CN106570113A - Cloud storage method and system for mass vector slice data - Google Patents
Cloud storage method and system for mass vector slice data Download PDFInfo
- Publication number
- CN106570113A CN106570113A CN201610939884.6A CN201610939884A CN106570113A CN 106570113 A CN106570113 A CN 106570113A CN 201610939884 A CN201610939884 A CN 201610939884A CN 106570113 A CN106570113 A CN 106570113A
- Authority
- CN
- China
- Prior art keywords
- data
- vector slice
- magnanimity
- slice
- magnanimity vector
- 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
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
- G06F16/134—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
- G06F16/18—File system types
- G06F16/182—Distributed file systems
Abstract
The invention discloses a cloud storage method for mass vector slice data. The method comprises the following steps of: establishing a distributed file system directory tree file; establishing all the metadata nodes corresponding to a distributed file system directory tree file; aggregating the mass vector slice data under the same directories in a distributed file system so as to generate a mass vector slice data packet; storing the mass vector slice data packet in the metadata nodes; establishing indexes for the mass vector slice data, and correlating the mass vector slice data through the indexes so as to form a net-structure data index table of the mass vector slice data, wherein the index table is used for recording paths of the mass vector slice data in the mass vector slice data packets; and providing mass vector slice data indexing services through a mass vector slice data packet index table.
Description
Technical field
The present invention relates to mass data storage field, more particularly, to a kind of magnanimity vector slice of data cloud storage side
Method and system.
Background technology
With the continuous development of science and technology, the mass data epoch have arrived.Therefore, how file optimizing system is negative
Carry, lifting the harmony of load becomes demand important at present.When the size of data set is more than an independent physical computer
Storage capacity when, it is therefore necessary to it carry out subregion and store on some independent computers.Google, Amazon, IBM
Substantial amounts of scientific research strength has been put into the international major company such as Microsoft in this field, it is proposed that the Mass Data Management skill of various innovations
Art.At present research work is concentrated mainly on accumulation layer, computation layer and interface layer this 3 levels.The Hadoop projects of prior art
Realize Hadoop distributed file system Hadoop DFS (abbreviation HDFS), and multiple programming framework Hadoop
MapReduce.Distributed file system framework introduces the complexity of network programming, therefore distributed field system on network
System is more increasingly complex than ordinary magnetic disc file.The target of distributed file system is to realize resource-sharing, program picture is stored and is visited
Ask that its Typical Representative is Google file system GFS, Hadoop similar to the mode of local file is accessed to Remote File Manipulation
File system HDFS, dynamo, TFS etc..Present distributed file system generally remains almost identical with local file system
Access interface and object model, this is primarily to provide a user with compatibility backward.
Prior art is primarily directed to super large rank (referring to that file size is hundreds of MB, GB or TB) data file and adopts and be based on
Distributed file system is stored and read.But carry out based on distributed file system for large amount of small documents data, by
It is slow in storage speed, it is impossible to meet the storage demand of large amount of small documents data.Currently without for storage large amount of small documents data
Carry out the technical scheme for being stored based on distributed file system and being read.
The content of the invention
In order to solve the speed issue that large amount of small documents data carry out when being stored based on distributed file system, this
Bright to provide a method that, methods described includes:
Set up all metadata nodes corresponding with distributive catalogue of document system tree;
To be polymerized with the magnanimity vector slice of data under first class catalogue in distributed file system, generated magnanimity vector
Slice of data bag;
The magnanimity vector slice of data bag is stored in the metadata node;
Set up for the magnanimity vector slice of data and index, the magnanimity vector slice of data sets up association by index,
Form the data directory of cancellated magnanimity vector slice of data;
The magnanimity vector slice of data index service is provided by the magnanimity vector slice of data bag concordance list.
Preferably, method according to claim 1, methods described includes:
The magnanimity vector slice of data index includes the magnanimity vector slice of data path, title and in the sea
Side-play amount in amount vector slice of data bag;
The magnanimity vector slice of data path includes that first site position, magnanimity vector slice of data line position are put and magnanimity
Vector slice of data column position.
Preferably, methods described includes:
The default unitary Data Node of each layer, concordance list is stored in each layer be pre-designed of metadata node;
The magnanimity vector slice of data concordance list stored in the metadata is transmitted to client, magnanimity arrow is set up
The lasting mapping table of amount slice of data concordance list.
Preferably, the magnanimity vector slice of data bag includes file header and at least one record;
The file header includes file type, version number, document keyword, file name, per recording corresponding described in bar
Position;
Per described in bar record correspondence one vector slice of data, it is described per bar record include vector slice of data length,
Key length, key and value.
Preferably, the magnanimity vector slice of data bag is stored using data file sequencing method.
Preferably, also include:Carry out adding storage in the afterbody of the magnanimity vector slice of data bag.
Preferably, methods described includes:The magnanimity vector slice of data is indexed into table cache to client, is reduced and is accessed
The metadata node number of times accesses the access times of magnanimity vector slice of data to improve.
Preferably, also include:The method that magnanimity vector slice of data is read out:
The corresponding unit of the magnanimity vector slice of data bag is determined by the magnanimity vector slice of data concordance list
Back end shortest path;
By it is determined that metadata node in file header in data APMB package, determine the vector slice of data
Position.
Based on embodiments of the present invention, the present invention provides a kind of cloud storage system for magnanimity vector slice of data,
The system includes:
First signal generating unit, for setting up distributive catalogue of document system tree file;
Second signal generating unit, for setting up all metadata nodes corresponding with distributive catalogue of document system tree;
Polymerized unit, for will be gathered with the magnanimity vector slice of data under first class catalogue based on distributed file system
Close, generate magnanimity vector slice of data bag;
Memory element, for the magnanimity vector slice of data bag to be stored in the metadata node;
3rd signal generating unit, for generating the magnanimity vector slice of data concordance list, by concordance list the sea is set up
The network structure of amount vector slice of data bag, for recording the magnanimity vector slice of data in the magnanimity vector slice of data
Path in bag;
Indexing units, for providing the magnanimity vector slice of data index by magnanimity vector slice of data index
Service.
Beneficial effects of the present invention are:To enter with the magnanimity vector slice of data under first class catalogue in distributed file system
Row polymerization, generates magnanimity vector slice of data bag so that magnanimity vector slice of data realizes quick storage.Propose simultaneously as sea
Amount vector slice of data sets up index, and magnanimity vector slice of data sets up association by index, forms cancellated magnanimity arrow
The data directory of amount slice of data.By the data directory of network structure, realization finds corresponding unit by shortest path
Back end, accelerates the access speed of data.
Description of the drawings
By reference to the following drawings, the illustrative embodiments of the present invention can be more fully understood by:
Fig. 1 is according to a kind of magnanimity vector slice of data cloud storage method system flow chart of embodiment of the present invention;And
Fig. 2 is according to a kind of magnanimity vector slice of data cloud storage method system structure chart of embodiment of the present invention.
Specific embodiment
With reference now to accompanying drawing, the illustrative embodiments of the present invention are introduced, however, the present invention can be with many different shapes
Formula is not limited to embodiment described herein implementing, there is provided these embodiments are to disclose at large and fully
The present invention, and fully pass on the scope of the present invention to person of ordinary skill in the field.For showing for being illustrated in the accompanying drawings
Term in example property embodiment is not limitation of the invention.In the accompanying drawings, identical cells/elements are attached using identical
Icon is remembered.
Unless otherwise stated, term (including scientific and technical terminology) used herein has to person of ordinary skill in the field
It is common to understand implication.Further it will be understood that the term limited with the dictionary being usually used, is appreciated that and it
The linguistic context of association area has consistent implication, and is not construed as Utopian or excessively formal meaning.
Fig. 1 is according to a kind of magnanimity vector slice of data cloud storage method system flow chart of embodiment of the present invention.This
It is bright to propose a kind of method that magnanimity vector slice of data based on distributed file system is stored.The solution of the present invention is with existing
Based on distributive catalogue of document system tree construction, the multiple magnanimity vector slice of datas in a catalogue are packaged into into magnanimity arrow
Amount slice of data bag is stored, and the magnanimity vector slice of data bag being packaged into is large data files, and file-level is in 100 MB
More than.Meanwhile, technical scheme life magnanimity vector slice of data sets up index, and record magnanimity vector slice of data is in sea
Path in amount vector slice of data bag, accesses magnanimity vector slice of data and provides interface for client.The method of the present invention
Make full use of in the advantage of the high fault-tolerant of master-salve distributed file system, extensibility and distributivity, it is super in object oriented file rank
On the basis of crossing the distributed file system of 100 MB, the efficient storage of massive vector data is realized.Method proposed by the present invention makes
Massive vector data is stored with distributed file system, while massive vector data is set up indexing, storage sea at present is solved
The slow-footed problem of amount vector data, and improve access speed by setting up index.
Preferably, method 100 starts to walk from step 101:Set up distributive catalogue of document system tree file.Build distributed text
Part system directory tree construction file, can make full use of the high fault-tolerant of distributed file system, extensibility and distributed excellent
Point.
Preferably, step 102:Set up all metadata nodes corresponding with distributive catalogue of document system tree.Metadata
Node is used for data storage.
Preferably, step 103:To be gathered with the magnanimity vector slice of data under first class catalogue in distributed file system
Close, generate magnanimity vector slice of data bag.The file structure of design magnanimity vector slice of data bag, magnanimity vector slice of data bag
Including file header and at least one record.File header includes file type, version number, document keyword, file name, remembers per bar
Record corresponding position.One vector slice of data of correspondence is recorded per bar, record includes length, the bond distance of vector slice of data per bar
Degree, key and value.The additional of magnanimity vector slice of data is stored as being added in the afterbody of magnanimity vector slice of data bag
Storage.Magnanimity vector slice of data bag is stored using data file sequencing method.Embodiment proposed by the present invention, uses
In the method in a distributed manner based on system architecture of massive vector data cloud storage, by a metadata node and metadata node
The back end composition of lower multilevel hierarchy.Embodiments of the present invention are by the magnanimity vector slice of data whole under same first class catalogue
It is saved in the data file under the catalogue, is the data file magnanimity vector slice of data bag in the present invention, is distributed text
File in part system.In embodiment of the present invention, the memory technology that is polymerized it is critical only that magnanimity vector slice of data APMB package
Design.Magnanimity vector slice of data APMB package is distributed using binary system key/value (Key/Value) perdurable data structure
File system files, it is made up of file header and one or more subsequent record.Magnanimity vector slice of data APMB package head
First three byte for SEQ file type, the version number of a byte representation file data structure followed by.File header is also
Including some other field, including the content such as the title of key and value respective type.Magnanimity vector slice of data is when being stored
Directly the afterbody in magnanimity vector slice of data APMB package is added.Per bar, record represents a vector slice of data.Record
It is made up of the length, key length, key, value four that record.Wherein the value of key is the filename of vector slice of data, is worth and is cut for vector
The content of sheet data.
Preferably, step 104:Magnanimity vector slice of data bag is stored in metadata node.Magnanimity vector number of slices
It is that base distributed file system is realized according to bag storage method, it depends on distribution to the operation that magnanimity vector slice of data is accessed
Formula file system.The additional of magnanimity vector slice of data is stored as carrying out adding in the afterbody of magnanimity vector slice of data bag depositing
Storage.Magnanimity vector slice of data bag is stored using data file sequencing method.Vector is cut into slices when there is a client
When data are write under certain catalogue, the client can carry out write operation, distributed file system note to the data file of the catalogue
What occupancy authority Lease for having recorded the data file was considered as file writes lock.If now another client is also required to certainly
Oneself vector slice of data is stored under identical catalogue, and equally it can also go application to the magnanimity vector number of slices under the catalogue
Write operation is carried out according to APMB package.Lock, and distributed document are write because magnanimity vector slice of data APMB package has had one
System is not carried out the maintenance of transactions requests queue, the result for directly returning operation failure to client.From from the point of view of user,
It is not in conflict that different magnanimity vector slice of data APMB packages are created under same catalogue, but is in fact in rear end
The operation that same magnanimity vector slice of data APMB package is carried out, because such lock mechanism just occurs multiple users same
To the problem of different vector slice of data write conflicts under one catalogue.The realization of magnanimity vector slice of data APMB package mainly adopts number
According to the sequence and unserializing method of file.So-called serializing, refers to and for structured object to be converted into byte stream, so as in network
Upper transmission is write and permanently stored on disk.Unserializing is referred to the inverse process of the byte stream meeting of conversion structured object.
Preferably, step 105:Set up for magnanimity vector slice of data and index, magnanimity vector slice of data is built by index
Vertical association, forms the data directory of cancellated magnanimity vector slice of data;Concordance list is used to record the section of magnanimity vector
Path of the data in magnanimity vector slice of data bag.Magnanimity vector slice of data index includes magnanimity vector slice of data road
Footpath, title and the side-play amount in magnanimity vector slice of data bag, magnanimity vector slice of data path include first site position,
Magnanimity vector slice of data line position is put and magnanimity vector slice of data column position.For example, a magnanimity vector slice of data road
Footpath includes<18,0506>, wherein 18 is metadata site position, 05 puts for magnanimity vector slice of data line position, and 06 is magnanimity arrow
Amount slice of data column position.When making a look up to this magnanimity vector slice of data, by positioning metadata site position 18, then
Corresponding row 05 is continued to search for, then makes a look up corresponding row 06.All magnanimity vector slice of datas are according to path in concordance list
Metadata site position, magnanimity vector slice of data line position is put and magnanimity vector slice of data column position Special composition is netted
Index structure.Embodiments of the present invention can realize the shortest path that magnanimity vector slice of data is searched.
Each layer of metadata node presets a metadata node for being used for data storage concordance list, by the section of magnanimity vector
Data directory is stored in corresponding metadata node.By the magnanimity vector slice of data concordance list recorded in metadata transmit to
Catalogue file, and set up the lasting mapping table of magnanimity vector slice of data index in client.
The index record of vector slice of data vector slice of data in concrete magnanimity vector slice of data APMB package
Position and other attributes of vector slice of data, it is that client is necessary after the data for having stored magnanimity vector slice of data
It to be its establishment.The magnanimity that the title comprising magnanimity vector slice of data, magnanimity vector slice of data are located in index record
Vector slice of data APMB package path and the side-play amount in magnanimity vector slice of data APMB package.Magnanimity vector slice of data
Digit shared by APMB package name determines the quantity of data file under a catalogue, and the digit shared by side-play amount determines data text
The capacity of data storage is limited under the size of part, therefore one catalogue of explanation.
Preferably, magnanimity vector slice of data index is distributed to each back end to manage.Magnanimity vector slice of data
Although index data it is very huge, after being distributed in metadata node, the index data in single metadata node is with regard to phase
To very little, and the ability of cluster-based storage magnanimity vector slice of data depends on the scale of cluster.The size of cluster scale is not only
The size of memory capacity can be determined, the size of storage magnanimity vector slice of data quantity can be more embodied.Metadata node is safeguarded
The index of vector slice of data, and provide index service to client.Vector is safeguarded in the index position description of vector slice of data
The metadata node of slice of data index.
Preferably, the index of magnanimity vector slice of data is classified according to its parent directory being located, and its objective is will be same
Magnanimity vector slice of data index under one catalogue is managed by the metadata node of same one-level.In view of the feature, the present invention
Embodiment creates index position mapping table to record the mapping relations of catalogue and metadata node.Index position mapping table is by unit
Back end is managed.Client is when magnanimity vector slice of data index is inquired about, it is necessary first to knows and safeguards the extra large vector
The metadata node position of slice of data index.It by the way that the path of magnanimity vector slice of data is passed to into metadata node, so
Afterwards metadata node finds metadata node position according to the parent directory search index position mapping table in extra large vector slice of data path
Put.The present invention designs index position maintenance module in metadata node, dedicated for distributing back end for catalogue, safeguards index
Position mapping table.
Preferably, index position maintenance module therefrom selects to distribute to according to the total data node that metadata node is safeguarded
Catalogue.Index position mapping table is persisted on local disk, and when its data change, the content on its disk also will
Re-start renewal.If index position maintenance module can not find enough metadata sections when metadata node is distributed to catalogue
Point, the module can wait unappropriated direct insertion to catalogue in distribution queue, while the content of the queue also wants persistence
To on disk, queue once has new catalogue to add or deletion is required for updating again on disk.When metadata node starts
Need queuing data on disk to be read in internal memory.The purpose of the queue is to wait for distributed file system new data section
Point is registered plus fashionable, and index position maintenance module is redistributed to the catalogue in queue.Same queue updates also need every time
Carry out persistence.
Embodiment of the present invention on back end by designing vector slice of data index module come maintenance management vector
The index of slice of data, to client index service is provided.Module safeguard the index record and index file in internal memory and with
The corresponding journal file of index file.Metadata node is ranked up to accelerate the lookup for indexing to visit to index record with B-tree
Ask.The renewal of index record first can modify to memory data structure, temporarily asynchronous to correspond to index file.But will more
New content record as needed arranges index file after back end starts in the corresponding Log files of the index file
Sequence reads in internal memory, and index data structure is updated according to Log, and will now index record stores data again in internal memory
Old index file is replaced on node, Log is emptied.The purpose of do so is in order to avoid the unexpected power-off of back end causes internal memory
In index data lose.
Preferably, magnanimity vector slice of data is indexed into table cache to client, reduces and access the metadata node time
Number accesses the access times of magnanimity vector slice of data to improve.Embodiments of the present invention, by client-cache user
Commonly used magnanimity vector slice of data index, it is possible to reduce access times of the client to metadata node, improves to magnanimity
The efficiency that vector slice of data is accessed.
Preferably, step 106:Magnanimity vector slice of data index clothes are provided by magnanimity vector slice of data bag concordance list
Business.The corresponding metadata node shortest path of magnanimity vector slice of data bag is determined by magnanimity vector slice of data concordance list.
By it is determined that metadata node in file header in data APMB package, determine the position of vector slice of data.
Fig. 2 is according to a kind of magnanimity vector slice of data cloud storage method system structure chart of embodiment of the present invention.System
200 include:
First signal generating unit 201, for setting up distributive catalogue of document system tree file;
Second signal generating unit 202, for setting up all metadata nodes corresponding with distributive catalogue of document system tree;
Polymerized unit 203, for will be entered with the magnanimity vector slice of data under first class catalogue based on distributed file system
Row polymerization, generates magnanimity vector slice of data bag;
Memory element 204, for magnanimity vector slice of data bag to be stored in metadata node;
3rd signal generating unit 205, for generating magnanimity vector slice of data concordance list, by concordance list magnanimity vector is set up
The network structure of slice of data bag, for recording path of the magnanimity vector slice of data in magnanimity vector slice of data bag;
Indexing units 206, for providing magnanimity vector slice of data index service by magnanimity vector slice of data index.
Magnanimity vector slice of data cloud storage method system 200 a kind of according to the embodiment of the present invention is another with the present invention
A kind of magnanimity vector slice of data cloud storage method system 100 of embodiment is corresponding, and here is no longer repeated.
Beneficial effects of the present invention are:To enter with the magnanimity vector slice of data under first class catalogue in distributed file system
Row polymerization, generates magnanimity vector slice of data bag so that magnanimity vector slice of data realizes quick storage.Propose simultaneously as sea
Amount vector slice of data sets up index, and magnanimity vector slice of data sets up association by index, forms cancellated magnanimity arrow
The data directory of amount slice of data.By the data directory of network structure, realization finds corresponding unit by shortest path
Back end, accelerates the access speed of data.
The present invention is described by reference to a small amount of embodiment.However, it is known in those skilled in the art, as
What subsidiary Patent right requirement was limited, except the present invention other embodiments disclosed above equally fall the present invention's
In the range of.
Normally, all terms for using in the claims are all solved according to them in the usual implication of technical field
Release, unless clearly defined in addition wherein.It is all of to be all opened ground with reference to " one/described/be somebody's turn to do [device, component etc.] "
At least one of described device, component etc. example is construed to, unless otherwise expressly specified.Any method disclosed herein
Step all need not be run with disclosed accurate order, unless explicitly stated otherwise.
In addition, those skilled in the art are it should be appreciated that embodiments herein can be provided as method, system or calculate
Machine program product.Therefore, the application can be using complete hardware embodiment, complete software embodiment or with reference to software and hardware side
The form of the embodiment in face.And, the application can be adopted and wherein include computer usable program code at one or more
The computer implemented in computer-usable storage medium (including but not limited to disk memory, CD-ROM, optical memory etc.)
The form of program product.
The application is the flow process with reference to method, equipment (system) and computer program according to the embodiment of the present application
Figure and/or block diagram are describing.It should be understood that can be by computer program instructions flowchart and/or each stream in block diagram
The combination of journey and/or square frame and flow chart and/or the flow process in block diagram and/or square frame.These computer programs can be provided
The processor of general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced for reality by the instruction of computer or the computing device of other programmable data processing devices
The device of the function of specifying in present one flow process of flow chart or one square frame of multiple flow processs and/or block diagram or multiple square frames.
These computer program instructions may be alternatively stored in can guide computer or other programmable data processing devices with spy
In determining the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory is produced to be included referring to
Make the manufacture of device, the command device realize in one flow process of flow chart or one square frame of multiple flow processs and/or block diagram or
The function of specifying in multiple square frames.
These computer program instructions also can be loaded in computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented process, so as in computer or
The instruction performed on other programmable devices is provided for realizing in one flow process of flow chart or multiple flow processs and/or block diagram one
The step of function of specifying in individual square frame or multiple square frames.
Claims (9)
1. a kind of cloud storage method for magnanimity vector slice of data, methods described includes:
Set up all metadata nodes corresponding with distributive catalogue of document system tree;
To be polymerized with the magnanimity vector slice of data under first class catalogue in distributed file system, generated the section of magnanimity vector
Packet;
The magnanimity vector slice of data bag is stored in the metadata node;
Set up for the magnanimity vector slice of data and index, the magnanimity vector slice of data sets up association by index, formed
The data directory of cancellated magnanimity vector slice of data;
The magnanimity vector slice of data index service is provided by the magnanimity vector slice of data bag concordance list.
2. method according to claim 1, methods described includes:
The magnanimity vector slice of data index includes the magnanimity vector slice of data path, title and in magnanimity arrow
Side-play amount in amount slice of data bag;
The magnanimity vector slice of data path includes that first site position, magnanimity vector slice of data line position are put and magnanimity vector
Slice of data column position.
3. method according to claim 1, methods described includes:
The default unitary Data Node of each layer, concordance list is stored in each layer be pre-designed of metadata node;
The magnanimity vector slice of data concordance list stored in the metadata is transmitted to client, magnanimity vector is set up and is cut
The lasting mapping table of sheet data concordance list.
4. method according to claim 1, the magnanimity vector slice of data bag includes file header and at least one record;
The file header includes file type, version number, document keyword, file name, per recording corresponding position described in bar;
Per one vector slice of data of correspondence is recorded described in bar, the record per bar includes length, the bond distance of vector slice of data
Degree, key and value.
5. method according to claim 1, the magnanimity vector slice of data bag is entered using data file sequencing method
Row storage.
6. method according to claim 1, also includes:Carry out adding in the afterbody of the magnanimity vector slice of data bag and deposit
Storage.
7. method according to claim 1, methods described includes:By magnanimity vector slice of data index table cache extremely
Client.
8. method according to claim 4, also includes:The method that magnanimity vector slice of data is read out:
The corresponding metadata of the magnanimity vector slice of data bag is determined by the magnanimity vector slice of data concordance list
Node shortest path;
By it is determined that metadata node in file header in data APMB package, determine the position of the vector slice of data
Put.
9. a kind of cloud storage system for magnanimity vector slice of data, the system includes:
First signal generating unit, for setting up distributive catalogue of document system tree file;
Second signal generating unit, for setting up all metadata nodes corresponding with distributive catalogue of document system tree;
Polymerized unit, for will be polymerized with the magnanimity vector slice of data under first class catalogue based on distributed file system,
Generate magnanimity vector slice of data bag;
Memory element, for the magnanimity vector slice of data bag to be stored in the metadata node;
3rd signal generating unit, for generating the magnanimity vector slice of data concordance list, by concordance list the magnanimity arrow is set up
The network structure of amount slice of data bag, for recording the magnanimity vector slice of data in the magnanimity vector slice of data bag
Path;
Indexing units, for providing the magnanimity vector slice of data index clothes by magnanimity vector slice of data index
Business.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610939884.6A CN106570113B (en) | 2016-10-25 | 2016-10-25 | Mass vector slice data cloud storage method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610939884.6A CN106570113B (en) | 2016-10-25 | 2016-10-25 | Mass vector slice data cloud storage method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106570113A true CN106570113A (en) | 2017-04-19 |
CN106570113B CN106570113B (en) | 2022-04-01 |
Family
ID=58536334
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610939884.6A Active CN106570113B (en) | 2016-10-25 | 2016-10-25 | Mass vector slice data cloud storage method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106570113B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107291889A (en) * | 2017-06-20 | 2017-10-24 | 郑州云海信息技术有限公司 | A kind of date storage method and system |
CN108172277A (en) * | 2017-12-19 | 2018-06-15 | 浙江大学 | A kind of more multiplying power digital slices image storages and the method and system of browsing |
CN109767274A (en) * | 2018-12-05 | 2019-05-17 | 航天信息股份有限公司 | A kind of pair of magnanimity invoice data is associated the method and system of storage |
CN111459882A (en) * | 2020-03-30 | 2020-07-28 | 北京百度网讯科技有限公司 | Namespace transaction processing method and device of distributed file system |
CN111782663A (en) * | 2020-05-21 | 2020-10-16 | 浙江邦盛科技有限公司 | Aggregation index structure and aggregation index method for improving aggregation query efficiency |
CN115373645A (en) * | 2022-10-24 | 2022-11-22 | 济南新语软件科技有限公司 | Complex data packet operation method and system based on dynamic definition |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102332027A (en) * | 2011-10-15 | 2012-01-25 | 西安交通大学 | Mass non-independent small file associated storage method based on Hadoop |
CN102385623A (en) * | 2011-10-25 | 2012-03-21 | 曙光信息产业(北京)有限公司 | Catalogue access method in DFS (distributed file system) |
CN102541985A (en) * | 2011-10-25 | 2012-07-04 | 曙光信息产业(北京)有限公司 | Organization method of client directory cache in distributed file system |
CN103020315A (en) * | 2013-01-10 | 2013-04-03 | 中国人民解放军国防科学技术大学 | Method for storing mass of small files on basis of master-slave distributed file system |
CN103473287A (en) * | 2013-08-30 | 2013-12-25 | 中国科学院信息工程研究所 | Method and system for automatically distributing, running and updating executable programs |
CN103577123A (en) * | 2013-11-12 | 2014-02-12 | 河海大学 | Small file optimization storage method based on HDFS |
US8825652B1 (en) * | 2012-06-28 | 2014-09-02 | Emc Corporation | Small file aggregation in a parallel computing system |
CN104731921A (en) * | 2015-03-26 | 2015-06-24 | 江苏物联网研究发展中心 | Method for storing and processing small log type files in Hadoop distributed file system |
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 |
-
2016
- 2016-10-25 CN CN201610939884.6A patent/CN106570113B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102332027A (en) * | 2011-10-15 | 2012-01-25 | 西安交通大学 | Mass non-independent small file associated storage method based on Hadoop |
CN102385623A (en) * | 2011-10-25 | 2012-03-21 | 曙光信息产业(北京)有限公司 | Catalogue access method in DFS (distributed file system) |
CN102541985A (en) * | 2011-10-25 | 2012-07-04 | 曙光信息产业(北京)有限公司 | Organization method of client directory cache in distributed file system |
US8825652B1 (en) * | 2012-06-28 | 2014-09-02 | Emc Corporation | Small file aggregation in a parallel computing system |
CN103020315A (en) * | 2013-01-10 | 2013-04-03 | 中国人民解放军国防科学技术大学 | Method for storing mass of small files on basis of master-slave distributed file system |
CN103473287A (en) * | 2013-08-30 | 2013-12-25 | 中国科学院信息工程研究所 | Method and system for automatically distributing, running and updating executable programs |
CN103577123A (en) * | 2013-11-12 | 2014-02-12 | 河海大学 | Small file optimization storage method based on HDFS |
CN104731921A (en) * | 2015-03-26 | 2015-06-24 | 江苏物联网研究发展中心 | Method for storing and processing small log type files in Hadoop distributed file system |
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 |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107291889A (en) * | 2017-06-20 | 2017-10-24 | 郑州云海信息技术有限公司 | A kind of date storage method and system |
CN108172277B (en) * | 2017-12-19 | 2020-07-07 | 浙江大学 | Method and system for storing and browsing multiple-magnification digital slice image |
CN108172277A (en) * | 2017-12-19 | 2018-06-15 | 浙江大学 | A kind of more multiplying power digital slices image storages and the method and system of browsing |
CN109767274B (en) * | 2018-12-05 | 2023-04-25 | 航天信息股份有限公司 | Method and system for carrying out associated storage on massive invoice data |
CN109767274A (en) * | 2018-12-05 | 2019-05-17 | 航天信息股份有限公司 | A kind of pair of magnanimity invoice data is associated the method and system of storage |
CN111459882A (en) * | 2020-03-30 | 2020-07-28 | 北京百度网讯科技有限公司 | Namespace transaction processing method and device of distributed file system |
CN111459882B (en) * | 2020-03-30 | 2023-08-29 | 北京百度网讯科技有限公司 | Namespace transaction processing method and device for distributed file system |
CN111782663A (en) * | 2020-05-21 | 2020-10-16 | 浙江邦盛科技有限公司 | Aggregation index structure and aggregation index method for improving aggregation query efficiency |
WO2021232645A1 (en) * | 2020-05-21 | 2021-11-25 | 浙江邦盛科技有限公司 | Aggregation index structure and aggregation index method for improving aggregate query efficiency |
CN111782663B (en) * | 2020-05-21 | 2023-09-01 | 浙江邦盛科技股份有限公司 | Aggregation index structure and aggregation index method for improving aggregation query efficiency |
US11928113B2 (en) | 2020-05-21 | 2024-03-12 | Zhejiang Bangsun Technology Co., Ltd. | Structure and method of aggregation index for improving aggregation query efficiency |
CN115373645A (en) * | 2022-10-24 | 2022-11-22 | 济南新语软件科技有限公司 | Complex data packet operation method and system based on dynamic definition |
CN115373645B (en) * | 2022-10-24 | 2023-02-03 | 济南新语软件科技有限公司 | Complex data packet operation method and system based on dynamic definition |
Also Published As
Publication number | Publication date |
---|---|
CN106570113B (en) | 2022-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106570113A (en) | Cloud storage method and system for mass vector slice data | |
US9558194B1 (en) | Scalable object store | |
US10275489B1 (en) | Binary encoding-based optimizations at datastore accelerators | |
Jiang et al. | THE optimization of HDFS based on small files | |
Padhy et al. | RDBMS to NoSQL: reviewing some next-generation non-relational database’s | |
CN106021266B (en) | Fast multi-tier index supporting dynamic updates | |
US8341130B2 (en) | Scalable file management for a shared file system | |
CN107247778B (en) | System and method for implementing an extensible data storage service | |
US8266136B1 (en) | Mechanism for performing fast directory lookup in a server system | |
Pirzadeh et al. | Performance evaluation of range queries in key value stores | |
Dong et al. | Towards a fast and secure design for enterprise‐oriented cloud storage systems | |
Xu et al. | Drop: Facilitating distributed metadata management in eb-scale storage systems | |
US11080207B2 (en) | Caching framework for big-data engines in the cloud | |
Xu et al. | Efficient and scalable metadata management in EB-scale file systems | |
US10558373B1 (en) | Scalable index store | |
CN102024019A (en) | Suffix tree based catalog organizing method in distributed file system | |
Nguyen et al. | Zing database: high-performance key-value store for large-scale storage service | |
US10592153B1 (en) | Redistributing a data set amongst partitions according to a secondary hashing scheme | |
Qi | S-store: A scalable data store towards permissioned blockchain sharding | |
Gao et al. | An efficient ring-based metadata management policy for large-scale distributed file systems | |
CN110502472A (en) | A kind of the cloud storage optimization method and its system of large amount of small documents | |
US9275091B2 (en) | Database management device and database management method | |
Lu et al. | Hybrid storage architecture and efficient MapReduce processing for unstructured data | |
Orhean et al. | Toward scalable indexing and search on distributed and unstructured data | |
US10628391B1 (en) | Method and system for reducing metadata overhead in a two-tier storage architecture |
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 |