CN105094695A - Storing method and system - Google Patents
Storing method and system Download PDFInfo
- Publication number
- CN105094695A CN105094695A CN201510370132.8A CN201510370132A CN105094695A CN 105094695 A CN105094695 A CN 105094695A CN 201510370132 A CN201510370132 A CN 201510370132A CN 105094695 A CN105094695 A CN 105094695A
- Authority
- CN
- China
- Prior art keywords
- file
- operated
- dbstore
- storage class
- inode
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000013507 mapping Methods 0.000 claims description 36
- 238000012163 sequencing technique Methods 0.000 claims description 12
- 238000012217 deletion Methods 0.000 claims description 9
- 230000037430 deletion Effects 0.000 claims description 9
- 230000000875 corresponding effect Effects 0.000 description 47
- 230000000694 effects Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a storing method and system. The method comprises acquiring a preset storage type of a file from a preset inode list before an application layer operates the file. and correspondingly processing the file according to the storage type corresponding to the file. By adopting the storing method and system, problems of frequent addressing a disk and low performance caused by a lot of write-in random small files can be solved, and the write-in efficiency can be raised.
Description
Technical field
The present invention relates to field of distributed storage, be specifically related to a kind of storage means and system.
Background technology
Along with developing rapidly of network application, network information data amount is increasing, and the mass data storage of PB rank becomes more and more important.The local disk storage mode of traditional sense can not meet the Large Copacity of existing application, high reliability, high-performance, the requirement such as extending transversely, and for meeting these new demands, network distributed memory system obtains extensive attention.
Society network, mobile communication, network video and audio, ecommerce, sensor network, the data that the various application such as scientific experiment produces, often from a few KB to several MB not etc., the size dimension of these data is uneven, tackle common large files, traditional file system can obtain excellent performance by height is concurrent, but efficiency can be very low when processing mass small documents write, this is because mass small documents often brings a large amount of random writings, greatly performance can be reduced because of the frequent addressing of disk when being written locally, owing to there is inborn network delay in the distributed storage become more and more popular at present, many copies are arranged and metadata write can make performance be even lower.And if merely consider small documents performance, can lose again the advantage of large files process, there is certain paradox in both.
Summary of the invention
In order to solve the problem, the present invention proposes a kind of storage means and system, the frequent addressing of disk a large amount of random small documents can being avoided to write bring, the problem of degraded performance, improve write efficiency.
In order to achieve the above object, the present invention proposes a kind of storage means, the method comprises:
Before application layer operates file, obtained the predetermined storage class of this file by the index node inode list preset.
The storage class corresponding according to this file processes accordingly to this file.
Preferably,
Storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
Preferably,
The method also comprises: pre-set inode list in the following manner:
By different file to be operated and different inode value one_to_one corresponding.
By the storage class one_to_one corresponding of inode value and corresponding file.
Corresponding relation between the storage class of this file, inode value and file is kept in inode list, and inode list is kept in DBStore.
Preferably,
The method also comprises: the action type pre-determining different file to be operated in the following manner:
The size of this file and the threshold value preset are compared.
When the size of file is less than default threshold value, this file is judged to be small documents, and this small documents is defined as DBStore storage class.
When the size of file is more than or equal to default threshold value, this file is judged to be large files, and this large files is defined as FileStore storage class.
Preferably,
Carry out corresponding process according to the storage class that file is corresponding to file to comprise:
In creation operation, determine that the first file to be operated is after DBStore storage class, file to be operated is directly stored in DBStore; Wherein, the first file to be operated is metadata, and metadata is small documents.
In deletion action, determine that the second file to be operated is after DBStore storage class, directly obtain the positional information of the second file in DBStore to be operated, and the second file to be operated is deleted from DBStore; Wherein, the second file to be operated is metadata, and metadata is small documents.
In read request operation, after determining that the 3rd file to be operated is DBStore storage class, determine the 3rd position of file in DBStore to be operated according to inode value, and from DBStore, read the 3rd file to be operated; After determining that the 3rd file to be operated is FileStore storage class, determine the 3rd position of file in FileStore to be operated according to inode value, and from FileStore, read the 3rd file to be operated.
In write request operation, after determining that the 4th file to be operated is DBStore storage class, multiple the 4th file existed with small documents form is carried out record according to the sequencing of write, and by the 4th file and inode value one_to_one corresponding, form mapping pair, 4th file is stored in buffer zone with the form of mapping pair, and mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, when the mapping pair mapping log recording reaches default amount threshold, continue to perform write operation according to the action type recorded in Operation Log, according to the sequencing of write, the 4th file is stored in DBStore, after determining that the 4th file to be operated is FileStore storage class, directly whole the 4th file existed with large files form is cut into slices, each section is distributed in FileStore and stores.
In order to achieve the above object, the invention allows for a kind of storage system, this system comprises: acquisition module and processing module.
Acquisition module, before operating file in application layer, obtains the predetermined storage class of this file by the index node inode list preset.
Processing module, processes this file accordingly for the storage class corresponding according to this file.
Preferably,
Storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
Preferably,
This system also comprises and arranges module; Module is set for pre-setting inode list in the following manner:
By different file to be operated and different inode value one_to_one corresponding.
By the storage class one_to_one corresponding of inode value and corresponding file.
Corresponding relation between the storage class of this file, inode value and file is kept in inode list, and inode list is kept in DBStore.
Preferably,
This system also comprises determination type module; Determination type module is used for the action type pre-determining different file to be operated in the following manner:
The size of file and the threshold value preset are compared.
When the size of file is less than default threshold value, file is judged to be small documents, and small documents is defined as DBStore storage class.
When the size of file is more than or equal to default threshold value, file is judged to be large files, and large files is defined as FileStore storage class.
Preferably,
Processing module is carried out corresponding process according to the storage class that file is corresponding to file and is referred to:
In creation operation, determine that the first file to be operated is after DBStore storage class, file to be operated is directly stored in DBStore; Wherein, the first file to be operated is metadata, and metadata is small documents.
In deletion action, determine that the second file to be operated is after DBStore storage class, directly obtain the positional information of the second file in DBStore to be operated, and the second file to be operated is deleted from DBStore; Wherein, the second file to be operated is metadata, and metadata is small documents.
In read request operation, after determining that the 3rd file to be operated is DBStore storage class, determine the 3rd position of file in DBStore to be operated according to inode value, and from DBStore, read the 3rd file to be operated; After determining that the 3rd file to be operated is FileStore storage class, determine the 3rd position of file in FileStore to be operated according to inode value, and from FileStore, read the 3rd file to be operated.
In write request operation, after determining that the 4th file to be operated is DBStore storage class, multiple the 4th file existed with small documents form is carried out record according to the sequencing of write, and by the 4th file and inode value one_to_one corresponding, form mapping pair, 4th file is stored in buffer zone with the form of mapping pair, and mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, when the mapping pair mapping log recording reaches default amount threshold, continue to perform write operation according to the write operation recorded in Operation Log, according to the sequencing of write, the 4th file is stored in DBStore, after determining that the 4th file to be operated is FileStore storage class, directly whole the 4th file existed with large files form is cut into slices, each section is distributed in FileStore and stores.
Compared with prior art, the present invention includes: before application layer operates file, obtained the predetermined storage class of this file by the index node inode list preset.The storage class corresponding according to this file processes accordingly to this file.By the solution of the present invention, the frequent addressing of disk a large amount of random small documents can being avoided to write bring, the problem of degraded performance, improve write efficiency.
Accompanying drawing explanation
Be described the accompanying drawing in the embodiment of the present invention below, the accompanying drawing in embodiment is for a further understanding of the present invention, is used from explanation the present invention, does not form limiting the scope of the invention with instructions one.
Fig. 1 is storage means process flow diagram of the present invention;
Fig. 2 is storage system composition frame chart of the present invention.
Embodiment
For the ease of the understanding of those skilled in the art, below in conjunction with accompanying drawing, the invention will be further described, can not be used for limiting the scope of the invention.
The present invention is intended to develop a kind of novel storage means and solves the large small documents write problem that current distributed storage faces, and the application scenarios for user's complexity provides general solution.
In order to achieve the above object, the present invention proposes a kind of storage means, as shown in Figure 1, the method comprises:
S101, before application layer operates file, obtain the predetermined storage class of this file by the index node inode list preset.
Preferably,
Storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
For the ease of user operation, in advance rear end is stored in the present invention program and be divided into two types, namely the storage DBStore based on database (as LevelDB) and the storage FileStore based on file, in an embodiment of the present invention, the former is applicable to small documents, and the latter is applicable to large files.Sum up according to reality test, take 128KB as boundary in the present invention, the data acquisition DBStore being less than or equal to this value stores, and the data acquisition FileStore being greater than this value stores.
Preferably,
The method also comprises: pre-set inode list in the following manner:
S201, by different file to be operated and different inode value one_to_one corresponding.
S202, by the storage class one_to_one corresponding of inode value and corresponding file.
S203, the corresponding relation between the storage class of this file, inode value and file to be kept in inode list, and inode list is kept in DBStore.
In embodiments of the present invention, by such scheme, make operator carry out also to find action type corresponding to this file fast when adding operation to application layer file in application layer, carry out fast operating, improve operating efficiency.
In addition, in order to when user operation, rapid pin carries out DBStore storage or FileStore storage to the operation file of user according to its file size, need the classification in advance operation file corresponding for various operation being carried out storage class, different operation files is corresponded in advance in its corresponding action type, when user carries out this operation, can action type that directly call operation file is corresponding store, the frequent addressing of disk a large amount of random small documents can being avoided to write bring, the problem of degraded performance, improve write efficiency.
Preferably,
The method also comprises: the action type pre-determining different file to be operated in the following manner:
S301, by the size of this file with preset threshold value compare.
In embodiments of the present invention, this threshold value preset can be set to above-mentioned 128KB, and certainly, this threshold value is not limited to this, also can be set to other numerical value as the case may be.
S302, when the size of file is less than default threshold value, this file is judged to be small documents, and this small documents is defined as DBStore storage class.
When the size of file is more than or equal to default threshold value, this file is judged to be large files, and this large files is defined as FileStore storage class.
S102, according to the storage class that this file is corresponding, this file to be processed accordingly.
Classification that rear end stores is being completed in advance and after determining action type corresponding to operation file corresponding to various operation by above-mentioned scheme, carry out creating user, delete, the operation such as read request, write request time, just can action type that directly call operation file is corresponding store, improve work efficiency, below we will be described in detail for different action types respectively.
Preferably,
Carry out corresponding process according to the storage class that file is corresponding to file to comprise:
Situation one, in creation operation, determine that the first file to be operated is after DBStore storage class, file to be operated is directly stored in DBStore; Wherein, the first file to be operated is metadata, and metadata is small documents.
In embodiments of the present invention, creation operation can think its correspondence operation file be that data volume is almost 0, it provide only the metadata information of file, and general metadata all thinks small documents data, therefore, in creation operation directly by metadata store in DBStore.
Situation two, in deletion action, determine that the second file to be operated is after DBStore storage class, directly obtain the positional information of the second file in DBStore to be operated, and the second file to be operated is deleted from DBStore; Wherein, the second file to be operated is metadata, and metadata is small documents.
In embodiments of the present invention, deletion action generally needs the metadata information reading deleted file, find the memory location of this metadata, and it is deleted, illustrate in above-mentioned creation operation, general metadata all thinks small documents data, therefore, operation file corresponding to deletion action is also necessarily stored in DBStore, therefore, in deletion action, we directly read the file metadata be stored in DBStore in DBStore, obtain its memory location, then the data of correspondence are deleted one by one.
Situation three, read-write requests operation in, the operation file of general meeting first corresponding to the read request operation of preserving in inode list or write request operation determines the storage class corresponding to this file, and then sends read request operation or write request operation in corresponding storage.Below safety pin to read request operation and write request operation be described.
In read request operation, after determining that the 3rd file to be operated is DBStore storage class, determine the 3rd position of file in DBStore to be operated according to inode value, and from DBStore, read the 3rd file to be operated; After determining that the 3rd file to be operated is FileStore storage class, determine the 3rd position of file in FileStore to be operated according to inode value, and from FileStore, read the 3rd file to be operated.
In embodiments of the present invention, operate for read request, the file searched corresponding to read request can be arrived first in buffer memory, if existed in buffer memory, then directly return results, if do not had in buffer memory, just go to the position getting the file corresponding to read request in DBStore according to inode value, then go to the relevant position in DBStore storage or FileStore storage to obtain this file object according to such scheme.
In write request operation, after determining that the 4th file to be operated is DBStore storage class, multiple the 4th file existed with small documents form is carried out record according to the sequencing of write, and by the 4th file and inode value one_to_one corresponding, form mapping pair, 4th file is stored in buffer zone with the form of mapping pair, and mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, afterwards when the mapping pair mapping log recording reaches default amount threshold, continue to perform write operation according to the action type recorded in Operation Log, according to the sequencing of write, the 4th file is stored in DBStore, the write efficiency of application layer is ensured with this, after determining that the 4th file to be operated is FileStore storage class, directly whole the 4th file existed with large files form is cut into slices, each section is distributed in FileStore and stores, obtain higher aggregate bandwidth.
In embodiments of the present invention, for write request operation, can ensure that size file data can rule efficiently by such scheme.
In order to achieve the above object, the invention allows for a kind of storage system 01, as shown in Figure 2, this system comprises: acquisition module 02 and processing module 03.
Acquisition module 02, before operating file in application layer, obtains the predetermined storage class of this file by the index node inode list preset.
Preferably,
Storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
Processing module 03, processes this file accordingly for the storage class corresponding according to this file.
Preferably,
This system also comprises and arranges module 04; Module 04 is set for pre-setting inode list in the following manner:
By different file to be operated and different inode value one_to_one corresponding.
By the storage class one_to_one corresponding of inode value and corresponding file.
Corresponding relation between the storage class of this file, inode value and file is kept in inode list, and inode list is kept in DBStore.
Preferably,
This system also comprises determination type module 05; Determination type module 05 is for pre-determining the action type of different file to be operated in the following manner:
The size of file and the threshold value preset are compared.
When the size of file is less than default threshold value, file is judged to be small documents, and small documents is defined as DBStore storage class.
When the size of file is more than or equal to default threshold value, file is judged to be large files, and large files is defined as FileStore storage class.
Preferably,
Processing module 03 is carried out corresponding process according to the storage class that file is corresponding to file and is referred to:
In creation operation, determine that the first file to be operated is after DBStore storage class, file to be operated is directly stored in DBStore; Wherein, the first file to be operated is metadata, and metadata is small documents.
In deletion action, determine that the second file to be operated is after DBStore storage class, directly obtain the positional information of the second file in DBStore to be operated, and the second file to be operated is deleted from DBStore; Wherein, the second file to be operated is metadata, and metadata is small documents.
In read request operation, after determining that the 3rd file to be operated is DBStore storage class, determine the 3rd position of file in DBStore to be operated according to inode value, and from DBStore, read the 3rd file to be operated; After determining that the 3rd file to be operated is FileStore storage class, determine the 3rd position of file in FileStore to be operated according to inode value, and from FileStore, read the 3rd file to be operated.
In write request operation, after determining that the 4th file to be operated is DBStore storage class, multiple the 4th file existed with small documents form is carried out record according to the sequencing of write, and by the 4th file and inode value one_to_one corresponding, form mapping pair, 4th file is stored in buffer zone with the form of mapping pair, and mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, when the mapping pair mapping log recording reaches default amount threshold, continue to perform write operation according to the write operation recorded in Operation Log, according to the sequencing of write, the 4th file is stored in DBStore, after determining that the 4th file to be operated is FileStore storage class, directly whole the 4th file existed with large files form is cut into slices, each section is distributed in FileStore and stores.
Compared with prior art, the present invention includes: before application layer operates file, obtained the predetermined storage class of this file by the index node inode list preset.The storage class corresponding according to this file processes accordingly to this file.By the solution of the present invention, the frequent addressing of disk a large amount of random small documents can being avoided to write bring, the problem of degraded performance, improve write efficiency.
Classification and the common cooperation of different action types of the storage class in the present invention program complete the read-write of size file data, and compared to traditional storage means, storage means of the present invention mainly contains following characteristics:
1) efficient metadata management.Metadata due to file has been put in DBStore as small block data to be deposited, so good performance can be obtained when random read-write, and the deposit position of management document each several part data object is carried out in usage data storehouse, efficiently, fast, and database has born transactional operating feature, namely, the continuity operations of affairs, when any one link is made mistakes, all can stop the operation of this things, this ensure that the consistance of data admirably.
2) aggregate bandwidth.In the present invention program when processing the write of small documents data, can first by action type (as, be read request operation or write request operation) and the data of operation file be recorded in the journal file of disk, and make a mark in the operating list safeguarded at internal memory, seek unity of action again after reaching predetermined value Deng the operation amount in journal file, realize the effect be sequentially written in, so both avoid a large amount of random small documents and write the frequent addressing of disk brought, improve write efficiency, can ensure that again system can carry out operation playback based on journal file when unexpected power-off, guarantee data consistency.When chunk data writes, the concurrent feature of the height of each operation and various respective behavior can be relied on again to obtain the effect of aggregate bandwidth.
It should be noted that; above-described embodiment is only understand for the ease of those skilled in the art; be not limited to protection scope of the present invention; under the prerequisite not departing from inventive concept of the present invention, any apparent replacement and improvement etc. that those skilled in the art make the present invention are all within protection scope of the present invention.
Claims (10)
1. a storage means, is characterized in that, described method comprises:
Before application layer operates file, obtained the predetermined storage class of described file by the index node inode list preset;
The storage class corresponding according to described file processes accordingly to described file.
2. storage means as claimed in claim 1, it is characterized in that, described storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
3. storage means as claimed in claim 2, it is characterized in that, described method also comprises: pre-set described inode list in the following manner:
By different file to be operated and different inode value one_to_one corresponding;
By the storage class one_to_one corresponding of described inode value and corresponding file;
Corresponding relation between the storage class of described file, described inode value and described file is kept in described inode list, and described inode list is kept in described DBStore.
4. storage means as claimed in claim 2, it is characterized in that, described method also comprises: the action type pre-determining different file to be operated in the following manner:
The size of described file and the threshold value preset are compared;
When the size of described file is less than described default threshold value, described file is judged to be small documents, and described small documents is defined as described DBStore storage class;
When the size of described file is more than or equal to described default threshold value, described file is judged to be large files, and described large files is defined as described FileStore storage class.
5. storage means as claimed in claim 4, is characterized in that, the described storage class corresponding according to described file is carried out corresponding process to described file and comprised:
In creation operation, determine that described first file to be operated is after described DBStore storage class, described file to be operated is directly stored in described DBStore; Wherein, described first file to be operated is metadata, and described metadata is small documents;
In deletion action, determine that described second file to be operated is after described DBStore storage class, the described positional information of the second file in described DBStore to be operated of direct acquisition, and described second file to be operated is deleted from described DBStore; Wherein, described second file to be operated is metadata, and described metadata is small documents;
In read request operation, after determining that described 3rd file to be operated is described DBStore storage class, determine described 3rd position of file in described DBStore to be operated according to described inode value, and from described DBStore, read described 3rd file to be operated; After determining that described 3rd file to be operated is described FileStore storage class, determine described 3rd position of file in described FileStore to be operated according to described inode value, and from described FileStore, read described 3rd file to be operated;
In write request operation, after determining that described 4th file to be operated is described DBStore storage class, multiple described 4th file existed with small documents form is carried out record according to the sequencing of write, and by described 4th file and described inode value one_to_one corresponding, form mapping pair, described 4th file is stored in buffer zone with the form of mapping pair, and described mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, when the described mapping pair of described mapping log recording reaches default amount threshold, continue to perform write operation according to the action type recorded in Operation Log, according to the sequencing of write, described 4th file is stored in described DBStore, after determining that described 4th file to be operated is described FileStore storage class, directly whole described 4th file existed with large files form is cut into slices, each section is distributed in described FileStore and stores.
6. a storage system, is characterized in that, described system comprises: acquisition module and processing module;
Described acquisition module, before operating file in application layer, obtains the predetermined storage class of described file by the index node inode list preset;
Described processing module, processes described file accordingly for the storage class corresponding according to described file.
7. storage system as claimed in claim 6, it is characterized in that, described storage class comprises: the database purchase DBStore based on database and the file based on file store FileStore.
8. storage system as claimed in claim 7, is characterized in that, described system also comprises and arranges module; The described module that arranges is for pre-setting described inode list in the following manner:
By different file to be operated and different inode value one_to_one corresponding;
By the storage class one_to_one corresponding of described inode value and corresponding file;
Corresponding relation between the storage class of described file, described inode value and described file is kept in described inode list, and described inode list is kept in described DBStore.
9. storage system as claimed in claim 7, it is characterized in that, described system also comprises determination type module; Described determination type module is used for the action type pre-determining different file to be operated in the following manner:
The size of described file and the threshold value preset are compared;
When the size of described file is less than described default threshold value, described file is judged to be small documents, and described small documents is defined as described DBStore storage class;
When the size of described file is more than or equal to described default threshold value, described file is judged to be large files, and described large files is defined as described FileStore storage class.
10. storage system as claimed in claim 9, is characterized in that, described processing module is carried out corresponding process according to the storage class that described file is corresponding to described file and referred to:
In creation operation, determine that described first file to be operated is after described DBStore storage class, described file to be operated is directly stored in described DBStore; Wherein, described first file to be operated is metadata, and described metadata is small documents;
In deletion action, determine that described second file to be operated is after described DBStore storage class, the described positional information of the second file in described DBStore to be operated of direct acquisition, and described second file to be operated is deleted from described DBStore; Wherein, described second file to be operated is metadata, and described metadata is small documents;
In read request operation, after determining that described 3rd file to be operated is described DBStore storage class, determine described 3rd position of file in described DBStore to be operated according to described inode value, and from described DBStore, read described 3rd file to be operated; After determining that described 3rd file to be operated is described FileStore storage class, determine described 3rd position of file in described FileStore to be operated according to described inode value, and from described FileStore, read described 3rd file to be operated;
In write request operation, after determining that described 4th file to be operated is described DBStore storage class, multiple described 4th file existed with small documents form is carried out record according to the sequencing of write, and by described 4th file and described inode value one_to_one corresponding, form mapping pair, described 4th file is stored in buffer zone with the form of mapping pair, and described mapping pair is recorded in mapping daily record, return to write and successfully reply ACK, when the described mapping pair of described mapping log recording reaches default amount threshold, continue to perform write operation according to the action type recorded in Operation Log, according to the sequencing of write, described 4th file is stored in described DBStore, after determining that described 4th file to be operated is described FileStore storage class, directly whole described 4th file existed with large files form is cut into slices, each section is distributed in described FileStore and stores.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510370132.8A CN105094695B (en) | 2015-06-29 | 2015-06-29 | A kind of storage method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510370132.8A CN105094695B (en) | 2015-06-29 | 2015-06-29 | A kind of storage method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105094695A true CN105094695A (en) | 2015-11-25 |
| CN105094695B CN105094695B (en) | 2018-09-04 |
Family
ID=54575255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510370132.8A Active CN105094695B (en) | 2015-06-29 | 2015-06-29 | A kind of storage method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105094695B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105404691A (en) * | 2015-12-14 | 2016-03-16 | 曙光信息产业股份有限公司 | File storage method and apparatus |
| CN105487824A (en) * | 2015-12-07 | 2016-04-13 | 联想(北京)有限公司 | Information processing method, storage device and electronic device |
| CN105787012A (en) * | 2016-02-23 | 2016-07-20 | 深圳市瑞驰信息技术有限公司 | Method for improving small file processing capability of storage system and storage system |
| CN107315533A (en) * | 2016-04-26 | 2017-11-03 | 杭州海康威视数字技术股份有限公司 | A kind of date storage method and device |
| CN107580066A (en) * | 2017-09-20 | 2018-01-12 | 郑州云海信息技术有限公司 | The method, apparatus and system of file access in a kind of distributed NAS storage system |
| CN107967121A (en) * | 2017-10-25 | 2018-04-27 | 华为技术有限公司 | Method for writing data and storage device |
| CN108897698A (en) * | 2018-06-29 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of file data blocks addressing method, system and equipment and storage medium |
| CN109241011A (en) * | 2018-09-21 | 2019-01-18 | 联想(北京)有限公司 | A kind of virtual machine file processing method and processing device |
| CN111324581A (en) * | 2020-02-14 | 2020-06-23 | 苏州浪潮智能科技有限公司 | File writing method, device, equipment and storage medium |
| CN112822084A (en) * | 2019-11-18 | 2021-05-18 | 瑞昱半导体股份有限公司 | Gateway control chip and network packet processing method thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110106806A1 (en) * | 2009-11-02 | 2011-05-05 | Stg Interactive | Process for optimizing file storage systems |
| CN103246700A (en) * | 2013-04-01 | 2013-08-14 | 厦门市美亚柏科信息股份有限公司 | Mass small file low latency storage method based on HBase |
| CN104111804A (en) * | 2014-06-27 | 2014-10-22 | 暨南大学 | Distributed file system |
-
2015
- 2015-06-29 CN CN201510370132.8A patent/CN105094695B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110106806A1 (en) * | 2009-11-02 | 2011-05-05 | Stg Interactive | Process for optimizing file storage systems |
| CN103246700A (en) * | 2013-04-01 | 2013-08-14 | 厦门市美亚柏科信息股份有限公司 | Mass small file low latency storage method based on HBase |
| CN104111804A (en) * | 2014-06-27 | 2014-10-22 | 暨南大学 | Distributed file system |
Non-Patent Citations (3)
| Title |
|---|
| JEROME H.SALTZER 等: "《计算机系统设计原理》", 31 December 2012 * |
| 徐四鹏: "基于NoSQL的PDM图文档分布式存储与并行查询", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 文艾 等: "《Hadoop分布式文件系统深度实践》", 31 May 2012 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105487824A (en) * | 2015-12-07 | 2016-04-13 | 联想(北京)有限公司 | Information processing method, storage device and electronic device |
| CN105404691A (en) * | 2015-12-14 | 2016-03-16 | 曙光信息产业股份有限公司 | File storage method and apparatus |
| CN105787012B (en) * | 2016-02-23 | 2019-08-27 | 深圳市瑞驰信息技术有限公司 | A kind of method and storage system improving storage system processing small documents |
| CN105787012A (en) * | 2016-02-23 | 2016-07-20 | 深圳市瑞驰信息技术有限公司 | Method for improving small file processing capability of storage system and storage system |
| CN107315533A (en) * | 2016-04-26 | 2017-11-03 | 杭州海康威视数字技术股份有限公司 | A kind of date storage method and device |
| CN107580066A (en) * | 2017-09-20 | 2018-01-12 | 郑州云海信息技术有限公司 | The method, apparatus and system of file access in a kind of distributed NAS storage system |
| CN107967121A (en) * | 2017-10-25 | 2018-04-27 | 华为技术有限公司 | Method for writing data and storage device |
| CN107967121B (en) * | 2017-10-25 | 2020-04-14 | 华为技术有限公司 | Data writing method and storage device |
| US11237765B2 (en) | 2017-10-25 | 2022-02-01 | Huawei Technologies Co., Ltd. | Data writing method and storage device |
| US11880602B2 (en) | 2017-10-25 | 2024-01-23 | Huawei Technologies Co., Ltd. | Data writing method and storage device |
| CN108897698A (en) * | 2018-06-29 | 2018-11-27 | 郑州云海信息技术有限公司 | A kind of file data blocks addressing method, system and equipment and storage medium |
| CN108897698B (en) * | 2018-06-29 | 2021-10-29 | 郑州云海信息技术有限公司 | File data block addressing method, system and equipment and storage medium |
| CN109241011A (en) * | 2018-09-21 | 2019-01-18 | 联想(北京)有限公司 | A kind of virtual machine file processing method and processing device |
| CN109241011B (en) * | 2018-09-21 | 2023-01-06 | 联想(北京)有限公司 | Virtual machine file processing method and device |
| CN112822084A (en) * | 2019-11-18 | 2021-05-18 | 瑞昱半导体股份有限公司 | Gateway control chip and network packet processing method thereof |
| CN111324581A (en) * | 2020-02-14 | 2020-06-23 | 苏州浪潮智能科技有限公司 | File writing method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105094695B (en) | 2018-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105094695A (en) | Storing method and system | |
| CN105574104B (en) | A kind of LogStructure storage system and its method for writing data based on ObjectStore | |
| US10649897B2 (en) | Access request processing method and apparatus, and computer device | |
| CN104657500A (en) | Distributed storage method based on KEY-VALUE pair | |
| US9996421B2 (en) | Data storage method, data storage apparatus, and storage device | |
| CN106951375B (en) | Method and device for deleting snapshot volume in storage system | |
| CN109445685B (en) | User mode file system processing method | |
| CN101576890B (en) | Novel method and system for sorting fragment | |
| CN104111804A (en) | Distributed file system | |
| CN109804359A (en) | For the system and method by write back data to storage equipment | |
| US9430492B1 (en) | Efficient scavenging of data and metadata file system blocks | |
| CN107391544B (en) | Processing method, device and equipment of column type storage data and computer storage medium | |
| CN112684975B (en) | Data storage method and device | |
| CN104268088A (en) | Vehicle DVR (Digital Video Recorder) hard disk data storage method | |
| CN103744875B (en) | Data quick migration method and system based on file system | |
| CN104360914A (en) | Incremental snapshot method and device | |
| CN111143290A (en) | Memory management method, system, equipment and computer readable storage medium | |
| CN104281717A (en) | Method for establishing massive ID mapping relation | |
| CN107506466B (en) | Small file storage method and system | |
| CN110147203B (en) | File management method and device, electronic equipment and storage medium | |
| CN105493080A (en) | Method and apparatus for context aware based data de-duplication | |
| CN104239231A (en) | Method and device for accelerating preheating of second-level cache | |
| CN103631682A (en) | Data backup implement method and device | |
| CN101808243A (en) | Data processing method and device based on snapshot | |
| US20170269868A1 (en) | Information processing apparatus, storage system, computer-readable recording medium, and information processing 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |