CN103678715B - The metadata information management method of snapshot is supported in a kind of distributed file system - Google Patents
The metadata information management method of snapshot is supported in a kind of distributed file system Download PDFInfo
- Publication number
- CN103678715B CN103678715B CN201310752681.2A CN201310752681A CN103678715B CN 103678715 B CN103678715 B CN 103678715B CN 201310752681 A CN201310752681 A CN 201310752681A CN 103678715 B CN103678715 B CN 103678715B
- Authority
- CN
- China
- Prior art keywords
- ino
- snapshot
- file
- inode
- seg
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/11—File system administration, e.g. details of archiving or snapshots
- G06F16/128—Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of metadata information management method of support snapshot in distributed file system, comprises the following steps:Section resource is managed;Establishment file snapshot and to create File Snapshot modify;INode resources are managed;Create directory snapshot and to create catalogue snapshot modify.The metadata information management method of snapshot is supported in the distributed file system that the present invention is provided, by iNode information, segment information are respectively adopted iNode managers and fragment manager is safeguarded in MDS, and the method safeguarded by manager, that lets nature take its course realizes the snapshot of distributed file system.
Description
Technical field
The present invention relates to a kind of management method, the metadata of snapshot is supported in particular to a kind of distributed file system
Approaches to IM.
Background technology
Distributed file system generally comprises client, meta data server (MetaData Servicer, MDS) sum
According to server (Data Servicer, DS), the access interface that client is responsible for file data is formulated, at meta data server device
Manage the layout and attribute of file, the data content of data server device storage file.
For distributed file system, snapping technique is the effective means for realizing taking precautions against loss of data.When storage device hair
When raw application and trouble or file corruption, it is possible to use snapshot restores data to the state of snapshot generation time point;Additionally, fast
Another data access channel is provided according to for storage user, when former data carry out application on site to be processed, user can access
Snapshot carries out dependence test work using snapshot.
Copy on write (Copy On Write, COW) is the basic thought for realizing File Snapshot, and modification formula accesses → division
→ access of modification formula is redirected, it is the basic procedure for realizing COW.Some file system are established soon in order to realize snapshot, additionally
According to table, all whether snapshot can be have accessed by snapshot table look-up when accessing every time, and decide whether to perform division treatment;Also
A little file system can all be examined when accessing every time by the extra metadata reference count table for increasing data block by reference count table
Look into and whether have accessed snapshot, and decide whether to perform division treatment.
Current snapping technique mainly has a disadvantage that:
1) technology based on snapshot table/reference count table look-up, though original metadata information structure need not be changed, when fast
According to table/reference count table it is very big when, check snapshot table will be than relatively time-consuming;
2) distributed file system snapshot realizes the realization sides for being confined to File Snapshot grade, lacking directory level snapshot more at present
Case.
The content of the invention
In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a kind of unit of support snapshot in distributed file system
Data information management method, by iNode information, segment information are respectively adopted iNode managers and fragment manager is tieed up in MDS
Shield, and the method safeguarded by manager, that lets nature take its course realizes the snapshot of distributed file system.
In order to realize foregoing invention purpose, the present invention is adopted the following technical scheme that:
The present invention provides a kind of metadata information management method of support snapshot in distributed file system, methods described bag
Include following steps:
Step 1:Section resource is managed;
Step 2:Establishment file snapshot and to create File Snapshot modify;
Step 3:INode resources are managed;
Step 4:Create directory snapshot and to create catalogue snapshot modify.
In the step 1, the base attribute of section resource includes reference count RC and division version SV;Step 1 is specifically included
Following steps:
Step 1-1:When corresponding file data content changes in segment layout, if reference count RC > 1, section is provided
Line splitting is entered in source;
Step 1-2:When file is deleted, meta data server will change all RC of corresponding iNode quoting resources, make RC
=RC-1;If the RC values of section resource are 0, notify that data server deletes the corresponding file of segment layout, and from fragment manager
Middle exhausting section resource.
If ino represents the iNode resources of current modification file, seg represents current modification corresponding section of resource of data content,
It is as follows that meta data server carries out fission process to seg:
If 1) seg layout associated metadatas server space is not enough, for seg_new builds new layout;Otherwise, copy
Seg gives new section resource seg_new, and sets seg_new attributes RC=1, SV=0;
2) if seg is consistent with seg_new layouts and related Data Services device supports the snapshot of data object file, build
Found the snapshot of corresponding objects file in seg and be mapped to seg_new corresponding objects files;Otherwise, corresponding objects text in copy seg
Part as seg_new obj ect file;
3) seg attributes RC=RC-1, SV=SV+1;
4) ino quotes seg_new.
The step 2 is comprised the following steps:
Step 2-1:If ino represents File Snapshot source, ino_snap represents the snapshot document that need to be created, then Metadata Service
Device establishment file snapshot processes are as follows:
1) all sections of quoting resource information are to ino_snap in copying ino;
2) all sections of resource RC attributes that ino is quoted are set:RC=RC+1.
Step 2-2:After the completion of establishment file snapshot, the modification basis section resource management to File Snapshot is processed.
In the step 3, the base attribute of the iNode resources includes reference count RC and division version SV;Step 3 has
Body is comprised the following steps:
Step 3-1:When client need change specified path file/directory when, client will since root, to
Meta data server obtains the iNode resources of each directory entry on the path step by step;If meeting the iNode Resource Properties RC of directory entry>
1, then need to perform the division treatment of iNode resources since the directory entry, until afterbody directory entry on path
INode resources are completed;
Step 3-2:When client deletes the catalogue of specified path, meta data server performs the flow that deltrees;
Step 3-3:If client deletes the file of specified path, according to modification catalogue handling process, file is completed
The division treatment of place catalogue;Then according to the handling process that file is deleted in section resource management, delete processing is completed.
In the step 3-1, if ino represents the iNode resources of current directory, ino_parent represents parent directory
INode resources, the process that meta data server enters line splitting to ino is as follows:
1) ino attributes RC=RC-1, SV=SV+1 are set;
2) copy ino resource informations give new iNode resources ino_new, and set ino_new attributes RC=1, SV=0;
3) ino_parent quotes ino_new;
4) because ino_new and ino quotes identical resource, if ino is file type, the section resource of all references
Attribute RC=RC+1;If ino is Directory Type, the iNode Resource Properties RC=RC+1 of all references.
The step 3-2 is comprised the following steps:
1) according to modification catalogue handling process, division treatment is completed;If ino_parent represents band and deltrees father's mesh
Record the iNode resources of item;
2) directory entry of correspondence ino in ino_parent is deleted, and ino attributes RC=RC-1 is set;
3) if RC values are 0, ino is reclaimed from iNode managers.
Described 4 comprise the following steps:
Step 4-1:If ino represents catalogue snapshot source, ino_snap_parent represents the parent directory of catalogue snapshot, then unit
The data server snapshot processes that create directory are as follows:
1) directory entry is increased in ino_snap_parent, the entitled catalogue snapshot title of directory entry, directory entry
INode quotes ino;
2) ino attributes RC=RC+1 is set;
Step 4-2:Create directory after the completion of snapshot, the modification basis iNode resource managements to catalogue snapshot are processed.
Compared with prior art, the beneficial effects of the present invention are:
1) metadata information of MDS in distributed file system is employed management as resource scheme by the present invention, is snapshot
Realization lay a good foundation;And by the way that section resource management and iNode resource managements is discussed in detail, illustrate this resource management side
Method being capable of natural support snapshot;
2) compared with traditional snapshot implementation, the present invention need not carry out specially treated to snapshot, while eliminating snapshot
Dependence between source and snapshot, effectively reduces the complexity of snapshot realization;
3) while, according to distributed file system the characteristics of, the present invention COW characteristics are refine to the aspect of section at MDS ends,
Reduce the cost of COW singles division;
4) invention has been that a kind of distributed file system realizes snapshot functions.
Brief description of the drawings
Fig. 1 is the section resource schematic diagram cited in the iNode resources ino_1 for creating snapshot preceding document;
Fig. 2 is file and segment information schematic diagram after the completion of establishment file snapshot;
Fig. 3 is file ino_1 modification seg_3 and file ino_1_0 modification seg_2 schematic diagrames;
Fig. 4 is to delete file and segment information schematic diagram after file ino_1;
Fig. 5 is create directory bibliographic structure and iNode resource information schematic diagrames before snapshot;
Fig. 6 is the/snapshot/snap_d_a schematic diagrames of d_a that create directory;
Fig. 7 is modification/snap_d_a/d_b directory entry information schematic diagrames;
Fig. 8 is deletion/snap_d_a/d_b_ext directory entry schematic diagrames.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
The present invention provides a kind of metadata information management method of support snapshot in distributed file system, methods described bag
Include following steps:
Step 1:Section resource is managed;
In the step 1, the base attribute of section resource includes reference count RC (Reference Count) and division version
SV(Split Version);Step 1 specifically includes following steps:
Step 1-1:When corresponding file data content changes in segment layout, if reference count RC > 1, section is provided
Line splitting is entered in source;
If ino represents the iNode resources of current modification file, seg represents current modification corresponding section of resource of data content,
It is as follows that meta data server carries out fission process to seg:
If 1) seg layout associated metadatas server space is not enough, for seg_new builds new layout;Otherwise, copy
Seg gives new section resource seg_new, and sets seg_new attributes RC=1, SV=0;
2) if seg is consistent with seg_new layouts and related Data Services device supports the snapshot of data object file, build
Found the snapshot of corresponding objects file in seg and be mapped to seg_new corresponding objects files;Otherwise, corresponding objects text in copy seg
Part as seg_new obj ect file;
3) seg attributes RC=RC-1, SV=SV+1;
4) ino quotes seg_new.
Step 1-2:When file is deleted, meta data server will change all RC of corresponding iNode quoting resources, make RC
=RC-1;If the RC values of section resource are 0, notify that data server deletes the corresponding file of segment layout, and from fragment manager
Middle exhausting section resource.
Step 2:Establishment file snapshot and to create File Snapshot modify;
The step 2 is comprised the following steps:
Step 2-1:If ino represents File Snapshot source, ino_snap represents the snapshot document that need to be created, then Metadata Service
Device establishment file snapshot processes are as follows:
1) all sections of quoting resource information are to ino_snap in copying ino;
2) all sections of resource RC attributes that ino is quoted are set:RC=RC+1.
Step 2-2:After the completion of establishment file snapshot, the modification basis section resource management to File Snapshot is processed.
Below by taking establishment file snapshot and modification File Snapshot as an example, the establishment of present document snapshot and the pipe of section resource
Reason.
Before creating snapshot, the section resource cited in the iNode resources ino_1 of file is as shown in Figure 1.If client is to MDS
When asking the layout information of ino_1, all segment layout information answers that MDS quotes ino_1 are to client.
First, MDS receives the snapshot document ino_1_0 requests of establishment file ino_1.Establishment process is by institute in ino_1
There is section quoting resource information copy to ino_1_0, and increase ino_1 and quote each section of RC value of resource.Establishment file snapshot is completed
Afterwards, file and segment information are as shown in Figure 2.
Then, the corresponding data of section seg_3 that client requirements modification file ino_1 is quoted.Due to seg_3 resources RC>
1, so, according to section resource management action, MDS need to enter line splitting to seg_3.Equally, client requirements modification file ino_1_0
The corresponding data of section seg_2 of reference, MDS also needs to enter line splitting to seg_2.After division, file and segment information as shown in figure 3,
Wherein, seg_4 is formed by seg_3 divisions, and seg_5 is formed by seg_2 divisions.
Finally, MDS receives the ino_1 requests of deletion file.Seg_1, seg_2, seg_4 section of resource that file ino_1 is quoted
RC values all subtract 1, wherein, seg_2 and seg_4 sections of resource RC value is 0, according to section resource management action, will carry out relevant recovery
Treatment.After deleting file ino_1, file and segment information are as shown in Figure 4.
Step 3:INode resources are managed;
In the step 3, the base attribute of the iNode resources includes reference count RC and division version SV;Step 3 has
Body is comprised the following steps:
Step 3-1:When client need change specified path file/directory when, client will since root, to
Meta data server obtains the iNode resources of each directory entry on the path step by step;If meeting the iNode Resource Properties RC of directory entry>
1, then need to perform the division treatment of iNode resources since the directory entry, until afterbody directory entry on path
INode resources are completed;
In the step 3-1, if ino represents the iNode resources of current directory, ino_parent represents parent directory
INode resources, the process that meta data server enters line splitting to ino is as follows:
1) ino attributes RC=RC-1, SV=SV+1 are set;
2) copy ino resource informations give new iNode resources ino_new, and set ino_new attributes RC=1, SV=0;
3) ino_parent quotes ino_new;
4) because ino_new and ino quotes identical resource, if ino is file type, the section resource of all references
Attribute RC=RC+1;If ino is Directory Type, the iNode Resource Properties RC=RC+1 of all references.
Step 3-2:When client deletes the catalogue of specified path, meta data server performs the flow that deltrees;
The step 3-2 is comprised the following steps:
1) according to modification catalogue handling process, division treatment is completed (not including ino);If ino_parent represents that band is deleted
The iNode resources of directory entry parent directory;
2) directory entry of correspondence ino in ino_parent is deleted, and ino attributes RC=RC-1 is set;
3) if RC values are 0, ino is reclaimed from iNode managers.
Step 3-3:If client deletes the file of specified path, according to modification catalogue handling process, file is completed
The division treatment of place catalogue;Then according to the handling process that file is deleted in section resource management, delete processing is completed.
Step 4:Create directory snapshot and to create catalogue snapshot modify.
Described 4 comprise the following steps:
Step 4-1:If ino represents catalogue snapshot source, ino_snap_parent represents the parent directory of catalogue snapshot, then unit
The data server snapshot processes that create directory are as follows:
1) directory entry is increased in ino_snap_parent, the entitled catalogue snapshot title of directory entry, directory entry
INode quotes ino;
2) ino attributes RC=RC+1 is set;
Step 4-2:Create directory after the completion of snapshot, the modification basis iNode resource managements to catalogue snapshot are processed.
Below by taking create directory snapshot and modification catalogue snapshot as an example, establishment and the iNode resources of catalogue snapshot are introduced
Management.
Create directory before snapshot, bibliographic structure and iNode resource informations are as shown in Figure 5.The iNode moneys that each directory entry is quoted
It is responsible for by iNode managers in source.
First, MDS receives the/snapshot snap_d_a of the d_a request that creaties directory under the root directory, as shown in Figure 6.Due to fast
It is identical with the content that/d_a is included according to snap_d_a, so snap_d_a directly quotes the iNode resources ino_ of d_a directory entries
2, and increase ino_2 attribute RC values.
Then, MDS reception/snap_d_a/d_b are renamed as/snap_d_a/d_b_ext requests.According to iNode resource pipes
The handling process of reason, the corresponding ino_2 Resource Properties RC of snap_d_a in path>1, so need to enter ino_2 line splitting, and
And snap_d_a quotes the iNode resources ino_6 after division.After division, the iNode that all directory entries of snap_d_a are quoted is provided
Source RC values increase 1 (such as:The RC values of ino_3 and ino_5 are for 2).
Equally, according to the handling process of iNode resource managements, the ino_3 that will continue right/snap_d_a/d_b references is carried out
Division.
In/snap_d_a/d_b paths after the completion of each catalogue term separation task, the treatment that renames is performed.Complete division and rename
Bibliographic structure and iNode resources afterwards is as shown in Figure 7.Wherein ino_2 and ino_3 Resource Properties SV values are 1, are represented by 1 time
Division.
Finally, MDS receives the request of deletion/snap_d_a/d_b_ext/f.dat directory entries.First, according to iNode resources
Management handling process, each directory entry in inspection/snap_d_a/d_b_ext paths simultaneously performs division treatment;Due to each directory entry
INode Resource Properties RC values are 1, so without division.Then, f.dat directory entries are deleted;After deletion, ino_4 resources
Attribute RC=1, so ino_4 resources need not be reclaimed.Bibliographic structure and iNode information after deletion is as shown in Figure 8.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention rather than its limitations, to the greatest extent
Pipe has been described in detail with reference to above-described embodiment to the present invention, and those of ordinary skill in the art should be understood:Still
Specific embodiment of the invention can be modified or equivalent, and without departing from any of spirit and scope of the invention
Modification or equivalent, it all should cover in the middle of scope of the presently claimed invention.
Claims (6)
1. the metadata information management method of snapshot is supported in a kind of distributed file system, it is characterised in that:Methods described bag
Include following steps:
Step 1:Section resource is managed;
Step 2:Establishment file snapshot and to create File Snapshot modify;
Step 3:INode resources are managed;
Step 4:Create directory snapshot and to create catalogue snapshot modify;
In the step 1, the base attribute of section resource includes reference count RC and division version SV;Step 1 specifically includes following
Step:
Step 1-1:When corresponding file data content changes in segment layout, if reference count RC > 1, enter to section resource
Line splitting;
Step 1-2:When file is deleted, meta data server will change all RC of corresponding iNode quoting resources, make RC=
RC-1;If the RC values of section resource are 0, notify that data server deletes the corresponding file of segment layout, and from fragment manager
Exhausting section resource;
If ino represents the iNode resources of current modification file, seg represents current modification corresponding section of resource of data content, first number
Fission process is carried out to seg according to server as follows:
If 1) seg layout associated metadatas server space is not enough, for seg_new builds new layout;Otherwise, seg is copied
To new section resource seg_new, and seg_new attributes RC=1, SV=0 are set;
2) if seg is consistent with seg_new layouts and related Data Services device supports the snapshot of data object file, set up
The snapshot of corresponding objects file and seg_new corresponding objects files are mapped in seg;Otherwise, corresponding objects file in copy seg
As the obj ect file of seg_new;
3) seg attributes RC=RC-1, SV=SV+1;
4) ino quotes seg_new.
2. the metadata information management method of snapshot, its feature are supported in distributed file system according to claim 1
It is:The step 2 is comprised the following steps:
Step 2-1:If ino_snap represents the snapshot document that need to be created, then meta data server establishment file snapshot processes are such as
Under:
1) all sections of quoting resource information are to ino_snap in copying ino;
2) all sections of resource RC attributes that ino is quoted are set:RC=RC+1;
Step 2-2:After the completion of establishment file snapshot, the modification basis section resource management to File Snapshot is processed.
3. the metadata information management method of snapshot, its feature are supported in distributed file system according to claim 1
It is:In the step 3, the base attribute of the iNode resources includes reference count RC and division version SV;Step 3 is specific
Comprise the following steps:
Step 3-1:When client needs to change the file/directory of specified path, client will be since root, to first number
Obtain the iNode resources of each directory entry on the path step by step according to server;If meeting the iNode Resource Properties RC of directory entry>1, then
The division that iNode resources are performed since the directory entry is needed to process, until the iNode moneys of afterbody directory entry on path
Source completes;
Step 3-2:When client deletes the catalogue of specified path, meta data server performs the flow that deltrees;
Step 3-3:If client deletes the file of specified path, according to modification catalogue handling process, file place is completed
The division treatment of catalogue;Then according to the handling process that file is deleted in section resource management, delete processing is completed.
4. the metadata information management method of snapshot, its feature are supported in distributed file system according to claim 3
It is:In the step 3-1, if ino_parent represents the iNode resources of parent directory, meta data server is carried out to ino
The process of division is as follows:
1) ino attributes RC=RC-1, SV=SV+1 are set;
2) copy ino resource informations give new iNode resources ino_new, and set ino_new attributes RC=1, SV=0;
3) ino_parent quotes ino_new;
4) because ino_new and ino quotes identical resource, if ino is file type, the section Resource Properties of all references
RC=RC+1;If ino is Directory Type, the iNode Resource Properties RC=RC+1 of all references.
5. the metadata information management method of snapshot, its feature are supported in distributed file system according to claim 3
It is:The step 3-2 is comprised the following steps:
1) according to modification catalogue handling process, division treatment is completed;If ino_parent represents band and deltrees parent directory
INode resources;
2) directory entry of correspondence ino in ino_parent is deleted, and ino attributes RC=RC-1 is set;
3) if RC values are 0, ino is reclaimed from iNode managers.
6. the metadata information management method of snapshot, its feature are supported in distributed file system according to claim 1
It is:The step 4 is comprised the following steps:
Step 4-1:If ino_snap_parent represents the parent directory of catalogue snapshot, then meta data server creaties directory snapshot
Process is as follows:
1) directory entry is increased in ino_snap_parent, the entitled catalogue snapshot title of directory entry, the iNode of directory entry draws
Use ino;
2) ino attributes RC=RC+1 is set;
Step 4-2:Create directory after the completion of snapshot, the modification basis iNode resource managements to catalogue snapshot are processed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310752681.2A CN103678715B (en) | 2013-12-31 | 2013-12-31 | The metadata information management method of snapshot is supported in a kind of distributed file system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310752681.2A CN103678715B (en) | 2013-12-31 | 2013-12-31 | The metadata information management method of snapshot is supported in a kind of distributed file system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103678715A CN103678715A (en) | 2014-03-26 |
CN103678715B true CN103678715B (en) | 2017-06-23 |
Family
ID=50316259
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310752681.2A Active CN103678715B (en) | 2013-12-31 | 2013-12-31 | The metadata information management method of snapshot is supported in a kind of distributed file system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103678715B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103984609B (en) * | 2014-05-28 | 2017-06-16 | 华为技术有限公司 | A kind of method and apparatus that checkpoint is reclaimed in file system based on copy-on-write |
CN106557263B (en) | 2015-09-25 | 2019-08-16 | 伊姆西公司 | For checking pseudo- shared method and apparatus in data block is deleted |
US10810168B2 (en) | 2015-11-24 | 2020-10-20 | Red Hat, Inc. | Allocating file system metadata to storage nodes of distributed file system |
CN105302922B (en) * | 2015-11-24 | 2018-07-06 | 无锡江南计算技术研究所 | A kind of distributed file system snapshot implementing method |
CN107291579B (en) * | 2017-07-05 | 2020-03-31 | 郑州云海信息技术有限公司 | Metadata cluster file snapshot method based on static subtree partition |
CN107958034A (en) * | 2017-11-20 | 2018-04-24 | 郑州云海信息技术有限公司 | Distribution method, device and the medium of the inode number of distributed file system |
CN111858474B (en) * | 2020-06-29 | 2022-12-27 | 苏州浪潮智能科技有限公司 | Distributed storage system Inode number allocation management method and related components |
CN113515487B (en) * | 2021-09-07 | 2021-11-19 | 联想凌拓科技有限公司 | Directory query method, computing device and distributed file system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101162469A (en) * | 2007-11-09 | 2008-04-16 | 清华大学 | Fine grit document and catalogs version management method based on snapshot |
CN102024018A (en) * | 2010-11-04 | 2011-04-20 | 曙光信息产业(北京)有限公司 | On-line recovering method of junk metadata in distributed file system |
CN102073739A (en) * | 2011-01-25 | 2011-05-25 | 中国科学院计算技术研究所 | Method for reading and writing data in distributed file system with snapshot function |
CN102722536A (en) * | 2012-05-22 | 2012-10-10 | 中兴通讯股份有限公司 | Method and device for realizing snapshot based on distributed file system |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7913046B2 (en) * | 2007-08-06 | 2011-03-22 | Dell Global B.V. - Singapore Branch | Method for performing a snapshot in a distributed shared file system |
US8352501B2 (en) * | 2010-01-28 | 2013-01-08 | Cleversafe, Inc. | Dispersed storage network utilizing revision snapshots |
US8972691B2 (en) * | 2011-11-03 | 2015-03-03 | International Business Machines Corporation | Addressing cross-allocated blocks in a file system |
-
2013
- 2013-12-31 CN CN201310752681.2A patent/CN103678715B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101162469A (en) * | 2007-11-09 | 2008-04-16 | 清华大学 | Fine grit document and catalogs version management method based on snapshot |
CN102024018A (en) * | 2010-11-04 | 2011-04-20 | 曙光信息产业(北京)有限公司 | On-line recovering method of junk metadata in distributed file system |
CN102073739A (en) * | 2011-01-25 | 2011-05-25 | 中国科学院计算技术研究所 | Method for reading and writing data in distributed file system with snapshot function |
CN102722536A (en) * | 2012-05-22 | 2012-10-10 | 中兴通讯股份有限公司 | Method and device for realizing snapshot based on distributed file system |
Also Published As
Publication number | Publication date |
---|---|
CN103678715A (en) | 2014-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103678715B (en) | The metadata information management method of snapshot is supported in a kind of distributed file system | |
US11093466B2 (en) | Incremental out-of-place updates for index structures | |
CN105677250B (en) | The update method and updating device of object data in object storage system | |
US11516288B2 (en) | Synchronized content library | |
CN109413127B (en) | Data synchronization method and device | |
US9141623B2 (en) | System for on-line archiving of content in an object store | |
CN106407250A (en) | Information query method, device and system, server and client side | |
US10817472B2 (en) | Storage organization system with associated storage utilization values | |
CN106462430A (en) | Method and device for obtaining application upgrade packages | |
CN102629247A (en) | Method, device and system for data processing | |
CN104158897B (en) | The update method of file layout in a kind of distributed file system | |
US10579642B2 (en) | Techniques for administering comingled content items | |
US10884980B2 (en) | Cognitive file and object management for distributed storage environments | |
US9411526B1 (en) | Filesystem management | |
CN103198108B (en) | A kind of index data update method, retrieval server and system | |
GB2512856A (en) | Method for file handling in a hierarchical storage environment and corresponding hierarchical storage environment | |
CN104281414A (en) | Distributed file system and small file access method thereof | |
US11307922B2 (en) | Application crash analysis techniques when memory dump and debug symbols are not co-located | |
CA3066254A1 (en) | Systems and methods of creation and deletion of tenants within a database | |
US9734171B2 (en) | Intelligent redistribution of data in a database | |
Sarkar et al. | Query language support for timely data deletion | |
WO2019222201A9 (en) | Cloud storage format to enable space reclamation while minimizing data transfer | |
CN102737082A (en) | Method and system for dynamically updating file data indexes | |
US8214336B2 (en) | Preservation of digital content | |
US20150220599A1 (en) | Automobile airbag deployment dependent on passenger size |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |