201811453737A CN109787808A CN 109787808 A CN109787808 A CN 109787808A CN 201811453737 A CN201811453737 A CN 201811453737A CN 109787808 A CN109787808 A CN 109787808A
- Authority
- CN
- China
- Prior art keywords
- backup
- center
- data center
- data
- wide
- 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.)
- Pending
Links
Abstract
Resist the present invention relates to a kind of multiple data centers for adapting to wide-area deployment and ruin the method for taking over, two adjacent data centers under wide area network deployment conditions pass through the backup relation of its common bearing system, cascade one by one, construct Ring-backup relationship；Inside data center, using bearer service system as minimum unit, a central interior separate service upstream back up by center, and other parts business downstream back up by center, forms the mutually standby structure of the two-way annular based on environment backup relation.Using " two-way annular is mutually standby " strategy, the multiple data centers under the conditions of wide-area deployment are constructed into backup loop.In central interior, using bearer service system as minimum unit, according to building backup loop, partial service system backs up to next data center according to " forward direction " direction, other parts system backs up to a upper data center according to " reversed " direction, forms data center backup system.The capability to resist destruction of system entirety will not be reduced because single centre destroys.
Description
Technical field
The present invention relates to resist to ruin the method for taking over the present invention relates to a kind of multiple data centers for adapting to wide-area deployment.
Background technique
The continuous improvement of the level of informatization, enterprise it is higher and higher to the degree of dependence of information system, the company of operation system
Continuous property and the importance of disaster protection are also more and more prominent.As the operation carrier of information system, data center needs to have anti-
The ability of taking over is ruined, the availability externally serviced with guarantee information system.
Currently, the common disaster tolerance protected mode of data center has local disaster tolerance, double center disaster recoveries and three center of two places to hold
Calamity.
1, data center local disaster tolerance refers to, in forms data central interior, by modes such as redundance units to information system
Protection is provided, when single device software and hardware breaks down, entire information providing service is switched on backup device.
2, the double center disaster recoveries of data center refer to, exist simultaneously Liang Ge data center, information system is in two operation systems
Between dispose, data are synchronous, construct main and standby relation, when a data central interior equipment or it is whole break down when, by
Another data center takes over.
3, three center disaster recovery of data center two places refers to, exists simultaneously three data centers, and two of them are primary data
Center, construct between Liang Ge data center it is active and standby take over relationship, business is taken over when failure.One is Backup Data center, according to
Strategy backs up primary data center, when Liang Ge data center simultaneous faults or when needing to restore historical data, takes over service.
Above-mentioned measure can satisfy the use demand of ordinary enterprises or unit substantially, still, in face of collapse area's wide area, majority
According under the scene of center, above-mentioned disaster tolerance guard method is unable to satisfy anti-ruin and takes over needs.
In face of collapsing under area's wide area, multiple data centers scene, need to establish between center orderly and simply anti-ruin takes over pass
System, on the one hand can reduce maintenance difficulties, on the other hand can avoid multinode to the unordered backup of single node, reduce to backbone communication net
The pressure of network reduces failure chain reaction risk.
Summary of the invention
Goal of the invention: in view of the above-mentioned problems, the object of the present invention is to provide a kind of multiple data centers for adapting to wide-area deployment
It is anti-to ruin the method for taking over.
Technical solution: in order to achieve the above object, two adjacent data under wide area network deployment conditions provided by the invention
Center passes through the backup relation of its common bearing system, cascades one by one, constructs Ring-backup relationship；Inside data center, with
Bearer service system be minimum unit, a central interior separate service upstream center back up, other parts business downstream in
Heart backup forms the mutually standby structure of the two-way annular based on environment backup relation.Using " two-way annular is mutually standby " strategy, by wide area portion
Multiple data centers under the conditions of administration construct backup loop.In central interior, using bearer service system as minimum unit, according to structure
Backup loop is built, partial service system backs up to next data center according to " forward direction " direction, and other parts system is according to " anti-
To " direction backs up to a upper data center, form data center backup system.
Described two adjacent data centers, by data, synchronous, state synchronized keeps data consistent, make bearing system across
Data center backup.
When any data center destruction in Ring-backup relationship or failure, downstream data center can take over offer clothes thereon
Business；According to Ring-backup direction, impacted operation system is destroyed for data center again and constructs backup traffic system.System is whole
The capability to resist destruction of body will not be reduced because single centre destroys.
Working principle: the two-way mutual standby structure of annular provided by the invention, if in backup loop, any data central fault,
Upper and lower Liang Ge data center adjacent thereto takes over fault data center service respectively, and realization business takes over service；Meanwhile therefore
Hinder adjacent upper and lower Liang Ge data center, data center and reconstruct disaster-tolerant relationships, again pull up Ring-backup link, reaches disaster tolerance pass
It is " self-healing ".
The utility model has the advantages that compared with prior art, the invention has the following advantages that
1. using Ring-backup mode between data center, Backup Data is orderly backed up according to loop network, backup relation letter
Single, relative to conventional mesh multi-point backup, network topology is succinct, and constructions cost is lower, small to fault network pressure；
2. using Ring-backup mode between data center, arbitrary data central fault, adjacent data center can take over service
And disaster-tolerant relationships are reconstructed, single centre failure will not reduce system entirety capability to resist destruction.
3. between data center as unit of operation system, using " two-way " Ring-backup mode, the service taken over pair can be reduced
Data center reserves resource requirement, helps to reducing the construction costs.
4. the occasion that the present invention can be used for having data center's wide-area deployment demand, can effectively whole anti-of lifting system ruining
Ability, and reducing the construction costs.
Detailed description of the invention
Fig. 1 data center Ring-backup schematic diagram
Across the data center backup relational graph of Fig. 2 system
Across the data center individual event of Fig. 3 operation system/two-way backup schematic diagram
Anti- ruin of the two-way backup of Fig. 4 data center takes over schematic diagram
Fig. 5 data center unidirectionally backs up anti-ruin and takes over schematic diagram.
Specific embodiment
Preferred embodiment according to the present invention.
A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over, two adjacent under wide area network deployment conditions
Data center passes through the backup relation of its common bearing system, cascades one by one, constructs Ring-backup relationship；In data center
Portion, using bearer service system as minimum unit, a central interior separate service upstream back up by center, and other parts business is downward
It is mutually standby to form the two-way annular based on environment backup relation for the backup of trip center.By the multiple data centers under the conditions of wide-area deployment
Construct backup loop.In central interior, using bearer service system as minimum unit, according to building backup loop, partial service system
System backs up to next data center according to " forward direction " direction, and other parts system backs up to a upper number according to " reversed " direction
According to center, data center backup system is formed, wherein inside data center, forward direction backs up to the part industry of next data center
The quantity of business, the quantity for other business for reversely backing up to a data center can be adjusted voluntarily according to demand, partial service
The accounting of the quantity of quantity and other business may is that 5:5,6:4,4:6,7:3,3:7 etc..
Between the data center for having backup relation, by data, the methods of synchronous, state synchronized keeps system mode
Unanimously, thus realize bearing system across data center backup.
When any data center destruction in Ring-backup relationship or failure, downstream data center can take over offer clothes thereon
Business；According to Ring-backup direction, impacted operation system is destroyed for data center again and constructs standby system.
The capability to resist destruction of system entirety will not be reduced because single centre destroys.
As shown in Fig. 2, system A occupies part resource in data center 1 and data center 2 respectively, and deployment system software, two
By data, synchronous, state synchronized keeps data consistent between person, to realize across the data center backup pass of operation system A
System.Building of the system across data center backup relationship is primary relationship of the invention.
If Fig. 3 is based on across the data center backup relationship of system shown in Fig. 2, list of the operation system across data center is described
Item/two-way backup.Assuming that system A, B is the operation system of data center n carrying, system C, D is the industry of data center n-1 carrying
Business system, system E, F are the operation system of center n+1 carrying.
Under the unidirectional backup relation of operation system, the system C and D under the n-1 of center is backed up to center n, the system under the n of center
A and B is to center n+1 backup, and downstream center is backed up by system E and F under the n+1 of center；Center n-1 also carries upstream center simultaneously
Standby system H1 and I1.
Under the two-way backup relation of operation system, for the system A under the n of center to center n+1 backup, system B is standby to center n-1
Part；System C under the n-1 of center is backed up to center n, and system D upstream back up by center；System E under the n+1 of center is standby to center n
Part, system F downstream back up by center.Based on above-mentioned backup strategy, data center's Ring-backup as shown in Figure 1 can be formed and closed
System.
A kind of multiple data centers adapting to wide-area deployment of the present invention are anti-to ruin the method for taking over, i.e., according to above-mentioned business
The two-way backup relation building of system.
As shown in figure 4, center n-2, n-1, n, n+1 carry two operation systems respectively, closed according to aforementioned two-way backup
System forms positive backup and reversely two backup rings of backup.
For the sake of being briefly described, it is assumed that system resource needed for source system and standby system is 1 unit, then in normal shape
Under state, each center carries the resource consumption of 4 units, respectively two sources operation systems and two backup traffic systems
System.
When a certain moment, center n-1 integrally destroys, and center n-2 and center n need to undertake the industry that center n-1 is carried respectively
Business system.Meanwhile the capability to resist destruction in order to guarantee system entirety, center n need the industry to center n+1 backup from center n-1
Business system, center n-2 need to back up the operation system from center n-1 to center n-3.Therefore, because center n-1 destroys, center
N-3 needs to carry the resource consumption of 5 units, and center n-2 needs to carry the resource consumption of 5 units, and center n needs to carry 5
The resource consumption of a unit, center n+1 need to carry the energy consumption of 5 units.
As described above, constructing two-way ring under the multiple data centers occasion of wide-area deployment based on the method for the invention
The data center backup of shape, when any center is destroyed, neighbouring data center can take over offer service, while be impacted
Operation system rebuild standby system, realize disaster-tolerant relationships " self-healing ".
On the other hand, " two-way " Ring-backup of the present invention can save number compared with " unidirectional " Ring-backup relationship
According to the prep resource at center.As shown in figure 5, center n-2, n-1, n, n+1 carry two operation systems respectively, according to aforementioned list
To backup relation, a positive backup ring is formd.When a certain moment, center n-1 integrally destroys, and center n needs to undertake center
The operation system of n-1 carrying.Meanwhile the capability to resist destruction in order to guarantee system entirety, center n need to come to center n+1 backup
The operation system of center n-1.Therefore, if center n-1 destroys, center n needs to carry the resource consumption of 6 units, and center n+1 is needed
Carry the resource consumption of 6 units.In the same circumstances, it is only needed according to " two-way " backup relation, center n and center n+1
When carrying the resource consumption, i.e. planning construction of 5 units, center needs to build the resource of 5 units, wherein 4 are use
Resource (carrying sources operation system and backup traffic system), 1 is used as reserved resource, relative to needed for " unidirectional " backup relation
The reserved resource of 2 units saves 50%, reduces investment demand.
Claims (4)
1. a kind of multiple data centers for adapting to wide-area deployment are anti-to ruin the method for taking over, it is characterised in that: under wide area network deployment conditions
Two adjacent data centers pass through the backup relation of its common bearing system, cascade one by one, construct Ring-backup relationship；
Inside data center, using bearer service system as minimum unit, a central interior separate service upstream back up by center,
Partial service downstream back up by center, forms the mutually standby structure of the two-way annular based on environment backup relation.
2. a kind of multiple data centers for adapting to wide-area deployment according to claim 1 are anti-to ruin the method for taking over, it is characterised in that:
Described two adjacent data centers, by data, synchronous, state synchronized keeps data consistent, makes bearing system across data center
Backup.
3. a kind of multiple data centers for adapting to wide-area deployment according to claim 1 are anti-to ruin the method for taking over, it is characterised in that:
In certain forms data central interior, using bearer service system as minimum unit, in partial service system and next data center
Portion's operation system establishes backup relation；Another part operation system and upper data central interior operation system establish backup pass
System.
4. a kind of multiple data centers for adapting to wide-area deployment according to claim 1 are anti-to ruin the method for taking over, it is characterised in that:
When any data center destruction in Ring-backup relationship or failure, downstream data center can take over offer service thereon；
According to Ring-backup direction, impacted operation system is destroyed for data center again and constructs backup traffic system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453737.3A CN109787808A (en) | 2019-04-03 | 2019-04-03 | A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453737.3A CN109787808A (en) | 2019-04-03 | 2019-04-03 | A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109787808A true CN109787808A (en) | 2019-05-21 |
Family
ID=66496540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811453737.3A Pending CN109787808A (en) | 2019-04-03 | 2019-04-03 | A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109787808A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111464612A (en) * | 2020-03-30 | 2020-07-28 | 中科九度(北京)空间信息技术有限责任公司 | Method for providing stable computing service in severe environment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1741489A (en) * | 2005-09-01 | 2006-03-01 | 西安交通大学 | High usable self-healing Logic box fault detecting and tolerating method for constituting multi-machine system |
| CN1889547A (en) * | 2006-08-03 | 2007-01-03 | 华为技术有限公司 | Base band unit ring cascade resource back-up method and base band unit |
| CN101848128A (en) * | 2010-04-08 | 2010-09-29 | 深圳市金宏威实业发展有限公司 | Method, system and topological structure for realizing stabilized communication among multi-ring topologies |
| CN104252397A (en) * | 2013-06-27 | 2014-12-31 | 广州中国科学院软件应用技术研究所 | Realizing method for supporting automatic backup of cloud platform user file |
| CN105763570A (en) * | 2016-04-26 | 2016-07-13 | 北京交通大学 | Virtualization-technology-based distributed real-time network simulation system |
-
2019
- 2019-04-03 CN CN201811453737.3A patent/CN109787808A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1741489A (en) * | 2005-09-01 | 2006-03-01 | 西安交通大学 | High usable self-healing Logic box fault detecting and tolerating method for constituting multi-machine system |
| CN1889547A (en) * | 2006-08-03 | 2007-01-03 | 华为技术有限公司 | Base band unit ring cascade resource back-up method and base band unit |
| CN101848128A (en) * | 2010-04-08 | 2010-09-29 | 深圳市金宏威实业发展有限公司 | Method, system and topological structure for realizing stabilized communication among multi-ring topologies |
| CN104252397A (en) * | 2013-06-27 | 2014-12-31 | 广州中国科学院软件应用技术研究所 | Realizing method for supporting automatic backup of cloud platform user file |
| CN105763570A (en) * | 2016-04-26 | 2016-07-13 | 北京交通大学 | Virtualization-technology-based distributed real-time network simulation system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111464612A (en) * | 2020-03-30 | 2020-07-28 | 中科九度(北京)空间信息技术有限责任公司 | Method for providing stable computing service in severe environment |
| CN111464612B (en) * | 2020-03-30 | 2022-05-03 | 中科九度(北京)空间信息技术有限责任公司 | Method for providing stable computing service in severe environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103843286A (en) | Triggering a redundant router master/backup status change based on switch connectivity | |
| CN100534024C (en) | Industry ethernet based fault processing method, system and a switching arrangement | |
| Huang et al. | Challenges of emergency communication network for disaster response | |
| CN101207512B (en) | Method for switching and redundancy backup of database system | |
| CN101605372B (en) | Method for dispatching and managing redundant path in centralized Mesh wireless sensor network | |
| CN107968843A (en) | A kind of new city Metro Network level integrates cloud platform system | |
| CN103560955B (en) | Redundance unit changing method and device | |
| CN101197747A (en) | System and method for industrial real-time control Ethernet redundancy fault-tolerant network | |
| CN106407042B (en) | A kind of across data center disaster tolerance solution system and method based on PostgreSQL database | |
| JP2012129759A (en) | Packet relay device | |
| CN105897472B (en) | The data processing system of business continuance protection is provided | |
| CN106844083A (en) | A kind of fault-tolerance approach and system perceived towards stream calculation system exception | |
| CN105915426B (en) | The fault recovery method and device of loop network | |
| CN109787808A (en) | A kind of multiple data centers adapting to wide-area deployment are anti-to ruin the method for taking over | |
| CN101656630A (en) | Service protection method and system | |
| CN101056254B (en) | An expansion method, system and device of network storage device | |
| CN100512220C (en) | Method for realizing service protection on elastic group ring | |
| CN105812161B (en) | A kind of controller failure backup method and system | |
| CN104135417A (en) | Quick recovery method for ethernet ring network link interruption and corresponding ethernet ring network | |
| CN103036756A (en) | Hybrid looped network protection method and system based on sharing channel | |
| CN101686167A (en) | Method for controlling recovery of faults of Ethernet | |
| CN110635940B (en) | Main/standby switching method of EAPS Ethernet ring network | |
| CN103840972B (en) | The system and method backed up between packet network link aggregation group | |
| CN101764714A (en) | Channel failure reversing method, node and signal transmission system | |
| CN103813369A (en) | Distributed telecommunication switching device backup method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: No.1 Lingshan South Road, Qixia District, Nanjing City, Jiangsu Province 210046 Applicant after: THE 28TH RESEARCH INSTITUTE OF CHINA ELECTRONICS TECHNOLOGY Group Corp. Address before: 210000 No. 1 East Garden Street, Qinhuai District, Jiangsu, Nanjing Applicant before: THE 28TH RESEARCH INSTITUTE OF CHINA ELECTRONICS TECHNOLOGY Group Corp. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190521 |
|
| RJ01 | Rejection of invention patent application after publication |