CN102566938A - Method for realizing relevance between original coil and multiple related coils - Google Patents
Method for realizing relevance between original coil and multiple related coils Download PDFInfo
- Publication number
- CN102566938A CN102566938A CN2011104253405A CN201110425340A CN102566938A CN 102566938 A CN102566938 A CN 102566938A CN 2011104253405 A CN2011104253405 A CN 2011104253405A CN 201110425340 A CN201110425340 A CN 201110425340A CN 102566938 A CN102566938 A CN 102566938A
- Authority
- CN
- China
- Prior art keywords
- volume
- virtual volume
- virtual
- child node
- coil
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method for realizing relevance between an original coil and multiple related coils. The method comprises the following steps: establishing the relevance between the original coil and n related coils, wherein n is a natural number; establishing n virtual coils; forming an (n+1)-level binary tree by use of the original coil, the n virtual coils and the n related coils, wherein a first virtual coil is taken as a parent node of the original coil and a first related coil, the original coil is taken as a left sub-node of the first virtual coil and the first related coil is taken as a right sub-node of the first virtual coil, the (i+1)th virtual coil is taken as the parent node of the ith virtual coil and the (i+1)th related coil, the ith virtual coil is taken as the left sub-node of the (i+1)th virtual coil and the (i+1)th related coil is taken as the right sub-node of the (i+1)th virtual coil, and i is the natural number less than n; and mapping the nth virtual coil into a server end, and then performing reading and writing.
Description
Technical field
The present invention relates to computer memory technical field, relate in particular to a kind of method that realizes that source book is related with a plurality of relevant volume.
Background technology
Along with the continuous development of storage and network technology, the capacity exponentially level of the network storage increases.The storage user of enterprise-level hopes that data have higher reliability, and data that can continuous access enterprises service management; Simultaneously, the system manager of storage server hopes that also more high efficiency memory management policy is arranged.
In the use of enterprise class storage server, the system manager hopes that storage resources can dynamic adjustments, rather than when bringing into use, will predict the capacity that certain storage resources possibly need in future.When certain storage resources is about to exhaust, can online dilatation, guarantee the sustainability visit of data.
Logical volume management (LVM, Logical Volume Manager) provide the scheme that addresses the above problem.LVM is a logical layer that is based upon on hard disk and the subregion, is used to improve the dirigibility of disk partition management.Shown in Figure 1 for the storage system logic structural representation with logical volume management of prior art.The position of volume management module 102 in storage system 100 is under the front end protocol-driven module 101, receives the read/write order, and through the function of self, with read/write order distribution, process cache module 103 is in physical store resource driver module 104.
Through the LVM system, the keeper is the hyperdisk subregion easily.As: several disk partition are connected to the volume group (volume group) of a monoblock, form a storage pool.The keeper can arbitrarily create logical volume group (logical volumes) on the volume group, and further on the logical volume group, creates file system.The keeper can adjust the size of storage volume group easily through LVM, and can name, manage and distribute according to the mode of group disk storage, thereby has improved the efficient of storage administration.
Shown in Figure 2 is the mapping relations synoptic diagram that the logical volume management physical zone arrives logic block.Logical volume management module 102 is formed volume group (VG, Volume Group) with physical volume (PV, Physical Volume), on the volume group, divides logical volume (LV, Logical Volume).After PV is distributed to the volume group, just can create the LV (as long as being no more than the size of VG) of any size.In order to realize that physics is rolled onto the mapping relations of logical volume, logical volume management module 102 also is divided into the more section of small grain size with physical volume and logical volume, is called physical zone (PE, Physical Extent) and logic block (LE, Logical Extent).Physical zone and logic block are one to one always.
Traditional logical volume mapping is main through following dual mode:
Mode one: single logical volume is mapped to server end through LUN, this logical volume is read and write use;
Mode two: a plurality of logical volumes are merged into more jumbo entity, are mapped to server through LUN and use.
This dual mode in fact all is man-to-man mapping.Fig. 3 shows volume mapping synoptic diagram of the prior art.Volume management module 302 is mapped to server 301 with logical volume 303 and uses, and logical volume 303 can be the unity logic volume, also can be the entity that a plurality of logical volumes are merged into.
When these two kinds of methods are used at the Premium Features of storing, all exist not enough.In the application of various storage Premium Features (mirror image, snapshot, clone, data de-duplication), can source book and one or more purpose volume be interrelated usually, for example: snapshot just exists a plurality of snapped volumes (purpose volume) to be associated with source book; Under traditional mode, be difficult to safeguard this non-structured relation.
Summary of the invention
The invention provides a kind of method that realizes that source book is related with a plurality of relevant volume, can facilitate for realizing the storage Premium Features.
A kind of method that realizes that source book is related with a plurality of relevant volume that the embodiment of the invention provides, source book is related with n relevant volume foundation, and n is a natural number; It is characterized in that, set up n virtual volume, said source book, a n virtual volume and n relevant volume are formed n+1 level binary tree; Wherein,
The 1st virtual volume is as the father node of source book and the 1st relevant volume, and source book is as the left child node of the 1st virtual volume, and the 1st relevant volume is as the right child node of the 1st virtual volume;
The i+1 virtual volume is as the father node of i virtual volume and i+1 relevant volume, and the i virtual volume is as the left child node of i+1 virtual volume, and the i+1 relevant volume is as the right child node of i+1 virtual volume, and i is the natural number less than n;
The n virtual volume is mapped to server end to be read and write.
Preferably, this method further comprises:
The n virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self
The i virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self;
Source book receives the IO request from the 1st virtual volume, carries out the IO operation, and returns IO to the 1st virtual volume and accomplish feedback;
After the i virtual volume is received the IO completion feedback from self left child node and right child node, feedback is sent to the father node of self;
After the n virtual volume received the IO completion feedback from self left and right sides child node simultaneously, the feedback that this IO is accomplished sent to server.
Can find out from above technical scheme,, realize the relationship safeguard between the associated logic volume, and then realize the stack of various Premium Features easily through the mode of using virtual logical volume and binary tree to manage; Simultaneously, the variation of the IO command stream that causes owing to Premium Features only realizes on virtual logical volume, and need not make any modification to original source book IO flow process, has guaranteed the basicly stable property of storage system.
Description of drawings
Fig. 1 is the storage system logic structural representation with logical volume management of prior art;
Fig. 2 is the mapping relations synoptic diagram of logical volume management physical zone to logic block;
Fig. 3 is a volume mapping synoptic diagram of the prior art;
The volume mapping synoptic diagram that Fig. 4 provides for the embodiment of the invention;
Fig. 5 is source book and two volume mapping synoptic diagram that relevant volume is related;
Embodiment
The present invention proposes a kind of improved logical volume management method, through the mode of using virtual logical volume and binary tree to manage, realizes the relationship safeguard between the associated logic volume, and then realizes the stack of various Premium Features easily; Simultaneously, the variation of the IO command stream that causes owing to Premium Features only realizes on virtual logical volume, and need not make any modification to original source book IO flow process, has guaranteed the basicly stable property of storage system.
Clearer for the know-why, characteristics and the technique effect that make technical scheme of the present invention, below the present invention program is set forth in detail through specific embodiment.
Fig. 4 shows the volume mapping synoptic diagram that the embodiment of the invention provides, and what this embodiment was directed against is a source book and a related situation of purpose volume.LUN that source book 404 is done mapping is transferred on the same virtual volume 405 of attribute and source book 404, and the disc information of therefore seeing at server 401 is less than variation; Source book 404 is as the left child node of virtual volume 403, and another resource volume that is associated with source book 404 (below be referred to as relevant volume 405), as the right child node of virtual volume 403 nodes.Wherein, virtual volume 403 does not take the actual physical space, and other attributes and source book 404 are just the same.
The IO logic is set on virtual volume 403, and virtual volume 403 judges in this IO request whether the IO sign is arranged after the IO request of receiving from server 401, if having, represent this IO to source book 404, then this IO request is redirected to source book 404.IO process in that server 401 sides are experienced is just operated as directly source book 404 being carried out IO.And the incidence relation between virtual volume 403 management source books 404 and the relevant volume 405, and according to said incidence relation realization Premium Features (like snapshot, clone, mirror image, data de-duplication).
If source book will with plural relevant volume opening relationships, secondary binary tree then shown in Figure 4 can't be managed its logical relation, need set up multistage binary tree.Fig. 5 shows source book and two volume mapping synoptic diagram that relevant volume is related.Comprising three grades of binary trees, as the left child node and the right child node of virtual volume 0505, virtual volume 0505 and relevant volume 1504 are as the left child node and the right child node of virtual volume 1503 respectively for source book 507 and relevant volume 0506.Volume management module 502 is mapped to server 501 with virtual volume 1503.Wherein, the IO logic is set on the virtual volume 1503, after virtual volume 1503 is received the IO request from server 501, judges in this IO request whether the IO sign is arranged,, represent this IO, then this IO request is redirected to left child node virtual volume 0505 to source book if having; Also dispose similar IO logic on the virtual volume 0505; After receiving the IO request that resets from virtual volume 1503 always, judge in this IO request whether the IO sign is arranged, if having; Represent this IO to source book, then this IO request is redirected to left child node source book 507.
Can find out that if source book is related with n relevant volume foundation, then need set up n virtual volume, said source book, a n virtual volume and n relevant volume are formed n+1 level binary tree; Wherein,
The 1st virtual volume is as the father node of source book and the 1st relevant volume, and source book is as the left child node of the 1st virtual volume, and the 1st relevant volume is as the right child node of the 1st virtual volume;
The i+1 virtual volume is as the father node of i virtual volume and i+1 relevant volume, and the i virtual volume is as the left child node of i+1 virtual volume, and the i+1 relevant volume is as the right child node of i+1 virtual volume, and i is the natural number less than n;
The n virtual volume is mapped to server end to be read and write.
When carrying out the IO operation,
2, method according to claim 1 is characterized in that, this method further comprises:
The n virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self;
The i virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self;
Source book receives the IO request from the 1st virtual volume, carries out the IO operation, and returns IO to the 1st virtual volume and accomplish feedback;
After the i virtual volume is received the IO completion feedback from self left child node and right child node, feedback is sent to the father node of self;
After the n virtual volume received the IO completion feedback from self left and right sides child node simultaneously, the feedback that this IO is accomplished sent to server.
The above is merely preferred embodiment of the present invention, and is in order to restriction the present invention, not all within spirit of the present invention and principle, any modification of being made, is equal to replacement, improvement etc., all should be included within the scope that the present invention protects.
Claims (2)
1. method that realizes that source book is related with a plurality of relevant volume, source book will be set up relatedly with n relevant volume, and n is a natural number; It is characterized in that, set up n virtual volume, said source book, a n virtual volume and n relevant volume are formed n+1 level binary tree; Wherein,
The 1st virtual volume is as the father node of source book and the 1st relevant volume, and source book is as the left child node of the 1st virtual volume, and the 1st relevant volume is as the right child node of the 1st virtual volume;
The i+1 virtual volume is as the father node of i virtual volume and i+1 relevant volume, and the i virtual volume is as the left child node of i+1 virtual volume, and the i+1 relevant volume is as the right child node of i+1 virtual volume, and i is the natural number less than n;
The n virtual volume is mapped to server end to be read and write.
2. method according to claim 1 is characterized in that, this method further comprises:
The n virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self
The i virtual volume receives the IO request from server, duplicates a new IO request, issues self left child node; According to the logic of virtual volume self, generate a new IO request simultaneously, send to the right child node of self;
Source book receives the IO request from the 1st virtual volume, carries out the IO operation, and returns IO to the 1st virtual volume and accomplish feedback;
After the i virtual volume is received the IO completion feedback from self left child node and right child node, feedback is sent to the father node of self;
After the n virtual volume received the IO completion feedback from self left and right sides child node simultaneously, the feedback that this IO is accomplished sent to server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104253405A CN102566938A (en) | 2011-12-16 | 2011-12-16 | Method for realizing relevance between original coil and multiple related coils |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104253405A CN102566938A (en) | 2011-12-16 | 2011-12-16 | Method for realizing relevance between original coil and multiple related coils |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102566938A true CN102566938A (en) | 2012-07-11 |
Family
ID=46412463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104253405A Pending CN102566938A (en) | 2011-12-16 | 2011-12-16 | Method for realizing relevance between original coil and multiple related coils |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102566938A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040015672A1 (en) * | 2002-07-17 | 2004-01-22 | Masse Wayne P. | Method, system, and storage medium for optimizing disk space and information access |
| CN101398822A (en) * | 2007-09-26 | 2009-04-01 | 新奥特硅谷视频技术有限责任公司 | Method for dynamically extending network memory space by virtual file systems technology |
| CN101840308A (en) * | 2009-10-28 | 2010-09-22 | 创新科存储技术有限公司 | Hierarchical memory system and logical volume management method thereof |
| US20110082997A1 (en) * | 2009-10-04 | 2011-04-07 | Infinidat Ltd. | Virtualized storage system and method of operating thereof |
| CN102117184A (en) * | 2010-01-05 | 2011-07-06 | 联想(北京)有限公司 | Controller for multimemory, terminal equipment and starting and access method thereof |
-
2011
- 2011-12-16 CN CN2011104253405A patent/CN102566938A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040015672A1 (en) * | 2002-07-17 | 2004-01-22 | Masse Wayne P. | Method, system, and storage medium for optimizing disk space and information access |
| CN101398822A (en) * | 2007-09-26 | 2009-04-01 | 新奥特硅谷视频技术有限责任公司 | Method for dynamically extending network memory space by virtual file systems technology |
| US20110082997A1 (en) * | 2009-10-04 | 2011-04-07 | Infinidat Ltd. | Virtualized storage system and method of operating thereof |
| CN101840308A (en) * | 2009-10-28 | 2010-09-22 | 创新科存储技术有限公司 | Hierarchical memory system and logical volume management method thereof |
| CN102117184A (en) * | 2010-01-05 | 2011-07-06 | 联想(北京)有限公司 | Controller for multimemory, terminal equipment and starting and access method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10042869B1 (en) | Method for data transfer between compute clusters and file system | |
| US10178174B2 (en) | Migrating data in response to changes in hardware or workloads at a data store | |
| KR101930117B1 (en) | Volatile memory representation of nonvolatile storage device set | |
| JP5411250B2 (en) | Data placement according to instructions to redundant data storage system | |
| CN110663019B (en) | File system for Shingled Magnetic Recording (SMR) | |
| US9442954B2 (en) | Method and apparatus for achieving optimal resource allocation dynamically in a distributed computing environment | |
| CN101840308B (en) | Hierarchical memory system and logical volume management method thereof | |
| CN104461685B (en) | Virtual machine processing method and virtual computer system | |
| JP2017091548A (en) | Dispersion storage system, and dispersion storage method, and dispersion facility | |
| CN102136003A (en) | Large-scale distributed storage system | |
| CN107291889A (en) | A kind of date storage method and system | |
| CN104462389B (en) | Distributed file system implementation method based on classification storage | |
| CN105242881A (en) | Distributed storage system and data read-write method for same | |
| CN104536903B (en) | A kind of mixing storage method and system stored classifiedly by data attribute | |
| CN103503414A (en) | Computing storage integration cluster system | |
| US10346362B2 (en) | Sparse file access | |
| US20200133555A1 (en) | Mechanisms for performing accurate space accounting for volume families | |
| CN106570113B (en) | Mass vector slice data cloud storage method and system | |
| CN104272242A (en) | Creating encrypted storage volumes | |
| CN103761059A (en) | Multi-disk storage method and system for mass data management | |
| CN105573674A (en) | Distributed storage method oriented to a large number of small files | |
| CN106960011A (en) | Metadata of distributed type file system management system and method | |
| US7792966B2 (en) | Zone control weights | |
| CN107408239B (en) | Architecture for managing mass data in communication application through multiple mailboxes | |
| KR101531564B1 (en) | Method and System for load balancing of iSCSI storage system used network distributed file system and 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 | ||
| ASS | Succession or assignment of patent right |
Owner name: BEIHAI UNITED INFORMATION TECHNOLOGY CO., LTD. Free format text: FORMER OWNER: COOL CLOUDZ INFORMATION TECHNOLOGY (TIANJIN) CO., LTD. Effective date: 20130722 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 300350 JINNAN, TIANJIN TO: 536000 BEIHAI, GUANGXI ZHUANG AUTONOMOUS REGION |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20130722 Address after: 536000, Taiwan Road, the Guangxi Zhuang Autonomous Region, Beihai, northeast of Jilin Road (China Electronics Beihai Industrial Park, East 14 room) Applicant after: UNITED INFORMATION TECHNOLOGY CO., LTD. Applicant after: Clouds creators information technology (Wuxi) Co., Ltd. Address before: 300350 9, Xing Xing Road, Haihe Industrial Zone, Jinnan District, Tianjin, 8013 Applicant before: Clouds creators information technology (Tianjin) Co., Ltd. Applicant before: Clouds creators information technology (Wuxi) Co., Ltd. |
|
| C05 | Deemed withdrawal (patent law before 1993) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120711 |