CN113111034B - Index pre-allocation method and device - Google Patents
Index pre-allocation method and device Download PDFInfo
- Publication number
- CN113111034B CN113111034B CN202110374789.7A CN202110374789A CN113111034B CN 113111034 B CN113111034 B CN 113111034B CN 202110374789 A CN202110374789 A CN 202110374789A CN 113111034 B CN113111034 B CN 113111034B
- Authority
- CN
- China
- Prior art keywords
- mds
- client
- index
- mapping table
- directory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/13—File access structures, e.g. distributed indices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/17—Details of further file system functions
- G06F16/172—Caching, prefetching or hoarding of files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a method and a device for pre-distributing indexes, wherein MDS responds to a request of a client to distribute an initial number of index nodes for the client, the client creates files to use the index nodes, the MDS monitors the condition that the client uses the index nodes and records the number of the current distributed index nodes through a first mapping table, and if the number of the current index nodes is lower than a distribution threshold value; the MDS allocates an inode to the client. The MDS records index node numbers of unoccupied index nodes in the allocated index nodes by using a second mapping table. The MDS reclaims inodes assigned to the clients that are disconnected or have no file creating active directory. The invention ensures the smooth progress of the client file creation through the continuous pre-allocation index nodes, and accelerates the processing speed of the client file. And moreover, index node resources are prevented from being allocated to unused clients or directories through an index node recycling mechanism, so that resources are saved.
Description
Technical Field
The invention relates to the technical field of distributed storage indexes, in particular to a method and a device for pre-distributing indexes.
Background
In the current distributed file system, when a client processes a file, the client needs to send a message to an MDS (Meta-data Server) to apply for an index node, and then wait for the MDS to reply to a certain index node of the client, so that the client can perform the next operation.
In application scenarios such as video monitoring of a distributed file system, a directory is generally created according to a date, and file creation is continuously performed in a large amount under one directory within a certain time. A typical relationship between a client and a directory is a one-to-many relationship, i.e., one client corresponds to multiple directories, while one directory does not involve multiple clients. In the application scenario of video monitoring, the amount of the related file data is often relatively large, the final amount of the related file data is unpredictable for the MDS, the index node allocated by the client side request MDS can be used up soon, and when the client side needs to process the file again, the client side still needs to repeatedly apply for the index node again, so that the problem of slow file processing is caused.
Disclosure of Invention
The present invention provides a method for pre-allocating indexes to solve the above problems, comprising:
constructing a first mapping table in the MDS, wherein the first mapping table records the mapping relation between the directory of the client and the allocation amount of the index nodes;
the MDS responds to the index node application of a client, an initial number of index nodes are configured for the client, the client takes one index node from the index nodes to fill directory information, the client sends the directory information to the MDS, the MDS feeds back the directory information to an upper computer, and the MDS updates the first mapping table;
the MDS counts the occupied amount of the index nodes in the file creation process under the creation catalog, and updates the first mapping table according to the occupied amount of the index nodes, wherein the current distribution amount of the index nodes in the first mapping table is equal to the distribution amount of the index nodes of the original record minus the occupied amount of the index nodes;
the MDS judges whether the current distribution amount of the index nodes is lower than a distribution amount threshold value or not;
if the current inode allocation is below the allocation threshold, the MDS allocates an inode to the directory of the client.
Further, when the upper computer stores the file in the created directory, the client selects an index node to fill the file information, the client sends a request for creating the file to the MDS, and the MDS counts the index node occupation amount according to the received request for creating the file.
Further, the MDS configures a free index space, the free index space records an unallocated and occupied inode number, the MDS configures a second mapping table, and the second mapping table records a mapping relationship between the directory and the allocated inode number; and when the MDS acquires the inode number from the idle index space and distributes the inode number to the catalog, the MDS updates the second mapping table according to the distributed inode number, and when the distributed inode number is occupied, the MDS deletes the occupied inode number from the second mapping table.
Further, the MDS monitors the creation activity of the files in the directory of the client, and if the MDS does not receive the request for creating the files under the directory within a set time, the MDS acquires the allocated and unoccupied inode numbers under the target directory, and recovers the inode numbers.
Further, the MDS obtains the allocated and unoccupied inode number in the target directory from the second mapping table, the MDS sends recycling information to the client, the recycling information includes the unoccupied inode number, and the client deletes the inode in the cache according to the recycling information.
Further, the MDS detects whether the client is disconnected from the MDS, if the client is disconnected from the MDS, the MDS obtains the directory created under the target client, and the MDS recovers all the allocated inode numbers under the target client directory according to the second mapping table.
Further, the MDS configures a third mapping table, wherein the third mapping table stores the mapping relation between the client and the directory created under the client, and the MDS obtains the directory created under the target client according to the third mapping table.
Further, when the client is connected to the MDS, the client sends the inode number in its own cache to the MDS, the MDS determines whether the inode number has been retracted, and if the inode number has been retracted, the client deletes the inode number in the cache.
Further, the MDS stores the acquired unoccupied inode number into the free index space.
The invention provides an index pre-allocation device, which comprises a cluster, wherein at least one instruction is configured in the cluster, and the instruction is executed to realize the index pre-allocation method.
The index pre-allocation method and device provided by the application have the following beneficial effects:
according to the index pre-allocation method, the MDS allocates an initial number of index nodes for the client, the client utilizes the initial index nodes to carry out file creation, the MDS monitors the occupation condition of the initial index nodes in the file creation process of the client, counts whether the current index node allocation amount of the client is lower than an allocation amount threshold, allocates index nodes for the client if the current index node allocation amount is lower than the allocation amount threshold, and ensures that the index nodes are always available for the client in the file creation process, so that the problem of slow file creation process caused by the fact that the index nodes are used up and re-apply index nodes to the MDS in the file creation process of the client is avoided.
In addition, the MDS of the invention can retrieve the allocated index node by monitoring the file creation activity under the directory of the client, if the directory has no file creation activity within the set time, thereby avoiding the waste of resources. The MDS monitors the connection condition of the client, and if the client is disconnected, the MDS recovers the index node distributed to the target client, so that the waste of resources is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of pre-allocation in a method of index pre-allocation in an embodiment of the present invention;
figure 2 is a flow chart of increasing node allocation for a directory by the MDS of figure 1 based on a current node allocation amount in a first mapping table;
FIG. 3 is a flow chart of pre-allocation inode reclamation in a method of index pre-allocation in accordance with an embodiment of the present invention;
fig. 4 is a flowchart of a third mapping table updating client in an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The present invention will be described with reference to the accompanying drawings, in which fig. 1 is a flowchart of pre-allocation in a method for pre-allocation of indexes in an embodiment of the present invention; figure 2 is a flow chart of increasing node allocation for a directory by the MDS of figure 1 based on a current node allocation amount in a first mapping table; FIG. 3 is a flow chart of pre-allocation inode reclamation in a method of index pre-allocation in accordance with an embodiment of the present invention; fig. 4 is a flowchart of a third mapping table updating client in an embodiment of the present invention.
The invention provides a method for pre-distributing indexes, which comprises the steps of constructing a first mapping table in MDS, wherein the first mapping table records the mapping relation between a directory of a client and the distribution amount of index nodes; configuring a second mapping table in the MDS, wherein the second mapping table records the mapping relation between the directory and the currently allocated inode number; and configuring a third mapping table in the MDS, wherein the third mapping table records the mapping relation between a client side connected with the MDS and a directory in the client side. The method comprises the following steps:
referring to fig. 1, the method includes a pre-allocation process of index nodes:
s100, the client acquires a catalog creation application of an upper computer, and the client sends an index node application to the MDS; s200, the MDS responds to the application of index nodes of a client, and configures an initial number of index nodes for the client; and S300, the client creates a catalog by using index nodes and feeds back the catalog to the MDS, specifically, the client takes one index node from the index nodes to fill catalog information, the client sends the catalog information to the MDS, the client stores the index node numbers of the index nodes which are distributed in the rest in a cache, the index node numbers in the cache are associated with the catalog, and the MDS feeds back the catalog information to an upper computer, so that the creation of the catalog is realized. After the directory is created, S400, the MDS updates the first mapping table and the second mapping table according to the directory, the allocated inode amount, and the inode number; specifically, updating the first mapping table includes adding the created directory to the first mapping table, and adding the number of index nodes of which the initial number is reduced by one to the first mapping table to correspond to the corresponding directory; updating the second mapping table comprises adding a directory to the second mapping table, and adding an index node occupied by an initially allocated non-directory to the second mapping table to correspond to the directory; updating the third mapping table includes adding a directory to the third mapping table under the corresponding client.
S500, the MDS counts the occupied amount of the index nodes in the file creation process under the created catalogue. Specifically, the client obtains a file creation application of the upper computer under the target directory, the client responds to the file creation application to select an index node from the index nodes of the corresponding directory to fill file information, the client sends a file creation request to the MDS, and the MDS counts the occupation amount of the index nodes according to the received file creation request. After the MDS obtains the index node occupation amount, the first mapping table is updated according to the index node occupation amount, specifically, the current index node distribution amount in the first mapping table is equal to the original recorded index node distribution amount minus the index node occupation amount. Updating the second mapping table according to the occupied inode number, specifically, when the allocated inode number is occupied, deleting the occupied inode number from the second mapping table by the MDS.
And S600, the MDS increases the allocation of the index nodes for the catalogue according to the current allocation amount of the index nodes in the first mapping table. Specifically, referring to fig. 2, the MDS obtains a current allocation amount of the index node in the first mapping table, and the MDS determines whether the current allocation amount of the index node is lower than an allocation amount threshold; a viable threshold for the amount of allocation is fifty percent of the initial amount.
If the current inode allocation is below the allocation threshold, the MDS allocates an inode to the directory of the client. One possible allocation method is to allocate index nodes to the directory of the client, so that the number of preallocated index nodes in the directory of the client is equal to the initial number.
In a specific implementation process, the MDS configures a free index space, the free index space records node numbers which are not allocated and occupied, the MDS calculates the number of the nodes to be allocated, and when the MDS takes the target number of node numbers from the free index space, the MDS sends the node numbers to the client, and the client stores the node numbers in the cache corresponding to the directory. Thereby implementing the MDS pre-allocation of inodes for the clients.
On the other hand, when index number allocation is performed, the MDS updates the first mapping table according to the number of allocated index nodes, and specifically, the current index node allocation amount in the first mapping table is equal to the index node allocation amount recorded by a person plus the newly allocated index node number; the MDS updates the second mapping table according to the allocated inode number, and specifically adds the allocated inode number to the second mapping table.
And the MDS increases the distribution process of the index nodes for the catalogue according to the distribution amount of the current index nodes in the first mapping table until the MDS judges that the index nodes which need to be recovered are ended.
Referring to fig. 3, the method further includes a recovery process of the index node:
during the use process, the MDS monitors the creation activity of the files in the directory of the client, and if the MDS does not receive the request for creating the files under the directory within the set time, the MDS acquires the allocated and unoccupied inode numbers under the target directory, and the inode numbers are recovered. In the implementation process, the MDS acquires the name of the directory, the MDS retrieves and acquires the index node number which is allocated and not occupied under the target directory from the second mapping table according to the directory name, the MDS sends recovery information to the client, the recovery information comprises the index node number which is acquired from the second mapping table and not occupied, and the client deletes the index node in the cache according to the index node number in the recovery information.
And after the MDS recovers the allocated inode numbers, the inode number in the first mapping table is zero, and the inode number in the second mapping table is null. And when the client receives the request of creating the upper computer under the catalog again, the client searches the cache according to the catalog, the index node pre-allocated in the cache is empty, and the client sends the index node application to the MDS.
In addition, the MDS detects whether the client is disconnected from the MDS, and in a specific implementation process, the MDS may send a ping to the client at regular time, and if the client cannot be ping-enabled, the client is disconnected. And if the client is disconnected from the MDS, the MDS acquires the directory created under the target client from the third mapping table, the MDS retrieves the corresponding index node in the second mapping table according to the directory, and the MDS recovers all the index node numbers allocated under the target client directory.
When the client is reconnected to the MDS, the client sends the inode number in the cache to the MDS, the MDS judges whether the inode number is recovered, and if the inode number is recovered, the client deletes the inode number in the cache. And when the client receives the request of creating the upper computer under the catalog again, the client searches the cache according to the catalog, the index node pre-allocated in the cache is empty, and the client sends the index node application to the MDS.
In the implementation process, the process of recovering the index node by the MDS is to store the acquired unoccupied index node number into the idle index space.
When a certain client is connected with the MDS, the MDS acquires the identity information of the client, the MDS acquires whether the identity information retrieval of the client exists in the third mapping table or not, if the identity information of the client does not exist in the third mapping table, the third client is added into the third mapping table, and when any client has catalog creation, the MDS acquires the catalog and adds the catalog into the third mapping table.
The invention also provides an index pre-allocation device, which comprises a cluster, wherein the cluster is configured with MDS and a client, at least one instruction is configured in the cluster, and the instruction is executed to realize the index pre-allocation method.
According to the index pre-allocation method, the MDS allocates an initial number of index nodes for the client, the client utilizes the initial index nodes to carry out file creation, the MDS monitors the occupation condition of the initial index nodes in the file creation process of the client, counts whether the current index node allocation amount of the client is lower than an allocation amount threshold, allocates index nodes for the client if the current index node allocation amount is lower than the allocation amount threshold, and ensures that the index nodes are always available for the client in the file creation process, so that the problem of slow file creation process caused by the fact that the index nodes are used up and re-apply index nodes to the MDS in the file creation process of the client is avoided.
In addition, the MDS of the invention can retrieve the allocated index node by monitoring the file creation activity under the directory of the client, if the directory has no file creation activity within the set time, thereby avoiding the waste of resources. The MDS monitors the connection condition of the client, and if the client is disconnected, the MDS recovers the index node distributed to the target client, so that the waste of resources is avoided.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (7)
1. A method of index pre-allocation, comprising:
constructing a first mapping table, a second mapping table and a third mapping table in the MDS; the first mapping table records the mapping relation between the directory of the client and the allocation amount of the index nodes, the second mapping table records the mapping relation between the directory and the index node numbers allocated currently, and the third mapping table records the mapping relation between the client connected with the MDS and the directory in the client;
the MDS responds to the application of index nodes of a client, an initial number of index nodes are configured for the client, the client takes one index node from the index nodes to fill directory information, the client sends the directory information to the MDS, the MDS feeds the directory information back to an upper computer to realize the creation of the directory, the client stores the index node numbers of the index nodes which are distributed in the rest in a cache, and the index node numbers in the cache are associated with the directory; after the directory is created, the MDS updates the first, second, and third mapping tables according to the directory, the allocated inode amount, and the inode number:
adding the created catalogue to the first mapping table, adding the index node number of which the initial number is reduced by one to the first mapping table corresponding to the corresponding catalogue,
adding a directory to the second mapping table, adding an index node occupied by an initially allocated non-directory to the second mapping table corresponding to the directory,
adding a catalog to the corresponding client side in the third mapping table;
the MDS counts the occupied quantity of the index nodes and the occupied index node numbers in the file creation process under the created catalogue, and updates the first mapping table and the second mapping table; the MDS judges whether the current distribution amount of the index nodes is lower than a distribution amount threshold value according to the first mapping table; if the current allocation amount of the index nodes is lower than the allocation amount threshold value, the MDS allocates index nodes to the catalogue of the client;
the MDS monitors the creation activity of the files in the directory of the client, if the MDS does not receive a request for creating the files under the target directory within a set time, the MDS acquires the index node numbers which are allocated and not occupied under the target directory from the second mapping table, and the index node numbers are recovered; and the MDS detects whether the client is disconnected with the MDS, if the client is disconnected with the MDS, the MDS acquires the directory created under the target client from the third mapping table, and the MDS recovers all the allocated inode numbers under the target client directory according to the second mapping table.
2. The method of index pre-allocation according to claim 1, wherein when the upper computer stores the file under the created directory, the client selects an inode to fill the file information, the client sends a request for creating the file to the MDS, and the MDS counts the inode occupation amount and the occupied inode number according to the received request for creating the file.
3. The method of index pre-allocation according to claim 1, wherein the MDS configures free index space that records inode numbers that are unassigned and occupied; and when the MDS acquires the inode number from the idle index space and distributes the inode number to the catalog, the MDS updates the second mapping table according to the distributed inode number, and when the distributed inode number is occupied, the MDS deletes the occupied inode number from the second mapping table.
4. The method of index pre-allocation of claim 1, wherein the MDS sends reclamation information to the client, the reclamation information content including unoccupied inode numbers, the client deleting inodes in a cache based on the reclamation information.
5. The method of index pre-allocation according to claim 1, wherein when the client connects to the MDS, the client sends the node number in its own cache to the MDS, the MDS determines whether the node number has been reclaimed, and if the node number has been reclaimed, the client deletes the node number in the cache.
6. The method of index pre-allocation of claim 1, wherein the MDS stores the retrieved unoccupied inode number into free index space.
7. An apparatus for index pre-allocation, comprising a cluster in which at least one instruction is configured, and executing the instruction implements the method for index pre-allocation according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110374789.7A CN113111034B (en) | 2021-04-07 | 2021-04-07 | Index pre-allocation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110374789.7A CN113111034B (en) | 2021-04-07 | 2021-04-07 | Index pre-allocation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113111034A CN113111034A (en) | 2021-07-13 |
| CN113111034B true CN113111034B (en) | 2023-08-04 |
Family
ID=76715371
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110374789.7A Active CN113111034B (en) | 2021-04-07 | 2021-04-07 | Index pre-allocation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113111034B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547818A (en) * | 2012-01-04 | 2012-07-04 | 大唐移动通信设备有限公司 | Resource identifier allocating method and equipment |
| CN103268291A (en) * | 2013-05-23 | 2013-08-28 | 清华大学 | Method for delaying persistent indexing metadata in flash memory storage system |
| CN103761256A (en) * | 2013-12-30 | 2014-04-30 | 中国科学院计算技术研究所 | Maintenance system and method of branch points of remote directories |
| CN106326170A (en) * | 2016-08-19 | 2017-01-11 | 东软集团股份有限公司 | Equipment index number distribution method, device and system |
| CN106960020A (en) * | 2017-03-09 | 2017-07-18 | 星环信息科技（上海）有限公司 | A kind of method and apparatus for creating concordance list |
| CN109344122A (en) * | 2018-10-15 | 2019-02-15 | 中山大学 | A kind of distributed meta-data management method and system for creating strategy in advance based on file |
| CN110278288A (en) * | 2019-05-17 | 2019-09-24 | 平安科技（深圳）有限公司 | Method, apparatus, computer equipment and the storage medium of pre-dispensed vessels IP |
| CN111858474A (en) * | 2020-06-29 | 2020-10-30 | 苏州浪潮智能科技有限公司 | Distributed storage system Inode number allocation management method and related components |
| CN112612751A (en) * | 2020-12-25 | 2021-04-06 | 北京浪潮数据技术有限公司 | Asynchronous directory operation method, device, equipment and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10162828B2 (en) * | 2012-11-16 | 2018-12-25 | Red Hat, Inc. | Striping files across nodes of a distributed file system |
-
2021
- 2021-04-07 CN CN202110374789.7A patent/CN113111034B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547818A (en) * | 2012-01-04 | 2012-07-04 | 大唐移动通信设备有限公司 | Resource identifier allocating method and equipment |
| CN103268291A (en) * | 2013-05-23 | 2013-08-28 | 清华大学 | Method for delaying persistent indexing metadata in flash memory storage system |
| CN103761256A (en) * | 2013-12-30 | 2014-04-30 | 中国科学院计算技术研究所 | Maintenance system and method of branch points of remote directories |
| CN106326170A (en) * | 2016-08-19 | 2017-01-11 | 东软集团股份有限公司 | Equipment index number distribution method, device and system |
| CN106960020A (en) * | 2017-03-09 | 2017-07-18 | 星环信息科技（上海）有限公司 | A kind of method and apparatus for creating concordance list |
| CN109344122A (en) * | 2018-10-15 | 2019-02-15 | 中山大学 | A kind of distributed meta-data management method and system for creating strategy in advance based on file |
| CN110278288A (en) * | 2019-05-17 | 2019-09-24 | 平安科技（深圳）有限公司 | Method, apparatus, computer equipment and the storage medium of pre-dispensed vessels IP |
| CN111858474A (en) * | 2020-06-29 | 2020-10-30 | 苏州浪潮智能科技有限公司 | Distributed storage system Inode number allocation management method and related components |
| CN112612751A (en) * | 2020-12-25 | 2021-04-06 | 北京浪潮数据技术有限公司 | Asynchronous directory operation method, device, equipment and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113111034A (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7530115B2 (en) | Access to content addressable data over a network | |
| EP0312865B1 (en) | Space management system for a data access system of a file access processor | |
| CN101334797B (en) | Distributed file systems and its data block consistency managing method | |
| US5237682A (en) | File management system for a computer | |
| CN103916467B (en) | Load transfer method and system in metadata cluster | |
| CN106294352B (en) | A kind of document handling method, device and file system | |
| CN109739815A (en) | Document handling method, system, device, equipment and storage medium | |
| JP2001522074A (en) | Communication system and method | |
| CN109766318B (en) | File reading method and device | |
| CN106446044A (en) | Storage space reclaiming method and device | |
| CN101354720B (en) | Distributed memory database data system and sharing method thereof | |
| US9104689B2 (en) | Method for synchronizing documents for disconnected operation | |
| US10929354B2 (en) | Metadata server and method for distributing metadata in directories using the same | |
| US8707318B2 (en) | Partitioning system including a generic partitioning manager for partitioning resources | |
| CN112612751A (en) | Asynchronous directory operation method, device, equipment and system | |
| WO2020215580A1 (en) | Distributed global data deduplication method and device | |
| CN114610680A (en) | Method, device and equipment for managing metadata of distributed file system and storage medium | |
| CN113111034B (en) | Index pre-allocation method and device | |
| US8521789B2 (en) | Undrop objects and dependent objects in a database system | |
| CN107181773A (en) | Data storage and data managing method, the equipment of distributed memory system | |
| CN101483668A (en) | Network storage and access method, device and system for hot spot data | |
| CN100357943C (en) | A method for inspecting garbage files in cluster file system | |
| CN116150160A (en) | Adjustment method and device for database cluster processing nodes and storage medium | |
| CN109684331A (en) | A kind of object storage meta data management device and method based on Kudu | |
| CN115408342A (en) | File processing method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |