CN101997823B - Distributed file system and data access method thereof - Google Patents
Distributed file system and data access method thereof Download PDFInfo
- Publication number
- CN101997823B CN101997823B CN2009100912919A CN200910091291A CN101997823B CN 101997823 B CN101997823 B CN 101997823B CN 2009100912919 A CN2009100912919 A CN 2009100912919A CN 200910091291 A CN200910091291 A CN 200910091291A CN 101997823 B CN101997823 B CN 101997823B
- Authority
- CN
- China
- Prior art keywords
- server
- meta data
- client
- identifier
- data server
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a distributed file system and a data access method thereof. The system comprises at least one client server, a metadata central server, at least one metadata server and at least one object server, wherein the at least one client server is used for receiving an access request from a client; the metadata central server is used for returning a currently available global uniquemetadata server identifier of the metadata server to the client server according to the access request of the client; the at least one metadata server is used for receiving file information transmitted according to the global unique identifier of the metadata server by the client server, and returning a currently available global unique object server identifier of the object server to the client server according to the file information; and the at least one object server is used for generating an access result in response to access contents transmitted according to the global unique identifier of the object server by the client server, and returning the access result to the client by the client server. The distributed file system and the data access method thereof greatly improve the reliability of the system at the same time of not influencing the data transmission efficiency.
Description
Technical field
The present invention relates to computer memory technical field, refer to a kind of distributed file system and data access method thereof especially.
Background technology
In the Computer Storage field, cloud computing (Cloud Computing) is the development of distributed treatment (DistributedComputing), parallel processing (Parallel Computing) and grid computing (Grid Computing), or perhaps the commerce of these computer science concepts of distributed treatment, parallel processing and grid computing realizes.The basic principle of cloud computing is: by calculating being distributed on a large amount of distributed computers, but not in local computer or the remote server, the operation of the data center of enterprise is then similar to the Internet, make the enterprise can be with resource switch to the application of needs, according to demand access computer and storage system.In other words, cloud computing is exactly to utilize the particular network computer cluster to realize the function that hardware on the present home PC, software and some memory device have.
As shown in Figure 1, be the cloud computing system configuration diagram, client is told the demand of this client of cloud, and last cloud just can satisfy the demand of this client, and client need not be managed the details of resource distribution and flow process.The demand that client proposes, such as: need great computing capability, need to dispose which type of software, need do which type of test.Remaining thing client just all need not have been managed.Can give IP address of client after the work of cloud is finished, client is used the IT environment that satisfies his demand by this IP address, and the client here can be to be distributed in the whole world individual PC or the PC of enterprise of local access network arbitrarily.Specifically, it is the demand that is received client by the client-server in the cloud storage system, by client-server this demand is analyzed, search meta data server, find corresponding object server by meta data server again, this object server is generally the equipment with memory function, and the data that this object server will satisfy client demand return to client.
What above-mentioned distributed file system was taked is the object storage mode, all be the object storage mode of taking as above-mentioned as these systems of Lustre, GlusterFS, this has improved data transmission efficiency to a certain extent, but the administrative section to metadata only provides the single-point pattern at present, be that cloud has only a meta data server, in case fault has appearred in this meta data server, the reliability of whole system can't guarantee.
The inventor finds that there are the following problems at least in the prior art in realizing process of the present invention:
Existing cloud computing memory technology can't improve the reliability of data preferably when improving data transmission efficiency.
Summary of the invention
The technical problem to be solved in the present invention provides a kind of distributed file system and data access method thereof, makes the reliability that can improve data based on the memory technology of distributed file system when improving data transmission efficiency.
For solving the problems of the technologies described above, embodiments of the invention provide technical scheme as follows:
On the one hand, provide a kind of distributed file system, comprising: at least one client-server, at least one meta data server, at least one object server, and metadata central server;
Described at least one client-server is for the access request that receives client;
Described metadata central server is used for the access request according to described client, returns overall unique meta data server identifier of current available meta data server to described at least one client-server;
Described at least one meta data server, be used for to receive the fileinfo that described at least one client-server sends according to overall unique meta data server identifier of described meta data server, and return overall unique object server identifier of current available object server according to described fileinfo to described client-server;
Described at least one object server, be used for described at least one client-server of response according to the accessed content of overall unique object server identifier transmission of described object server, produce the visit result, and described visit result is returned described client via described client-server.
Preferably, above-mentioned distributed file system also comprises:
Metadata checks server, is used at least three meta data servers of described at least one meta data server are made the data redundancy backup.
Preferably, above-mentioned meta data server comprises:
Memory cell is for overall unique object server identifier of described at least one object server of storage;
Described meta data server obtains overall unique object server identifier of current available object server according to described fileinfo from described memory cell, and returns overall unique object server identifier of described object server to described client-server.
Preferably, above-mentioned distributed file system also comprises:
The object centers server, the overall unique object server identifier that is used for described at least one object server of storage, and receive the fileinfo that described at least one client-server that described meta data server transmits sends according to overall unique meta data server identifier of described meta data server, and return overall unique object server identifier of current available object server to described at least one meta data server according to described fileinfo.
Preferably, above-mentioned distributed file system also comprises:
Object checks server, is used at least three object servers of described at least one object server are made the data redundancy backup.
Preferably, above-mentioned distributed file system also comprises:
The backup object central server is used for described object centers server is backed up.
Preferably, above-mentioned distributed file system also comprises:
Backup metadata central server is used for described metadata central server is backed up.
On the other hand, also provide a kind of data access method of distributed file system, comprising:
Receive the access request of client, and send described access request;
Receive overall unique meta data server identifier of the current available meta data server that the metadata central server returns according to described access request;
Overall unique meta data server identifier according to described meta data server produces fileinfo, and sends described fileinfo;
Receive overall unique object server identifier of the current available object server that described meta data server returns according to described fileinfo;
Overall unique object server identifier according to described object server sends accessed content;
Receive the visit result that described object server produces according to described accessed content, and described visit result is returned to described client.
Preferably, described overall unique meta data server identifier according to described meta data server produces fileinfo, and also comprises after sending described fileinfo:
At least three meta data servers are made the data redundancy backup.
Preferably, also comprise before overall unique meta data server identifier of the current available meta data server that returns according to described access request of described reception metadata central server:
The metadata central server receives the registration request of meta data server, produce overall unique meta data server identifier of described meta data server, and overall unique meta data server identifier of described meta data server is returned to described meta data server;
Meta data server receives the registration request of object server, produces overall unique object server identifier of described object server, and overall unique object server identifier of described object server is returned to described object server.
Embodiments of the invention have following beneficial effect:
Pass through to increase the metadata central server in the said system scheme, be used for managing a plurality of meta data servers, the UUID of current available meta data server is returned to client-server, because what store in the metadata central server is a small amount of key messages such as the UUID of these meta data servers and machine address, therefore, can guarantee the efficiency of transmission of data, simultaneously owing to there are a plurality of meta data servers to exist, when one of them breaks down, the available meta data server that can search other by the metadata central server has guaranteed the reliability of data like this as substitute.
Description of drawings
Fig. 1 is existing cloud computing system configuration diagram;
Fig. 2 is the configuration diagram of embodiments of the invention distributed file system;
Fig. 3 is a specific embodiment schematic diagram of the framework of distributed file system shown in Figure 2;
Fig. 4 is another specific embodiment schematic diagram of the framework of distributed file system shown in Figure 2;
Fig. 5 is the another specific embodiment schematic diagram of the framework of distributed file system shown in Figure 2;
Fig. 6 is the data access method schematic flow sheet of embodiments of the invention distributed file system;
Fig. 7 is the concrete application flow schematic diagram of method shown in Figure 6.
Embodiment
For technical problem, technical scheme and advantage that embodiments of the invention will be solved is clearer, be described in detail below in conjunction with the accompanying drawings and the specific embodiments.
Embodiments of the invention can't improve the reliability problems of data at cloud storage system in the prior art when improving data transmission efficiency, the method for a kind of distributed file system and data access thereof is provided.
As shown in Figure 2, the embodiments of the invention distributed file system, comprise: at least one client-server (Client Server), at least one meta data server (Meta Data Server), at least one object server (Object Server), and metadata central server (Meta Data CenterMaster);
Wherein, at least one client-server, for the access request that receives client, this client can be distribute individual PC or the PC of enterprise or other communication apparatus of any access network in the world; The access request of client mainly comprises: some demands of client, as the great computing capability of needs, need to dispose which type of software, and need do which type of test etc.
The metadata central server, be used for the access request according to described client, return the GUID UUID (Universally Unique Identify) of current available meta data server, i.e. meta data server identifier to described at least one client-server; This metadata central server is responsible at least one meta data server is managed, store the UUID of these meta data servers, the data that machine address etc. are considerably less guaranteeing the operational efficiency of this metadata central server, and then guarantee data transmission efficiency;
At least one meta data server is used for receiving described at least one client-server according to the fileinfo of the UUID transmission of meta data server, and returns the UUID of described object server to described client-server according to described fileinfo;
Client-server is behind the UUID that receives the meta data server that the metadata central server returns, can be according to the UUID spanned file information of this meta data server, this document information spinner will comprise: filename, file ID and relevant information, and this document information is sent to meta data server; Meta data server is again according to the UUID of the current available object server of this document information searching, and the UUID of the object server that finds is returned to client-server;
At least one object server is used for described at least one client-server of response according to the accessed content of the UUID transmission of object server, produces the visit result, and described visit result is returned described client via described client-server;
Client-server is behind the UUID that receives the object server that meta data server returns, can send accessed content according to UUID and the file ID in the fileinfo of this object server, for example read file to the object server written document or from object server;
Object server responds this accessed content, produce the visit result, for example accessed content if client-server to the object server written document, then client-server can send file content to object server, when writing when finishing, object server can return one to client-server and write the report information of finishing; If client-server is read file content from object server, then client-server can send to object server with file ID, object server returns to client-server according to this document ID with relevant file content, the file content that will be read from object server by client-server returns to client again, and then satisfies the demand of client.
This embodiment is by increasing the metadata central server in distributed file system, be used for managing a plurality of meta data servers, store the UUID of these meta data servers, when one of them meta data server breaks down, this metadata central server can be searched current available meta data server, and the UUID of this meta data server that finds returned to client-server, because what store in the metadata central server is a small amount of key messages such as the UUID of these meta data servers and machine address, therefore, can guarantee the efficiency of transmission of data, simultaneously owing to there are a plurality of meta data servers to exist, when one of them breaks down, the available meta data server that can search other by the metadata central server has guaranteed the reliability of data like this as substitute.
As shown in Figure 3, on above-mentioned basis embodiment illustrated in fig. 2, above-mentioned distributed file system also can comprise:
Metadata checks server (Meta Data Check), is used at least three meta data servers of described at least one meta data server are made the data redundancy backup.
This metadata checks that the effect of server is that data and corresponding parity information are stored in each meta data server, after wherein a meta data server is damaged, utilize remaining data and corresponding parity information to go to recover impaired data, be similar to RAID 5 technology, RAID 5 is storage solutions that a kind of memory property, data security and carrying cost are taken into account.RAID 5 does not back up the data of storage, but data and corresponding parity information are stored on each disk of forming RAID 5, and parity information is stored in respectively on the different disks with corresponding data.After the data in magnetic disk of RAID 5 is damaged, utilize remaining data and corresponding parity information to go to recover impaired data.The further like this reliability that improves the distributed file system data.
On above-mentioned Fig. 2 or basis embodiment illustrated in fig. 3, in the above-mentioned distributed file system, meta data server also can comprise: memory cell, UUID (the UniversallyUnique Identify that is used for all object servers of storage, GUID), i.e. object server identifier.
That is to say that meta data server is in charge of all object servers, store the UUID of object server, machine address, filename, file ID and file attribute (comprising file size etc.).
Metadata is exactly the information of descriptor, can comprise name space, access control information and the file of file to the map information of piece as metadata, and the position at the current place of piece etc.; Meta data server is just stored the server of these metadata, and above-mentioned object server is the server that stores concrete file content.
Certainly, in an embodiment of the present invention, the UUID of all object servers also can store in the object centers server specially, and this object centers server is responsible for the managing of these object servers, therefore,
As shown in Figure 4, on above-mentioned Fig. 2 or basis embodiment illustrated in fig. 3, above-mentioned distributed file system also can comprise:
The object centers server, the UUID that is used for all object servers of storage, and receive the fileinfo that described at least one client-server that described meta data server transmits sends according to the UUID of described meta data server, and return the UUID of current available object server to described at least one meta data server according to described fileinfo.Like this by all object servers of this object centers server admin, when wherein an object server breaks down, this object centers server can be searched other current available object server, to guarantee returning the data message that client needs to client.
As shown in Figure 5, same, also can in above-mentioned distributed file system, increase by an object and check server, at least three object servers at least one object server are made the data redundancy backup.
This object checks that server stores data and corresponding parity information in each object server into equally, after wherein an object server is damaged, utilizes remaining data and corresponding parity information to go to recover impaired data.The further like this reliability that improves the distributed file system data.
In order further to improve the reliability of embodiments of the invention distributed file system, in this system, the object centers server need be done a Hot Spare, and therefore, above-mentioned distributed file system also comprises:
The backup object central server is used for described object centers server is backed up, and Hot Spare preferably is so that this object centers server when breaking down, can be enabled this backup object central server, to guarantee the reliability of data.
Equally, for the metadata central server, also need to have a backup, preferably Hot Spare when the metadata central server breaks down in the system, is enabled this backup metadata central server, described metadata central server is backed up, further guarantee the reliability of data.
To sum up, above-mentioned distributed file system of the present invention can be applied in the cloud computing storage system, by increasing the metadata central server, a plurality of meta data servers are managed, when one of them meta data server breaks down, can from other available meta data server, search relevant metadata by this metadata central server, to satisfy the demand of client, not influencing substantially under the prerequisite of data transmission performance, improve the security reliability of system greatly; Also can further increase metadata simultaneously and check server, each meta data server is done the data redundancy backup, meta data server is being broken down, during corrupted data, can be restored, further guarantee the reliability of data in the system.As a same reason, also can check the server admin object server by increasing object centers server and object, further guarantee the reliability of data in the system; In addition, embodiments of the invention distributed file system administration extensions is more convenient, can increase or reduce meta data server and object server arbitrarily.
As shown in Figure 6, embodiments of the invention also provide a kind of data access method of distributed file system, comprising:
Step S61 receives the access request of client, and sends described access request; The access request of client mainly comprises: some demands of client, as the great computing capability of needs, need to dispose which type of software, and need do which type of test etc.;
Step S62 receives overall unique meta data server identifier UUID of the current available meta data server that the metadata central server returns according to described access request;
Step S63 according to overall unique meta data server identifier UUID generation fileinfo of described meta data server, and sends described fileinfo; This document information spinner will comprise: filename, file ID and relevant information;
Step S64 receives overall unique object server identifier UUID of the current available object server that described meta data server returns according to described fileinfo;
Step S65 is according to overall unique object server identifier UUID transmission accessed content of described object server;
Step S66 receives the visit result that described object server produces according to described accessed content, and described visit result is returned to described client.
This embodiment returns to client-server by UUID that will this current available meta data server, make client-server according to the UUID spanned file information of this meta data server, and the UUID that is returned current available object server by meta data server to client-server, make client-server send accessed content according to the UUID of this object server, returned the file content that satisfies client demand by object server, substantially do not influence under the prerequisite of data transmission efficiency, improved the reliability of data greatly.
In addition, as shown in Figure 7, the application flow with regard to above-mentioned method shown in Figure 6 is elaborated below:
Step S70, the metadata central server receives the registration request of meta data server;
Step S71, the metadata central server produces the UUID of meta data server according to described registration request, and returns the UUID of described meta data server to described meta data server;
Step S72, meta data server receives the registration request of object server;
Step S73, meta data server produces the UUID of object server according to described registration request, and returns the UUID of described object server to described object server;
Step S74, metadata central server receive the access request of the client of client-server forwarding;
Step S75, the metadata central server is searched the UUID of meta data server, and returns the UUID of current available meta data server to client-server;
After step S76, client-server receive the UUID of meta data server, produce fileinfo, and send this document information to meta data server;
Step S77, meta data server produces file ID according to this document information, and searches object server UUID; And to metadata inspection server transmission check result;
Step S78, metadata checks that server carries out the data redundancy backup to meta data server, to the meta data server return results, wherein this meta data server is at least three;
Step S79, meta data server return the UUID of object server to client-server;
Step S80, client-server sends accessed content according to object server UUID;
Step S81, object server be according to accessed content, produces the visit result, and will visit the result and return to client-server, returns the visit result who satisfies the client needs to client by client-server;
Among this step S81, client-server can be to the object server written document to the accessed content that object server sends, also can be to read file from object server, if client-server is to the object server written document, then client-server can send file content to object server, when writing when finishing, object server can return one to client-server and write the report information of finishing; If client-server is read file content from object server, then client-server can send to object server with file ID, object server returns to client-server according to this document ID with relevant file content, the file content that will be read from object server by client-server returns to client again, and then satisfies the demand of client.In this routine accompanying drawing 7, being that example is described to object server written document content.
To sum up, method embodiment of the present invention returns to client-server by the metadata central server with the UUID of current available meta data server, by the UUID generation relevant fileinfo of client-server according to this meta data server, and send to meta data server, return current available object server UUID by meta data server again, client-server sends accessed content according to the UUID of this object server again, object server responds this accessed content, return the visit result who satisfies client demand at last, this method is when guaranteeing data transmission efficiency, guaranteed the reliability of data, metadata checks that server also carries out redundancy backup to meta data server simultaneously, when meta data server breaks down, can utilize Backup Data that impaired data are recovered, further guarantee the reliability of system data.
The above is preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from principle of the present invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (10)
1. distributed file system comprises: at least one client-server, and at least one meta data server, at least one object server is characterized in that, also comprises the metadata central server;
Described at least one client-server is for the access request that receives client;
Described metadata central server is used for the access request according to described client, returns overall unique meta data server identifier of current available meta data server to described at least one client-server;
Described at least one meta data server, be used for to receive the fileinfo that described at least one client-server sends according to overall unique meta data server identifier of described meta data server, and return overall unique object server identifier of current available object server according to described fileinfo to described client-server;
Described at least one object server, be used for described at least one client-server of response according to the accessed content of overall unique object server identifier transmission of described object server, produce the visit result, and described visit result is returned described client via described client-server.
2. distributed file system according to claim 1 is characterized in that, also comprises:
Metadata checks server, is used at least three meta data servers of described at least one meta data server are made the data redundancy backup.
3. distributed file system according to claim 1 is characterized in that, described meta data server comprises:
Memory cell is for overall unique object server identifier of described at least one object server of storage;
Described meta data server obtains overall unique object server identifier of current available object server according to described fileinfo from described memory cell, and returns overall unique object server identifier of described object server to described client-server.
4. distributed file system according to claim 1 is characterized in that, also comprises:
The object centers server, the overall unique object server identifier that is used for described at least one object server of storage, and receive the fileinfo that described at least one client-server that described meta data server transmits sends according to overall unique meta data server identifier of described meta data server, and return overall unique object server identifier of current available object server to described at least one meta data server according to described fileinfo.
5. distributed file system according to claim 4 is characterized in that, also comprises:
Object checks server, is used at least three object servers of described at least one object server are made the data redundancy backup.
6. distributed file system according to claim 4 is characterized in that, also comprises:
The backup object central server is used for described object centers server is backed up.
7. according to each described distributed file system of claim 1-6, it is characterized in that, also comprise:
Backup metadata central server is used for described metadata central server is backed up.
8. the data access method of a distributed file system is characterized in that, comprising:
Receive the access request of client, and send described access request;
Receive overall unique meta data server identifier of the current available meta data server that the metadata central server returns according to described access request;
Overall unique meta data server identifier according to described meta data server produces fileinfo, and sends described fileinfo;
Receive overall unique object server identifier of the current available object server that described meta data server returns according to described fileinfo;
Overall unique object server identifier according to described object server sends accessed content;
Receive the visit result that described object server produces according to described accessed content, and described visit result is returned to described client.
9. method according to claim 8 is characterized in that, described overall unique meta data server identifier according to described meta data server produces fileinfo, and also comprises after sending described fileinfo:
At least three meta data servers are made the data redundancy backup.
10. according to Claim 8 or 9 described methods, it is characterized in that, also comprise before overall unique meta data server identifier of the current available meta data server that described reception metadata central server returns according to described access request:
The metadata central server receives the registration request of meta data server, produce overall unique meta data server identifier of described meta data server, and overall unique meta data server identifier of described meta data server is returned to described meta data server;
Meta data server receives the registration request of object server, produces overall unique object server identifier of described object server, and overall unique object server identifier of described object server is returned to described object server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100912919A CN101997823B (en) | 2009-08-17 | 2009-08-17 | Distributed file system and data access method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100912919A CN101997823B (en) | 2009-08-17 | 2009-08-17 | Distributed file system and data access method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101997823A CN101997823A (en) | 2011-03-30 |
| CN101997823B true CN101997823B (en) | 2013-10-02 |
Family
ID=43787423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100912919A Active CN101997823B (en) | 2009-08-17 | 2009-08-17 | Distributed file system and data access method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101997823B (en) |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102866995B (en) * | 2011-07-05 | 2017-05-24 | 中兴通讯股份有限公司 | File access method for PPN (Peer-to-Peer Network), management method and distributed file system |
| CN102360382B (en) * | 2011-10-13 | 2013-04-10 | 中国人民解放军国防科学技术大学 | High-speed object-based parallel storage system directory replication method |
| CN103095758B (en) * | 2011-11-01 | 2016-09-21 | 北大方正集团有限公司 | A kind of method processing file data in distributed file system and this system |
| US10176184B2 (en) * | 2012-01-17 | 2019-01-08 | Oracle International Corporation | System and method for supporting persistent store versioning and integrity in a distributed data grid |
| US9305004B2 (en) * | 2012-06-05 | 2016-04-05 | International Business Machines Corporation | Replica identification and collision avoidance in file system replication |
| CN103581229B (en) * | 2012-07-26 | 2018-06-15 | 腾讯科技（深圳）有限公司 | Distributed file system, file access method and client |
| CN103577500B (en) * | 2012-08-10 | 2018-07-27 | 腾讯科技（深圳）有限公司 | Distributed file system carries out the method and the distributed file system of data processing |
| CN103778120B (en) * | 2012-10-17 | 2018-01-16 | 腾讯科技（深圳）有限公司 | Global profile mark generating method, generating means and corresponding distributed file system |
| CN103064927B (en) * | 2012-12-21 | 2016-10-05 | 曙光信息产业(北京)有限公司 | The data access method of distributed file system and device |
| CN103888499B (en) * | 2012-12-21 | 2018-08-14 | 中国科学院深圳先进技术研究院 | A kind of method and system of distributed objects processing |
| CN103237046B (en) * | 2013-02-25 | 2016-08-17 | 中国科学院深圳先进技术研究院 | Support distributed file system and the implementation method of mixed cloud storage application |
| CN103167035A (en) * | 2013-03-15 | 2013-06-19 | 汉柏科技有限公司 | Method for solving NameNode single-point failure in hadoop distributed file system (HDFS) |
| CN104348793B (en) * | 2013-07-30 | 2018-04-06 | 阿里巴巴集团控股有限公司 | The storage method of storage server system and data message |
| CN104660643A (en) * | 2013-11-25 | 2015-05-27 | 南京中兴新软件有限责任公司 | Request response method and device and distributed file system |
| CN104796412B (en) * | 2014-04-06 | 2018-08-17 | 惠州Tcl移动通信有限公司 | End-to-end cloud service system and access method to its sensitive data |
| CN105446794A (en) * | 2014-09-30 | 2016-03-30 | 北京金山云网络技术有限公司 | Disc operation method, apparatus and system based on virtual machine |
| CN104320401B (en) * | 2014-10-31 | 2018-06-22 | 北京思特奇信息技术股份有限公司 | A kind of big data storage based on distributed file system accesses system and method |
| CN104536988B (en) * | 2014-12-10 | 2017-10-24 | 杭州斯凯网络科技有限公司 | MonetDB Distributed Calculation storage methods |
| CN106155921B (en) * | 2015-04-03 | 2019-06-25 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN105468295B (en) * | 2015-11-14 | 2019-03-05 | 华中科技大学 | A kind of security protection access method and system for realizing object storage |
| CN107026876B (en) * | 2016-01-29 | 2019-07-19 | 杭州海康威视数字技术股份有限公司 | A kind of file data access system and method |
| CN106775459B (en) * | 2016-11-28 | 2020-06-02 | 北京同有飞骥科技股份有限公司 | Safe distributed file system based on FT-1500A processor |
| CN107968786A (en) * | 2017-12-05 | 2018-04-27 | 北京奇艺世纪科技有限公司 | Distributed RPC systems and data processing method and processing device |
| CN109783462B (en) * | 2018-12-13 | 2021-01-05 | 创新科技术有限公司 | Data access method and device based on distributed file system |
| CN113190523B (en) * | 2021-04-08 | 2022-09-13 | 金钱猫科技股份有限公司 | Distributed file system, method and client based on multi-client cooperation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1622087A (en) * | 2003-04-29 | 2005-06-01 | 国际商业机器公司 | Managing file system versions |
| CN1670726A (en) * | 2004-03-17 | 2005-09-21 | 联想(北京)有限公司 | A method for inspecting garbage files in cluster file system |
| CN1771497A (en) * | 2003-04-07 | 2006-05-10 | 皇家飞利浦电子股份有限公司 | Content directory service import container |
| CN101013381A (en) * | 2007-01-26 | 2007-08-08 | 华中科技大学 | Distributed lock based on object memory system |
| CN101040282A (en) * | 2004-08-13 | 2007-09-19 | 潘纳萨斯公司 | Distributed object-based storage system that stores virtualization maps in object attributes |
| CN101316273A (en) * | 2008-05-12 | 2008-12-03 | 华中科技大学 | Distributed safety memory system |
| CN101449559A (en) * | 2006-05-23 | 2009-06-03 | 诺燕控股公司 | Distributed storage |
-
2009
- 2009-08-17 CN CN2009100912919A patent/CN101997823B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1771497A (en) * | 2003-04-07 | 2006-05-10 | 皇家飞利浦电子股份有限公司 | Content directory service import container |
| CN1622087A (en) * | 2003-04-29 | 2005-06-01 | 国际商业机器公司 | Managing file system versions |
| CN1670726A (en) * | 2004-03-17 | 2005-09-21 | 联想(北京)有限公司 | A method for inspecting garbage files in cluster file system |
| CN101040282A (en) * | 2004-08-13 | 2007-09-19 | 潘纳萨斯公司 | Distributed object-based storage system that stores virtualization maps in object attributes |
| CN101449559A (en) * | 2006-05-23 | 2009-06-03 | 诺燕控股公司 | Distributed storage |
| CN101013381A (en) * | 2007-01-26 | 2007-08-08 | 华中科技大学 | Distributed lock based on object memory system |
| CN101316273A (en) * | 2008-05-12 | 2008-12-03 | 华中科技大学 | Distributed safety memory system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101997823A (en) | 2011-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101997823B (en) | Distributed file system and data access method thereof | |
| CN112313916B (en) | Method and system for pseudo-storage of anti-tampering logs by fusing block chain technology | |
| Lakshman et al. | Cassandra: a decentralized structured storage system | |
| US20170075741A1 (en) | Prioritizing Data Reconstruction in Distributed Storage Systems | |
| CN103116661B (en) | A kind of data processing method of database | |
| US11567899B2 (en) | Managing dependent delete operations among data stores | |
| CN103942112A (en) | Magnetic disk fault-tolerance method, device and system | |
| EP4213038A1 (en) | Data processing method and apparatus based on distributed storage, device, and medium | |
| US11409711B2 (en) | Barriers for dependent operations among sharded data stores | |
| CN103166785A (en) | Distributed type log analysis system based on Hadoop | |
| CN102142032B (en) | Method and system for reading and writing data of distributed file system | |
| US20210165768A1 (en) | Replication Barriers for Dependent Data Transfers between Data Stores | |
| WO2021112909A1 (en) | Managing replication state for deleted objects | |
| CN104965835B (en) | A kind of file read/write method and device of distributed file system | |
| RU2643642C2 (en) | Use of cache memory and another type of memory in distributed memory system | |
| CN103150225B (en) | Disk full abnormity fault tolerance method of object parallel storage system based on application level agent | |
| CN110278222A (en) | The method, system and relevant device of data management in distributed file storage system | |
| CN104951475A (en) | Distributed file system and implementation method | |
| US11341159B2 (en) | In-stream data load in a replication environment | |
| CN114996053A (en) | Remote volume replication transmission method, system, device and storage medium | |
| Cheng et al. | A practical cross-datacenter fault-tolerance algorithm in the cloud storage system | |
| CN102497359A (en) | Thin client streaming media-based service system running method | |
| Li et al. | A hybrid disaster-tolerant model with DDF technology for MooseFS open-source distributed file system | |
| Lv et al. | RotaryDS: fast storage for massive data streams via a rotation storage model | |
| Pandey et al. | Replication in distributed systems and its improvements |
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 |