CN102012933B - Distributed file system and method for storing data and providing services by utilizing same - Google Patents
Distributed file system and method for storing data and providing services by utilizing same Download PDFInfo
- Publication number
- CN102012933B CN102012933B CN 201010570126 CN201010570126A CN102012933B CN 102012933 B CN102012933 B CN 102012933B CN 201010570126 CN201010570126 CN 201010570126 CN 201010570126 A CN201010570126 A CN 201010570126A CN 102012933 B CN102012933 B CN 102012933B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- file system
- server
- node
- 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
Images
Abstract
The invention discloses a distributed file system which comprises a metadata server, a data server, a file system client and an application program client, wherein the metadata server is used for saving a list of nodes of the data server, which correspond to all user identifiers, and the corresponding relationship between the nodes and the identifiers, maintaining the load situation of each node, guiding the data server to implement the operations of load balancing and redundancy recovery and representing the load situation by data size; the data server is used for storing data of users and corresponding metadata of the users, leading the data of each user to be stored under an individual directory, and using the identifiers to differentiate the directories; the file system client is used for forming a local file directory and forwarding the operations to the corresponding data server; and the application program client is used for leading the user to access the file system through the file directory formed by the file system client and further providing an application program for the user. The invention further provides a method for storing the data and providing the services by utilizing the system. By adopting the system and the method, the efficiency of storing the data of the individual users can be improved.
Description
Technical field
The present invention relates to the field of filesystems of operating system, the method for particularly serving magnanimity personal user's distributed file system and utilizing it to store personal user's data and service is provided for the personal user.
Background technology
Along with the development of cloud computing technology and development of Mobile Internet technology, the remote service end has been transferred to from local device (such as desktop computer, net book, smart mobile phone etc.) in the memory location of increasing users personal data.This has brought a lot of facilities for user's use.
At first, such memory module is to the synchronous chance of having created between the data on the distinct device.The data such as user's document, picture, music may create or produce at desktop computer, but also want to check or revise at net book or smart mobile phone.Because data have a always readily available data at the remote service end, application program can be carried out by certain agreement the automatic synchronization of data, trouble and the problem of having avoided the manual copy data of user to bring.
Secondly, such memory module has been created chance to the recovery of data.Delete local data when the user misses, or after the lost devices, because the remote service end has also kept user's data, therefore can recover data by application program.
The 3rd, such memory module has been created chance to the data sharing between the different user.When certain user want to the shared data of other users, he can pass to data the remote service end, and is labeled as shared.Other can be accessed these through the user who authorizes and be marked as shared data.
Above-mentioned memory module can provide to the user facility of above these usage datas, has greatly improved the availability of user data, is typical application model of mobile Internet.A problem of meanwhile bringing is for a service, support the user of magnanimity just must have a large capacity, file system highly reliable, easy expansion that storage is provided.Google company set forth the distributed file system of using in Google intra-company in 2003 in the paper The Google File System that The ACM Symposium on Operating Systems Principles (SOSP) meeting is delivered, have large capacity, the characteristic such as highly reliable, fault-tolerant, and be optimized for a large amount of index of searching engine field and data cached characteristics, for the business of Google company provides powerful support platform.Its problem is that mainly the data characteristics of searching engine field is a large amount of index and the data cached one-time write that is, and mainly to append as main, efficient can be lower when the personal data of processing a large number of users need frequent situation of rewriting for alter mode.
Summary of the invention
The technical matters that (one) will solve
The technical problem to be solved in the present invention is: the efficient that how to improve storage personal user data.
(2) technical scheme
For solving the problems of the technologies described above, the invention provides a kind of distributed file system, comprising:
Meta data server, be used for preserving the tabulation of data server node corresponding to all user identifiers and the corresponding relation of this node and its identifier, safeguard the loading condition of each node, and guide data server to carry out load balancing and the redundant operation that recovers, described loading condition characterizes with data volume;
Data server is used for storage user's data and corresponding user's metadata, and wherein each user's data are stored under the independent catalogue, and use described identifier that catalogue is distinguished;
File system client is used to form local file directory, and operation is forwarded to corresponding data server;
Application client is used for making user's access file system by the formed file directory of described file system client, thereby provides application program for the user.
The present invention also provides a kind of method of utilizing above-mentioned distributed file system storage user data, and each user's data are stored under the independent catalogue, and user's identifier is distinguished catalogue.
Wherein, described user is the personal user.
The present invention also provides a kind of method of utilizing above-mentioned distributed file system that service is provided for the user, described distributed file system is offered application program use, and provides service by this application program to the user again.
Wherein, described user is the personal user.
(3) beneficial effect
The present invention is optimized for mass users data storage mode, each user's personal data are stored under the independent catalogue, and utilize user identifier (User Identifier is called for short later on user ID) that these catalogues are carried out division on data server.Can save metadata like this, raise the efficiency.Simultaneously, when the user uses his file, because file all is placed on same number of units according on the server, so can look ahead efficiently and buffer memory, further raise the efficiency.Distributed file system directly is not exposed to the terminal user and uses, use but offer traditional application program, provide service by these service routines to the terminal user again, given application program design the semantical definition of concrete application like this, simplified the design of distributed file system.
In addition, system of the present invention can compatible existing local file system, and the interface of Virtual File System (Virtual File System, VFS) is provided, make things convenient for existing application to be transplanted to very soon in the file system, do not need too many odd word.Another advantage is to provide extra function by file system structure, for example can set up for each user the index of oneself, can carry out full-text search etc.Can also when a part of data server node breaks down, still guarantee the high reliability of system.This system need not keeper's participation, can automatically carry out the configuration effort of system, automatically realizes data balancing and to the processing of misdata.
Description of drawings
Fig. 1 is the system architecture schematic diagram of the embodiment of the invention;
Fig. 2 has shown the organizational form of user data on DS.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention will be further described.
As shown in Figure 1, distributed file system provided by the invention comprises:
Meta data server (Metadata Server, MDS), be used for preserving the tabulation of data server node corresponding to all user identifiers and the corresponding relation of this node and its identifier, safeguard the loading condition of each node, and guide data server to carry out load balancing and the redundant operation that recovers, described loading condition characterizes with data volume;
Data server (Data Server, DS) is used for storage user's data and corresponding user's metadata, and wherein each user's data are stored under the independent catalogue, and use described identifier that catalogue is distinguished;
File system client (File System Client, FSC) is used to form local file directory, and operation is forwarded to corresponding data server;
Application client (Application Client, AC) is used for making user's access file system by the formed file directory of described file system client, thereby provides application program for the user.
Above-mentioned user is preferably the personal user, and personal user's data refer to the data that the non-overall situation is unified.
The below introduces the modules of said system in detail.
MDS is the core of whole system, and its groundwork is: (1) response FSC is for some users' query requests; (2) send heartbeat message to each DS node, and receive response; Situation about (3) going offline according to DS node is upgraded DataServerStatusList (tabulation of DS node information), and the information among the corresponding UserStatusList (user profile of safeguarding) is upgraded; (4) upgrade above-mentioned UserStatusList, and needs are carried out user that data heavily distribute initiate the data work that heavily distributes, according to the status architecture of the information updating user data of response; (5) processing needs when new node adds report for work to MDS first for the adding of DS new node; (6) MDS inquires that each DS goes the data structure in the internal memory of structure own when just beginning to start; (7) for not online for a long time, and the DS that reached the standard grade again afterwards startup garbage collection program, out-of-date user data (moving in some rubbish catalogues) deleted; (8) if find that some nodes are not online in one period considerable time, by Email or other means reporting systems keeper, so that the keeper can decide by some technological means next step work.
DS is used for the storage user data, and each user's data are assigned to the upper storage of several DS with the form of a top-level directory.DS itself is a passive equipment, waits for the read-write requests of FSC, waits for the various command of MDS, and makes corresponding answer or reaction.
The client of system of the present invention is divided into two parts, a part is FSC, it is provided at this distributed file system of carry on the local file system by some existing technological means, forms local folders or a driver that can conduct interviews with the DLL (dynamic link library) that operating system itself provides.Another part is AC, and it is the program that service is provided to the terminal user, stores user's personal data by the file system interface that FSC provides.
The distribution of the data of storing in the native system on DS is take the user as unit, and all data of unique user logically form the top file of a file system, and take so top file as unit, assign on the different DS by the coordination of MDS.Such distribution mode with take Single document as unit, perhaps the data block of file is that the mode of unit is compared, and can greatly reduce the amount of extra metadata, simultaneously for the buffer memory of file with look ahead and offer an opportunity.
The present invention also provides a kind of method of utilizing above-mentioned distributed file system storage personal user data, and each user's data are stored under the independent catalogue, and user's identifier is distinguished catalogue.
The present invention also provides a kind of method of utilizing above-mentioned distributed file system that service is provided for the personal user, described distributed file system is offered application program use, and provides service by this application program to the user again.
In specific implementation process, need to safeguard relevant data in each module, and operate according to certain flow process.
MDS will safeguard the information of the whole distributed file system overall situation, and the event that affects the overall situation is made a response.System's overall situation has several configurable parameters, respectively user file number of copies n (recommended value is 3) under normal circumstances, the time interval h of heartbeat detection (recommended value is 5 minutes), and judge the time interval d (recommended value is 2 hours) that goes offline.
The data that need to safeguard in MDS are divided into two parts, are respectively the data of DS situation and the data of user situation, to adapt to different processing application scenarioss.
(1) tabulation of DS node information:
Table 1
(2) user profile of safeguarding:
Table 2
When MDS just starts, the DS tabulation of a startup need to be arranged, in order to inquire the situation of each DS at MDS, thereby obtain concrete server situation.
The groundwork flow process of distributed file system of the present invention:
1, the request result of cache request (FSC can) of querying node is carried out in response to certain user.
A) AC initiates a request, creates a new catalogue below/user/, as/user/newuserid.
B) corresponding request is forwarded to FSC, and FSC will initiate a request this moment, to MDS request user profile.
C) MDS receives above-mentioned request, and the inquiry local data base returns to FSC with the corresponding situation of node in the database, and FSC receives after the information and corresponding DS communication, and FSC is the buffer memory relevant information always, until information is expired.MDS does not participate in the work of following relevant data read-write.
D) MDS receives above-mentioned request, the inquiry local data base, and finding does not have corresponding user, then begins to enter the work that creates user's corresponding with service device node mapping.(User Catalog does not have in this locality, and client need to be asked, and should be the new user who creates, and therefore need to create in metadata for corresponding of user.)
E) MDS is according to the tabulation of DS node information, generates the tabulation of n DS according to some algorithms, and with this list storage in the database at the place of MDS own.
F) MDS returns to FSC with the tabulation of this n node.Afterwards, FSC will carry out communication with the DS of correspondence.
2, regularly send heartbeat message to each DS that in DataServerStatusList, occurs, and collect response message.Regularly the server on backstage is monitored that (initiation of monitoring constantly can be regularly, also can be the MDS free time, be removed to determine the normal metadata operation of Priority Service by these two factors), if find the node of operation irregularity, then carry out the operation that data heavily distribute.
A) safeguard the heartbeat message of all nodes, periodic monitor, and refresh the status information of the corresponding node on MDS.(in the time of refreshing information, not only need to carry out the inspection whether node moves, need also to check whether corresponding DS works.Want to respond heartbeat message at DS.) when finding that node goes offline, need updating user data available copies number.
B) in the above when Flushing status information, check the state of the abnormal DS node of operation.Time surpasses d if abnormal node rolls off the production line, and then starts the request that heavily distributes for data.
3, upgrade node status information, if find that some nodes are not online for a long time, then go to upgrade corresponding user data tabulation, and if necessary, initiate the heavy distributed operation of data.At this moment, exactly the corresponding ReplicaStatus of the corresponding all data backups of this node is upgraded.Also need at last to delete the data structure that the DS for error condition keeps.
4, upgrade UserStatusList, and needs are carried out user that data heavily distribute initiate the data work that heavily distributes, according to the status architecture of the information updating user data of response.When initiating renewal, choose at random current available back end (being partial to the fewer node of data), and be inserted among the UserHostList of User Status, and the ReplicaStatus state of correspondence is set to RECOVERING, is set to HEALTHY until receive a response message state.
5, process the situation that the DS new node adds, when receiving that some nodes add the message of system, the information of node is inserted into DataServerStatusList.
6, MDS inquires that each DS goes the data structure in the internal memory of structure own when just beginning to start.The process of reconstruct will be inquired the state of each DS, reconstructs two above-mentioned tabulations, i.e. DataServerStatusList and UserStatusList.
If 7 some DataServer are not online for a long time, but reached the standard grade again afterwards, and in MDS reconstruct data itself, can run into the inconsistent situation of data.This situation solves by version number, is chosen at the formed version number of most data in the system, abandons inconsistent data, puts in the refuse collection catalogue.
8, refuse collection.For not online for a long time, and the DS that reached the standard grade again afterwards startup garbage collection program, out-of-date user data deleted.
9, status monitoring and notice.If find some nodes not line duration surpass d, by Email or other means reporting systems keeper, so that the keeper can decide by some technological means next step work.
DS itself is a passive equipment, waits for the read-write requests of FSC, waits for the various command of MDS, and makes corresponding answer or reaction.
DS itself has preserved user's data.The organizational form of user data on DS as shown in Figure 2, the user data of "/user data/user ID 1 " expression user ID 1 correspondence, App1～App3 represents that respectively application program 1 is to application program 3.There is the configuration file of an overall situation to specify the deposit position of these data at DS, and the catalogue of depositing the junk data that may occur.
Access interface and the treatment scheme of DS are as follows:
The access interface that DS provides comprises that one is the read-write operation for file under the corresponding catalogue that comes from FSC; One is to carry out the request interface that data heavily distribute from what MDS came; Last is the user data that sends over from other DS, the request that response heavily distributes with data; Also have one to be the request (the scanning of home user ID is tabulated, and reports to MDS) of reconstruct MDS.Heartbeat and newly-increased DS notice MDS is unidirectional in addition.
1. the information that sends according to FSC is carried out read-write operation to the catalogue of correspondence.
2. the order of sending according to MFS weighs distributed operation to data.The heavy distribution command that MFS sends is divided into two parts, namely needs the ID users of the user data that heavily distributes, and destination node.
3. the user data that sends over from other DS.
4. respond the heartbeat message that MDS sends over.In heartbeat message incidentally for the information of user data operating position, for finishing information that data heavily distribute etc.
5. new DS reports from trend MDS.
AC goes to read and write the particular category under the respective user.Client's file data carries out unified management according to/user/useridxxxx/service/file.Each AC needs to rewrite automatically the file under the corresponding catalogue is read and write.In the realization under the (SuSE) Linux OS, read-write operation is transmitted to FSC by FUSE (file system of user's space).FSC is the application program of a FUSE, plays two effects:
1. be used for producing a special directory in the local file system, the file under this special directory in fact all is the special file system of transmitting by FSClient.
2. be used for long-range file is read and write.
The workflow of FSC is as follows:
1.FSC at first remove to read and write file in some files by VFS (Virtual File System) interface, for example/and user/useridxxxx/service/file, all file read-write interfaces under/user/useridxxxx/service will be forwarded to the service routine of FSC so.
2.AC be transmitted to FSC for the operation meeting of file or the catalogue interface by FUSE.
If FSC buffer memory the position of DS at corresponding user place, then need not from MDS, to obtain user's metadata, otherwise to the metadata of MDS request relative users.
4. if read operation, FSC selects nearest server to read according to the strategy of this locality, perhaps according to the load state of node, selects load the lightest.
5. if write operation carries out according to the agreement that writes hereinafter.
Based on above-mentioned composition module, the detailed process of the operations such as writing, read, open, close of file is described below.
The write operation of file:
1.FSC receive write operation.
If in FSC this locality buffer memory this user's nodal information, then directly use; Otherwise inquire about this user's nodal information to MDS.
3. the order of tabulating according to user's nodal information attempts sending write request successively.
4. during the node data writing, at first be written to the buffer memory of corresponding document in the internal memory, then write operation be written to local disk as daily record, wait for writing the write operation success of returning after finishing on this DS node for twice.
5. if a certain step writes and mistake occurs, then return mistake to FSC, FSC returns the write operation failure.
6. log management service: will be to the log management service registry when file writes daily record first, size of audit log all when at every turn writing daily record is if surpass threshold value then notify the daily record management service.The daily record of all registrations of log management service scan is applied to all journalizings in the original file.
The read operation of file:
1.FSC receive read operation.
If in FSC this locality buffer memory this user's nodal information, then directly use; Otherwise inquire about this user's nodal information to MDS.
3. the main DS in tabulating to user node sends the read operation request.
4. this end of main DS is looked for first data from memory cache, if then do not look for from Disk Logs, if also there not be file on the direct reading disk then.Reply request after reading corresponding contents.
5.FSC return the content that reads.
The opening operation of file:
1.FSC receive the request of opening file.
If in FSC this locality buffer memory this user's nodal information, then can directly return OK.
3. otherwise FSC sends request to MDS, the nodal information that the Asynchronous Request user is corresponding and buffer memory also return OK.
The shutoff operation of file:
1.FSC receive the close file request.
2.FSC return OK.
Above embodiment only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; in the situation that do not break away from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (5)
1. a distributed file system is characterized in that, comprising:
Meta data server, be used for preserving the tabulation of data server node corresponding to all user identifiers and the corresponding relation of this node and its identifier, safeguard the loading condition of each node, and by come the guide data server to carry out load balancing and the redundant operation that recovers to the data server sending controling instruction, described loading condition characterizes with data volume;
Data server, be used for storage user's data and corresponding user's metadata, wherein each user's data are stored under the independent catalogue, and each user's All Files leaves same number of units in according on the server, and use described user identifier that catalogue is distinguished;
File system client is used to form local file directory, and the file read-write operation is forwarded to corresponding data server;
Application client is used for making user's access file system by the formed file directory of described file system client, thereby provides application program for the user.
2. method of utilizing distributed file system claimed in claim 1 storage user data, it is characterized in that, each user's data are stored under the independent catalogue, and each user's All Files leaves same number of units in according on the server, and user's identifier is distinguished catalogue.
3. method as claimed in claim 2 is characterized in that, described user is the personal user.
4. a method of utilizing distributed file system claimed in claim 1 that service is provided for the user is characterized in that, described distributed file system is offered application program use, and provides service by this application program to the user again.
5. method as claimed in claim 4 is characterized in that, described user is the personal user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010570126 CN102012933B (en) | 2010-12-02 | 2010-12-02 | Distributed file system and method for storing data and providing services by utilizing same |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010570126 CN102012933B (en) | 2010-12-02 | 2010-12-02 | Distributed file system and method for storing data and providing services by utilizing same |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102012933A CN102012933A (en) | 2011-04-13 |
| CN102012933B true CN102012933B (en) | 2013-01-30 |
Family
ID=43843106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010570126 Active CN102012933B (en) | 2010-12-02 | 2010-12-02 | Distributed file system and method for storing data and providing services by utilizing same |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102012933B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102385633B (en) * | 2011-12-13 | 2014-03-19 | 方正国际软件有限公司 | File management system using virtual storage directory |
| CN102546780B (en) * | 2011-12-28 | 2014-04-16 | 山东大学 | Operation method for file distributed storage based on thin client |
| CN102571959B (en) * | 2012-01-11 | 2015-05-06 | 北京奇虎科技有限公司 | System and method for downloading data |
| CN103685348B (en) * | 2012-09-03 | 2016-12-21 | 百度在线网络技术（北京）有限公司 | The system and method that cloud performs |
| CN103036948B (en) * | 2012-11-21 | 2015-12-02 | 北京航空航天大学 | Namely network file processing method, XM, software serve SaaS platform |
| CN103064635B (en) * | 2012-12-19 | 2016-08-24 | 华为技术有限公司 | Distributed storage method and distributed storage devices |
| CN103973744A (en) * | 2013-02-01 | 2014-08-06 | 北京英富森信息技术有限公司 | Distributed file progressive storage technology |
| CN103327116B (en) * | 2013-07-05 | 2016-06-01 | 山东大学 | The dynamic copies storage means of a kind of network file |
| CN104050282B (en) * | 2014-06-27 | 2017-08-01 | 北京思特奇信息技术股份有限公司 | Resource file update method and device |
| CN105069024B (en) * | 2015-07-16 | 2018-08-03 | 清华大学 | Distributed file system write access method towards parallel data acquisition |
| CN105404546A (en) * | 2015-11-10 | 2016-03-16 | 上海交通大学 | RDMA and HTM based distributed concurrency control method |
| CN105955801B (en) * | 2015-12-21 | 2018-12-18 | 上海交通大学 | A kind of distributed optimistic concurrency control method based on RDMA and HTM |
| CN107612763B (en) * | 2017-11-08 | 2020-10-02 | 浪潮通用软件有限公司 | Metadata management method, application server, service system, medium and controller |
| CN108366087B (en) * | 2017-12-26 | 2021-03-05 | 深圳创新科技术有限公司 | ISCSI service realization method and device based on distributed file system |
| CN111324293B (en) * | 2018-12-14 | 2022-08-05 | 杭州海康威视系统技术有限公司 | Storage system, data storage method, data reading method and device |
| CN109981798A (en) * | 2019-04-12 | 2019-07-05 | 苏州浪潮智能科技有限公司 | A kind of method for writing data, system and electronic equipment and storage medium |
| CN111309796B (en) * | 2020-02-07 | 2023-09-26 | 腾讯科技（深圳）有限公司 | Data processing method, device and computer readable storage medium |
| CN112597120B (en) * | 2020-11-30 | 2022-05-27 | 新华三大数据技术有限公司 | Catalog management method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1444149A (en) * | 2002-03-12 | 2003-09-24 | 中国科学院计算技术研究所 | Server system based on network storage and expandable system structure and its method |
| US7363352B2 (en) * | 2003-03-13 | 2008-04-22 | Hitachi, Ltd. | Method for accessing distributed file system |
| CN101227460A (en) * | 2007-01-19 | 2008-07-23 | 秦晨 | Method for uploading and downloading distributed document and apparatus and system thereof |
| WO2010079713A1 (en) * | 2009-01-08 | 2010-07-15 | 日本電気株式会社 | Distributed file name resolution system, distributed file name resolution method, and distributed file name resolution program |
-
2010
- 2010-12-02 CN CN 201010570126 patent/CN102012933B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1444149A (en) * | 2002-03-12 | 2003-09-24 | 中国科学院计算技术研究所 | Server system based on network storage and expandable system structure and its method |
| US7363352B2 (en) * | 2003-03-13 | 2008-04-22 | Hitachi, Ltd. | Method for accessing distributed file system |
| CN101227460A (en) * | 2007-01-19 | 2008-07-23 | 秦晨 | Method for uploading and downloading distributed document and apparatus and system thereof |
| WO2010079713A1 (en) * | 2009-01-08 | 2010-07-15 | 日本電気株式会社 | Distributed file name resolution system, distributed file name resolution method, and distributed file name resolution program |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102012933A (en) | 2011-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102012933B (en) | Distributed file system and method for storing data and providing services by utilizing same | |
| US11704290B2 (en) | Methods, devices and systems for maintaining consistency of metadata and data across data centers | |
| JP5918244B2 (en) | System and method for integrating query results in a fault tolerant database management system | |
| US20180046552A1 (en) | Variable data replication for storage implementing data backup | |
| CN101888405B (en) | Cloud computing file system and data processing method | |
| TWI232382B (en) | A distributed storage system for data-sharing among client computers running different operating system types | |
| CN101743546B (en) | Hierarchical storage management for a file system providing snapshots | |
| JP5567342B2 (en) | Network data storage system and data access method thereof | |
| US8793531B2 (en) | Recovery and replication of a flash memory-based object store | |
| US8412685B2 (en) | Method and system for managing data | |
| CN101103355B (en) | Methods and apparatus for managing deletion of data | |
| CN101866305B (en) | Continuous data protection method and system supporting data inquiry and quick recovery | |
| CN102629247B (en) | Method, device and system for data processing | |
| US7376681B1 (en) | Methods and apparatus for accessing information in a hierarchical file system | |
| US20080010325A1 (en) | Data migration apparatus, method, and program | |
| EP0663641A1 (en) | Summary catalogs | |
| JP2013545162A5 (en) | ||
| JP2008518284A (en) | Metadata management for fixed content distributed data storage | |
| KR101078287B1 (en) | Method Recovering Data Server at the Applying Multiple Reproduce Dispersion File System and Metadata Storage and Save Method Thereof | |
| CN103460197A (en) | Computer system, file management method and metadata server | |
| US7366836B1 (en) | Software system for providing storage system functionality | |
| CN100498781C (en) | Method for storing metadata of logic document system by adhesion property | |
| JPWO2007032046A1 (en) | HSM control program, HSM control device, and HSM control method | |
| CN113377868A (en) | Offline storage system based on distributed KV database | |
| JP5241298B2 (en) | System and method for supporting file search and file operations by indexing historical file names and locations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |