CN103139224B - The access method of a kind of NFS and NFS - Google Patents
The access method of a kind of NFS and NFS Download PDFInfo
- Publication number
- CN103139224B CN103139224B CN201110373307.2A CN201110373307A CN103139224B CN 103139224 B CN103139224 B CN 103139224B CN 201110373307 A CN201110373307 A CN 201110373307A CN 103139224 B CN103139224 B CN 103139224B
- Authority
- CN
- China
- Prior art keywords
- archive sit
- file
- pointer
- access destination
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of NFS, this system comprises: local file buffer memory, for preserving the access destination of accessing in user's nearest a period of time; Local file database, for preserving all access destination that user had accessed, and carries out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server; Far-end server, for preserving original user file system; Client, according to the operational order of user, when determining the position of access destination, first inquires about, if do not inquire, then inquires about in local file database in local file buffer memory, if inquire not yet, then and distally server lookup.The invention also discloses a kind of access method of NFS, this system and method can reduce the mutual of client and far-end server, improves stability and the access speed of file access.
Description
Technical field
The present invention relates to network communication technology field, particularly the access method of a kind of NFS and NFS.
Background technology
NFS, as cloud storage system, refers to that user file is stored in far-end server, and client, by access to netwoks far-end server, carries out the file system of file operation.
Current NFS generally takes simple primary structure, namely client directly and far-end server mutual, client can by Internet Transmission to far-end server to the operational order of file, responded by far-end server and perform, such as, when client needs file reading, the reading command of file will be sent directly to far-end server, distally after whois lookup to corresponding file destination, pull corresponding file data.
For existing NFS, all need directly to access far-end server to the operation of file at every turn, add the processing load of server, and can not ensure during Network Abnormal that file can be accessed smoothly, the poor stability of file access; Meanwhile, by the restriction of network transfer speeds, gratifying file access speed cannot be reached.
Summary of the invention
In view of this, the invention provides the access method of a kind of NFS and NFS, the mutual of client and far-end server can be reduced, and improve stability and the access speed of file access.
For reaching above-mentioned purpose, technical scheme of the present invention is specifically achieved in that
A kind of NFS, this system comprises: client, local file buffer memory, local data base and far-end server;
Described client, for the operational order according to user, determines the position of access destination in described operational order; When determining access destination position, first inquire about in described local file buffer memory, if do not inquire, then inquire about in described local file database, if inquire not yet, then inquire about to described far-end server;
Described local file buffer memory, for preserving the access destination of accessing in user's nearest a period of time;
Described local file database, for preserving all access destination that user had accessed, and carry out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server, make to be consistent in access destination and original user file system;
Far-end server, by network and described client's side link, for preserving original user file system.
An access method for NFS, the method comprises:
According to the operational order of user, when determining the position of access destination in described operational order;
First inquire about in the local file buffer memory preserving the access destination of accessing in user's nearest a period of time;
If do not inquire, all access destination of then preserving user and accessed, and carry out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server, make to inquire about in the local file database be consistent in the access destination of preservation and original user file system;
If inquire not yet, then to by network and described client's side link, preserve the far-end server inquiry of original user file system.
As seen from the above technical solution, user can control to complete in client this locality to the access of NFS by the access method of this NFS of the present invention and NFS to greatest extent, thus avoid the impact of network condition on NFS, improve stability and the access speed of file access.
Accompanying drawing explanation
Fig. 1 is the NFS structural representation of the embodiment of the present invention;
Fig. 2 is the linked list type file system index schematic diagram of embodiment of the present invention local file cache maintenance;
Fig. 3 is LRU chained list schematic diagram in the linked list type file system index of embodiment of the present invention local file cache maintenance;
Fig. 4 is in the linked list type file system index of embodiment of the present invention local file cache maintenance, set membership chained list schematic diagram.
Embodiment
For making object of the present invention, technical scheme and advantage clearly understand, to develop simultaneously embodiment referring to accompanying drawing, the present invention is described in more detail.
The present invention mainly adds the file cache of client this locality and the document data bank of client this locality between the client of NFS and far-end server; The document data bank of client this locality preserves all access destination that user had accessed; Client local file buffer memory preserves the access destination that in nearest a period of time, user accessed; When carrying out file access, first inquiring client terminal local file buffer memory, if miss, then inquiring client terminal local file database, if still miss, visits again far-end server; Meanwhile, keep the access destination of client file database preservation consistent with the original user file system that far-end server is preserved by synchronization policy.The present invention can make user occur in client this locality as much as possible to the access of network file, thus reduces the mutual of client and far-end server, reduces the dependence to network, improves stability and access speed that network file accesses.
Fig. 1 is the NFS structural representation of the embodiment of the present invention, and as shown in Figure 1, this system comprises: client 101, local file buffer memory 102, local file database 103 and far-end server 104;
Client 101, for the operational order according to user, as created, read, amendment etc., determine in described operational order, the position of access destination.
Wherein, access destination is the needs of specifying in the operational order file destination that carries out operating or catalogue.
For different operational orders, although concrete operations are different, but all need before executable operations instruction to inquire about access destination, determine the position of access destination, just can carry out follow-up operational order and perform and response, so, if access destination is local in client 101, instead of the server of network far-end, namely can carry out file operation in client 101 this locality, and not need to access far-end server 104.
Therefore, in the present embodiment, when client 101 determines access destination position, first can inquire about in local file buffer memory 102, if inquired, then determine that the position of access destination is in local file buffer memory 102; Otherwise, inquire about in local file database 103, if found, then determine that the position of access destination is in local file database 103, if do not find, then last more distally server 104 inquire about.
Local file buffer memory 102, for preserving the access destination of accessing in user's nearest a period of time;
Local file buffer memory 102 can be arranged in the internal memory of client 101 this locality, hard disk or other memory device (for improving access speed, within save as good); In addition, in local file buffer memory 102, need maintenance file system index, for client 101 queried access target particular location in the buffer.
Local file database 103, for preserving all access destination that user had accessed, and carry out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server 104, make to be consistent in access destination and original user file system, concrete synchronization policy is any, and this synchronizing process can adopt background mode to carry out.
Local file database 103 can be arranged in hard disk or other non-volatile memory apparatus of client 101 this locality, preserve access destination in a database format, the position that the search method queried access target that client 101 can be provided by database records in a database.
Far-end server 104, is connected with client 101 by network, for preserving original user file system, and carries out synchronous with the access destination of preserving in local file database 103, makes to be consistent in access destination and original user file system.
What preserve in far-end server 104 is original user file, namely final regular user data, all access destination that the user preserved in local file database 103 had accessed, being equivalent to a subset of original user file system, is the copy of the certain customers' file in original user file system; And the access destination of accessing in user's nearest a period of time of preserving in local file buffer memory 102, a subset of all access destination that the user being equivalent to again to preserve in local file database 103 had accessed also is the copy of the certain customers' file in original user file system;
In addition, the file system index safeguarded in local file buffer memory 102 can adopt chain sheet form to carry out managing (chain sheet form is generally used for the situation that local file buffer memory 102 is arranged in the local internal memory of client), also database form (as local file buffer memory 102 is arranged in the situation of client local hard drive) can be adopted, or other self-defining indexed mode; For chain sheet form, concrete chained list way to manage, the pointer field that comprises etc. arbitrarily, are lifted one below and are adopted chain sheet form to be the object lesson protecting file system index.
As shown in Figure 2, each access destination, i.e. a single file or catalogue, represent with an Archive sit in chained list, the position of Archive sit in internal memory is indicated by pointer.The index of whole file system is from root, in Fig. 2, pRoot is root node pointer, and the pointer of all child nodes under root node is kept in Hash table, pHashTable is the pointer of Hash table, and client 101 can find the position of Hash table in internal memory by pHashTable; Lelru is least recently used (lru) territory (explained later in detail) of Archive sit;
Wherein, Hash table comprises the pointer in all entrances (entry) territory, and each entry territory chain enters Hash (leHash) territory of respective file node; Archive sit enters corresponding entry territory in Hash table by its leHash territory chain, forms doubly linked list structure, the concrete position of each Archive sit in Hash table, by this node
father node pointerand should
the file name (filename or directory name) of nodedetermine, the cryptographic Hash such as making the pointer of this Archive sit be calculated by the father node pointer of this Archive sit and the title of this Archive sit represents.
Be subject to the restriction of memory size, all access destination that user accessed possibly cannot be preserved in local file buffer memory 102, therefore, the Archive sit safeguarded in local file buffer memory 102 can not infinitely increase, therefore need to adopt least recently used (LRU) algorithm to carry out eliminating of Archive sit, only preserve the access destination that in nearest a period of time, user accessed, to reduce taking of local file buffer memory 102 pairs of internal memories.
In this example, be realize with LRU chained list, as shown in Figure 3, each Archive sit comprises a LRU territory (representing with lelru), is entered by these mutual chains in LRU territory, forms LRU chained list.Whether certain Archive sit adds LRU chained list, can count decision by reference, and when certain Archive sit is accessed, reference count adds 1; At the end of access, counting subtracts 1; Can decision node whether finish using like this; For the node of finishing using (reference count is 0), LRU chained list can be added into, etc. to be recycled, when the memory headroom that local file buffer memory 102 takies exceedes preset value, removal process will be started, delete an Archive sit in LRU chained list, simultaneously by the data field content of the Archive sit of recovery, namely concrete access destination (file or catalogue) is saved in local file database 103.If Archive sit is cited again before startup is reclaimed, this Archive sit can remove from LRU chained list.
Particularly, as needed the Archive sit that increase by is new in local file buffer memory 102, now, if local file buffer memory 102 takes up room exceed preset value, 90% (configurable) of such as memory size, then remove an existing file node and return as new node from LRU chained list; Otherwise, from memory pool, distribute a new node, and be linked in local file buffer memory 102, specifically remove which Archive sit any, can random selecting, also can according to certain rule interestingness.
In addition, in order to energy fast finding Archive sit, inquire about according to the tree-shaped relation between Archive sit, the pointer field for reflecting set membership between Archive sit can be increased further in Archive sit, such as shown in Fig. 4, child node enters subdirectory (leSubDirs) chained list of father node by child (leChild) territory chain; Child node upwards can access the concrete data in the data field of its father node by his father (pParent) territory.
By above-mentioned file system index, the fast finding of access destination can be realized, such as accessed path be net dish I document readme.txt " the search procedure of file be:
First according to separator " " path decomposing is become multiple Archive sit, i.e. " net dish ", " my document ", " readme.txt "; From most superior node, the father node pointer according to this node character string and this node calculates cryptographic Hash, and this cryptographic Hash, to should the pointer of node in Hash table, utilizes this cryptographic Hash from Hash table, find the pointer of this node; Calculate cryptographic Hash by the character string of the next stage child node of this node pointer found and this node again, utilize this cryptographic Hash from Hash table, find the pointer of the next stage child node of this node; Circulation like this is until all nodes are all found, and the node finally found is the Archive sit that will inquire about, and the pointer of this Archive sit is the position of access destination in internal memory.
Particularly, first calculate its hash value according to root node pointer and " net dish " character string, the pointer of " net dish " node is found according to this hash value, the pointer of " net dish " node and " my document " character string is used to calculate hash value again, and find the pointer of " net dish my document " node, pointer according to " net dish my document " node calculates hash value with " readme.txt " character string, and find " net dish I document readme.txt " pointer of node, thus find corresponding readme.txt file according to this pointer.
In addition, for avoiding data collision, can also file lock be introduced in local file buffer memory 102, comprising read lock and writing lock; Before performing certain operational order to access destination, attempt locking to access destination, the type of lock is by operational order type decided; Such as reading command is read lock, and amendment and establishment instruction are for writing lock etc.; As locking request failure then shows that access destination is used, and point out user; Successful then access destination is locked, and before this operational order completes, cannot be accessed by other operational orders.
In addition, file system does not allow the existence of the file of the same name in same paths, subdirectory of the same name, therefore also need to introduce duplication of name testing mechanism, when relating to the operations such as newly-increased access destination, mobile access target or amendment access destination title in execution, will search according to new title; If found, indicate duplication of name; Current operation can not continue, and points out user.
In addition, for the creation operation of access destination, need to be performed at far-end server 104 by network, but due to network operation free on delay, so need to create a temporary file node to represent that this access destination is created at local file buffer memory 102; Other concurrent operation wants that the access destination creating same title will be interrupted by duplication of name testing mechanism like this; If the access destination on far-end server 104 creates successfully, then the temporary file node that local file buffer memory 102 creates can become effective Archive sit; Otherwise, cancel temporary file node and point out user to create failure.
For local file database 103, its access destination that any database file form preservation user can be adopted to access, concrete preserving type is not the emphasis that the present invention discusses, here repeat no more, only introduce the concrete grammar of queried access target in local file database 103, such as, need to search the All Files under " net dish my document "; Concrete grammar is as follows:
Because " net dish " is root, there is no father node, so first search father node to be recorded as sky, name is called the data-base recording RowId1 of " net dish ", then search father node and be recorded as RowId1, name is called the data-base recording RowId2 of " my document ", searches record entries all in RowId2 successively more afterwards.
Wherein, RowId is the recording mechanism of Archive sit in database file, the corresponding unique recording mechanism of each Archive sit; For the subfile under a catalogue, in the record of subfile node, save the recording mechanism of its father node, therefore can add that the name of subfile is searched according to the recording mechanism of father node, obtain the record of subfile or subdirectory.
From the above embodiments, user can control to complete in client this locality to the access of NFS by the access method of this NFS of the present invention and NFS to greatest extent, thus avoid the impact of network condition on NFS, improve stability and the access speed of file access.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (13)
1. a NFS, is characterized in that, this system comprises: client, local file buffer memory, local file database and far-end server;
Described client, for the operational order according to user, determines the position of access destination in described operational order; When determining access destination position, first inquire about in described local file buffer memory, if do not inquire, then inquire about in described local file database, if inquire not yet, then inquire about to described far-end server;
Described local file buffer memory, for preserving the access destination of accessing in user's nearest a period of time, described local file buffer memory safeguards the index of whole file system in chained list mode, described access destination represents with Archive sit in chained list; Wherein, the pointer of all child nodes under root node is kept in Hash table, and Hash table comprises the pointer in all entrance territories, and each entrance territory chain enters the Hash territory of respective file node; Archive sit enters corresponding entrance territory in Hash table by its Hash territory chain, forms doubly linked list structure; The cryptographic Hash that the pointer of each Archive sit is calculated by the father node pointer of this Archive sit and the title of this Archive sit represents; Described local file database, for preserving all access destination that user had accessed, and carry out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server, related content in access destination and original user file system is consistent, and described related content is content corresponding with described access destination in original user file system;
Far-end server, by network and described client's side link, for preserving original user file system.
2. NFS as claimed in claim 1, is characterized in that, safeguard a file system index in described local file buffer memory, for described client query access destination particular location in the buffer.
3. NFS as claimed in claim 1 or 2, it is characterized in that, described local file database preserves access destination in a database format, makes the position that described client is recorded in a database by the search method queried access target that database provides.
4. NFS as claimed in claim 1, it is characterized in that, described local file buffer memory is eliminated described Archive sit by least recently used lru algorithm.
5. NFS as claimed in claim 4, it is characterized in that, in the file system index of described local file cache maintenance, each Archive sit comprises LRU territory, is entered by these mutual chains in LRU territory, forms LRU chained list;
When Archive sit is accessed, reference count adds 1; At the end of access, counting subtracts 1; Be the Archive sit of 0 for reference count, LRU chained list can be added into, when the memory headroom that described local file buffer memory takies exceedes preset value, from LRU chained list, delete an Archive sit reclaim, the concrete data of the Archive sit of recovery are saved in local file database simultaneously.
6. NFS as claimed in claim 4, it is characterized in that, in the file system index of described local file cache maintenance, each Archive sit comprises the pointer field for reflecting set membership between Archive sit;
Its child nodes enters the subdirectory chained list of father node by its child territory chain; Child node upwards accesses the data field of its father node by father field.
7. NFS as claimed in claim 4, it is characterized in that, the path decomposing of access destination is become multiple Archive sit by described client; From most upper file node, the father node pointer according to this file node character string and this Archive sit calculates cryptographic Hash, utilizes this cryptographic Hash from Hash table, find the pointer of this Archive sit; Calculate cryptographic Hash by the character string of the next stage child node of this file node pointer found and this Archive sit again, utilize this cryptographic Hash from Hash table, find the pointer of the next stage child node of this Archive sit; Circulation like this is until all Archive sits are all found, and the pointer of the Archive sit finally found is the position of access destination in internal memory.
8. an access method for NFS, is characterized in that, the method comprises:
According to the operational order of user, when determining the position of access destination in described operational order;
First inquire about in the local file buffer memory preserving the access destination of accessing in user's nearest a period of time, by the file system index safeguarded in local file buffer memory, queried access target particular location in the buffer, described index is safeguarded in chained list mode, and described access destination represents with Archive sit in chained list; Wherein, the pointer of all child nodes under root node is kept in Hash table, and Hash table comprises the pointer in all entrance territories, and each entrance territory chain enters the Hash territory of respective file node; Archive sit enters corresponding entrance territory in Hash table by its Hash territory chain, forms doubly linked list structure; The cryptographic Hash that the pointer of each Archive sit is calculated by the father node pointer of this Archive sit and the title of this Archive sit represents;
If do not inquire, all access destination of then preserving user and accessed, and carry out synchronous by the synchronization policy preset with the original user file system of preserving in far-end server, inquire about in the local file database that related content in the access destination of preservation and original user file system is consistent, described related content is content corresponding with described access destination in original user file system;
If inquire not yet, then to by network and client's side link, preserve the far-end server inquiry of original user file system.
9. the access method of NFS as claimed in claim 8, is characterized in that, describedly inquires about in local file database, comprising:
By the search method that described local file database provides, inquire about the access destination of preserving in a database format, the position of recording in a database.
10. the access method of NFS as claimed in claim 8, it is characterized in that, described Archive sit is eliminated by least recently used lru algorithm.
The access method of 11. NFSs as claimed in claim 10, is characterized in that, described eliminating comprises:
Each Archive sit comprises a LRU territory, is entered by these mutual chains in LRU territory, forms LRU chained list;
When Archive sit is accessed, reference count adds 1; At the end of access, counting subtracts 1; Be the Archive sit of 0 for reference count, LRU chained list can be added into, when the memory headroom that described local file buffer memory takies exceedes preset value, from LRU chained list, delete an Archive sit reclaim, the concrete data of the Archive sit of recovery are saved in local file database simultaneously.
The access method of 12. NFSs as claimed in claim 8, it is characterized in that, described each Archive sit comprises the pointer field for reflecting set membership between Archive sit further;
Its child nodes enters the subdirectory chained list of father node by its child territory chain; Child node upwards accesses the data field of its father node by its father field.
The access method of 13. NFSs as claimed in claim 8, is characterized in that, by the file system index safeguarded in local file buffer memory, queried access target particular location in the buffer comprises:
The path decomposing of access destination is become multiple Archive sit; From most upper file node, the father node pointer according to this file node character string and this Archive sit calculates cryptographic Hash, utilizes this cryptographic Hash from Hash table, find the pointer of this Archive sit; Calculate cryptographic Hash by the character string of the next stage child node of this file node pointer found and this Archive sit again, utilize this cryptographic Hash from Hash table, find the pointer of the next stage child node of this Archive sit; Circulation like this is until all Archive sits are all found, and the pointer of the Archive sit finally found is the position of access destination in internal memory.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110373307.2A CN103139224B (en) | 2011-11-22 | 2011-11-22 | The access method of a kind of NFS and NFS |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110373307.2A CN103139224B (en) | 2011-11-22 | 2011-11-22 | The access method of a kind of NFS and NFS |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103139224A CN103139224A (en) | 2013-06-05 |
CN103139224B true CN103139224B (en) | 2016-01-27 |
Family
ID=48498529
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110373307.2A Active CN103139224B (en) | 2011-11-22 | 2011-11-22 | The access method of a kind of NFS and NFS |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103139224B (en) |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103546574B (en) * | 2013-10-30 | 2016-05-25 | 中国航空工业集团公司第六三一研究所 | A kind of airborne built-in network file access control method based on static configuration table |
CN103631914A (en) * | 2013-11-30 | 2014-03-12 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Offline browsing method and system of massive internet content |
CN104750715A (en) * | 2013-12-27 | 2015-07-01 | 中国移动通信集团公司 | Data elimination method, device and system in caching system and related server equipment |
CN103944958A (en) * | 2014-03-14 | 2014-07-23 | 中国科学院计算技术研究所 | Wide area file system and implementation method |
WO2015151082A1 (en) * | 2014-03-31 | 2015-10-08 | Varonis Systems, Ltd. | Delegating portions of an index of an enterprise |
CN104182487A (en) * | 2014-08-11 | 2014-12-03 | 浪潮软件股份有限公司 | Unified storage method supporting various storage modes |
CN104462194A (en) * | 2014-10-28 | 2015-03-25 | 北京国双科技有限公司 | Service data processing method, device and server |
CN104935636A (en) * | 2015-04-29 | 2015-09-23 | 广州杰赛科技股份有限公司 | Network channel acceleration method and system |
CN105718291B (en) * | 2016-01-25 | 2020-07-07 | 山东蓝创网络技术有限公司 | Multilevel cache acceleration method for mixed desktop application |
CN105975528B (en) * | 2016-04-29 | 2021-05-21 | 努比亚技术有限公司 | Synchronization method, device and system for background configuration file |
CN106469186A (en) * | 2016-08-29 | 2017-03-01 | 北京像素软件科技股份有限公司 | A kind of method and device of character string comparison |
CN106357769A (en) * | 2016-09-20 | 2017-01-25 | 百度在线网络技术(北京)有限公司 | Resource downloading method, device and network disk server |
CN106452893A (en) * | 2016-10-25 | 2017-02-22 | 广东西奥物联网科技股份有限公司 | Method for loading configuration in startup of intermediate equipment |
CN108021339B (en) * | 2017-11-03 | 2021-05-04 | 网宿科技股份有限公司 | Method and device for reading and writing magnetic disk and computer readable storage medium |
CN108984128B (en) * | 2018-07-19 | 2022-03-08 | 郑州云海信息技术有限公司 | Data reading method and device |
CN111001157B (en) * | 2019-11-29 | 2021-09-28 | 腾讯科技(深圳)有限公司 | Method and device for generating reference information, storage medium and electronic device |
CN113656366B (en) * | 2021-08-03 | 2024-05-03 | 北京数码大方科技股份有限公司 | Data processing method and device for electronic warehouse |
CN116541365B (en) * | 2023-07-06 | 2023-09-15 | 成都泛联智存科技有限公司 | File storage method, device, storage medium and client |
CN117708179B (en) * | 2024-02-02 | 2024-05-03 | 成都深瑞同华科技有限公司 | Method, device, equipment and medium for caching measurement point data of electric power comprehensive monitoring system |
CN117914867B (en) * | 2024-03-19 | 2024-06-18 | 苏州元脑智能科技有限公司 | Data buffering method, device, equipment and computer readable storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1370103A1 (en) * | 2002-06-07 | 2003-12-10 | Evolium S.A.S. | Connecting a terminal over a mobile radio access network or a local access network to the core network of a radio communications system |
CN1997015A (en) * | 2006-11-24 | 2007-07-11 | 华为技术有限公司 | Cache application method and device, and file transfer system |
CN101122885A (en) * | 2007-09-11 | 2008-02-13 | 腾讯科技(深圳)有限公司 | Data cache processing method, system and data cache device |
CN101539918A (en) * | 2008-03-19 | 2009-09-23 | 天下互联(北京)科技有限公司 | Method and system for internet search |
CN102014158A (en) * | 2010-11-29 | 2011-04-13 | 北京兴宇中科科技开发股份有限公司 | Cloud storage service client high-efficiency fine-granularity data caching system and method |
-
2011
- 2011-11-22 CN CN201110373307.2A patent/CN103139224B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1370103A1 (en) * | 2002-06-07 | 2003-12-10 | Evolium S.A.S. | Connecting a terminal over a mobile radio access network or a local access network to the core network of a radio communications system |
CN1997015A (en) * | 2006-11-24 | 2007-07-11 | 华为技术有限公司 | Cache application method and device, and file transfer system |
CN101122885A (en) * | 2007-09-11 | 2008-02-13 | 腾讯科技(深圳)有限公司 | Data cache processing method, system and data cache device |
CN101539918A (en) * | 2008-03-19 | 2009-09-23 | 天下互联(北京)科技有限公司 | Method and system for internet search |
CN102014158A (en) * | 2010-11-29 | 2011-04-13 | 北京兴宇中科科技开发股份有限公司 | Cloud storage service client high-efficiency fine-granularity data caching system and method |
Also Published As
Publication number | Publication date |
---|---|
CN103139224A (en) | 2013-06-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103139224B (en) | The access method of a kind of NFS and NFS | |
CN106874383B (en) | Decoupling distribution method of metadata of distributed file system | |
CN103150394B (en) | Distributed file system metadata management method facing to high-performance calculation | |
US7454405B2 (en) | File management program, file management process, and file management apparatus | |
CN101697168B (en) | Method and system for dynamically managing metadata of distributed file system | |
CN102473127B (en) | An integrated approach for deduplicating data in a distributed environment that involves a source and a target | |
US20150347553A1 (en) | Object Storage System with Local Transaction Logs, a Distributed Namespace, and Optimized Support for User Directories | |
CN109299113B (en) | Range query method with storage-aware mixed index | |
US9547706B2 (en) | Using colocation hints to facilitate accessing a distributed data storage system | |
US20070094269A1 (en) | Systems and methods for distributed system scanning | |
CN103338242B (en) | A kind of mixed cloud storage system based on multi-level buffer and method | |
CN110647497A (en) | HDFS-based high-performance file storage and management system | |
CN108021717B (en) | Method for implementing lightweight embedded file system | |
CN104516974B (en) | A kind of management method and device of file system directories item | |
CN104408111A (en) | Method and device for deleting duplicate data | |
CN101072178A (en) | Switching system routing table management method | |
CN106294772A (en) | The buffer memory management method of distributed memory columnar database | |
CN109726177A (en) | A kind of mass file subregion indexing means based on HBase | |
CN106354805A (en) | Optimization method and system for searching and caching distribution storage system NoSQL | |
CN103077199A (en) | File resource searching and locating method and device | |
CN110888837B (en) | Object storage small file merging method and device | |
CN102866995A (en) | File access method for PPN (Peer-to-Peer Network), management method and distributed file system | |
CN109522283A (en) | A kind of data de-duplication method and system | |
CN112148680B (en) | File system metadata management method based on distributed graph database | |
CN105159845A (en) | Memory reading method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230920 Address after: 518057 Tencent Building, No. 1 High-tech Zone, Nanshan District, Shenzhen City, Guangdong Province, 35 floors Patentee after: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. Patentee after: TENCENT CLOUD COMPUTING (BEIJING) Co.,Ltd. Address before: 2, 518044, East 403 room, SEG science and Technology Park, Zhenxing Road, Shenzhen, Guangdong, Futian District Patentee before: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. |