CN108073356A - A kind of data storage, lookup method, device and data handling system - Google Patents
A kind of data storage, lookup method, device and data handling system Download PDFInfo
- Publication number
- CN108073356A CN108073356A CN201611037957.9A CN201611037957A CN108073356A CN 108073356 A CN108073356 A CN 108073356A CN 201611037957 A CN201611037957 A CN 201611037957A CN 108073356 A CN108073356 A CN 108073356A
- Authority
- CN
- China
- Prior art keywords
- data
- stored
- disk
- filtered
- matched
- 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
- G06F16/14—Details of searching files based on file metadata
- G06F16/148—File search processing
-
- 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
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0643—Management of files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0674—Disk device
- G06F3/0676—Magnetic disk device
Abstract
The embodiment of the invention discloses a kind of data storage, lookup method, device and data handling system, storage methods to include:According to default at least one dimension, data to be stored is classified, for every a kind of data to be stored, is divided into N parts, wherein, N is the quantity of identified disk, and N is more than 1;The N parts of data to be stored is stored respectively into N number of disk.It can be seen that, in this programme, of a sort data are stored into N number of disk, realize the scattered storage of similar data, when needing to carry out data search, the searching data in N number of disk, the read or write speed that N number of disk allows is far longer than the read or write speed that a disk allows, therefore, it is possible to reduce the generation of I/O congestion situations.
Description
Technical field
The present invention relates to technical field of data processing, at more particularly to a kind of data storage, lookup method, device and data
Reason system.
Background technology
Existing data storage scheme is in centrally stored to one disk of similar data, for example, by the time mostly
Or data similar in feature are stored into a disk.In this way, when requiring to look up the data of storage, typically in a magnetic
It is searched in disk.
Data search is carried out in a disk, the read or write speed which allows is limited, when the number stored in the disk
During according to measuring very big, there is a situation where I/O (input/output) blocks.When particularly application is to scheme to search diagram technology, such as:Certain
Multiple images and its corresponding attribute information are stored in a disk;Determine the first image (target data) that user uploads,
The second image with first images match is searched in the disk, and the attribute information of the second image found is showed into use
Family.
In said program, if the data volume stored in the disk is very big, searched and first figure in the disk
, it is necessary to which first image is matched with the every image stored in the disk during as matched second image, easily occur
The situation of I/O obstructions.
The content of the invention
The embodiment of the present invention is designed to provide a kind of data storage, lookup method, device and data handling system, with
It realizes the scattered storage of similar data, reduces the generation of I/O congestion situations.
In order to achieve the above objectives, the embodiment of the invention discloses a kind of date storage method, including:
Obtain data to be stored;
According to default at least one dimension, classify to the data to be stored;
For every a kind of data to be stored, N part are divided into, wherein, the N is the quantity of identified disk, institute
N is stated more than 1;
The N parts of data to be stored is stored respectively into N number of disk, wherein, every part of data to be stored is corresponding
Disk is different.
Optionally, the basis presets at least one dimension, and the step of classifying to the data to be stored can wrap
It includes:
According to the time dimension and/or characteristic dimension of the data to be stored, the data to be stored is divided into N classes.
Optionally, described the step of being directed to per a kind of data to be stored, being divided into N parts, can include:
For every a kind of data to be stored, it is averaged and is divided into N parts.
Optionally, described store the N parts of data to be stored to the step in N number of disk respectively includes:
Respectively described N number of disk distributes its corresponding first thread;
For each disk, using its corresponding first thread, its corresponding data to be stored is stored.
Optionally, the data to be stored is image file and its corresponding attribute information.
In order to achieve the above objectives, the embodiment of the invention also discloses a kind of data search method, including:
Determine target data;
The matched data of the target data is searched in identified N number of disk respectively, wherein, the N is more than 1;Institute
The date storage method according to claim 1 of the data in N number of disk is stated to be stored.
Optionally, described the step of searching the matched data of the target data in identified N number of disk respectively, can
To include:
According to default filtering dimension, parameter of the target data under the filtering dimension is determined;
According to the parameter, the data in N number of disk are filtered, obtain filtered data;
The matched data of the target data is searched in the filtered data.
Optionally, it is described according to default filtering dimension in the case where the filtering dimension is time dimension, determine institute
The step of stating parameter of the target data under the filtering dimension can include:
Determine the corresponding time parameter of the target data;
It is described according to the parameter, the data stored in N number of disk are filtered, obtain the step of filtered data
Suddenly can include:
According to the time parameter, determine object time parameter section, by it is being stored in N number of disk, positioned at described
The data in section are determined as filtered data.
Optionally, it is described the target data is searched in the filtered data matched data the step of can wrap
It includes:
For each filtered data, itself and the target data are subjected to similarity calculation, obtain its corresponding calculating
As a result;
According to the corresponding result of calculation of each filtered data, each filtered data is ranked up;
According to ranking results, the matched data of the definite target data in each filtered data.
Optionally, it is described for each filtered data, itself and the target data are subjected to similarity calculation, obtain it
Corresponding result of calculation;According to the corresponding result of calculation of each filtered data, each filtered data is carried out
The step of sequence, can include:
Cache each filtered data;
It is for each filtered data cached, each filtered data is similar to target data progress
Degree calculates, and obtains the corresponding result of calculation of each filtered data;
According to the corresponding result of calculation of each filtered data, using Binary ordering method, each filtering is determined
The position of data in the sequence afterwards;Wherein, the sequence is made of each filtered data, after each filtering in the sequence
Data are ranked up according to its corresponding result of calculation.
Optionally, described the step of searching the matched data of the target data in identified N number of disk respectively, can
To include:
Respectively described N number of disk distributes its corresponding second thread;
For each disk, using corresponding second thread of each disk, searched in each disk described in
The matched data of target data.
Optionally, described the step of searching the matched data of the target data in identified N number of disk respectively, can
To include:
Respectively in N number of disk target data described in parallel search matched data.
Optionally, the target data is image file;The matched data is believed for image file and its corresponding attribute
Breath;
After described the step of searching the matched data of the target data in identified N number of disk respectively, also
It can include:
Matched data to be output is determined in the matched data found;
The attribute information included in the matched data to be output is determined as the corresponding attribute information of the target data.
In order to achieve the above objectives, the embodiment of the invention also discloses a kind of data storage device, including:
Acquisition module, for obtaining data to be stored;
Sort module, for according to default at least one dimension, classifying to the data to be stored;
Division module for being directed to every a kind of data to be stored, is divided into N parts, wherein, the N is identified
The quantity of disk, the N are more than 1;
Memory module, for being stored the N parts of data to be stored respectively into N number of disk, wherein, it waits for every part to deposit
It is different to store up the corresponding disk of data.
Optionally, the sort module, specifically can be used for:
According to the time dimension and/or characteristic dimension of the data to be stored, the data to be stored is divided into N classes.
Optionally, the division module, specifically can be used for:
For every a kind of data to be stored, it is averaged and is divided into N parts.
Optionally, the memory module, specifically can be used for:
Respectively described N number of disk distributes its corresponding first thread;
For each disk, using its corresponding first thread, its corresponding data to be stored is stored.
Optionally, the data to be stored is image file and its corresponding attribute information.
In order to achieve the above objectives, the embodiment of the invention also discloses a kind of data search device, including:
First determining module, for determining target data;
Searching module, for the matched data of the target data to be searched in identified N number of disk respectively, wherein,
The N is more than 1;Data in N number of disk are stored using the data storage device described in claim 14.
Optionally, the searching module can include:
First determination sub-module, for according to default filtering dimension, determining the target data in the filtering dimension
Under parameter;
Filter submodule, for according to the parameter, being filtered to the data in N number of disk, after obtaining filtering
Data;
Submodule is searched, for searching the matched data of the target data in the filtered data.
Optionally, in the case where the filtering dimension is time dimension, first determination sub-module can specifically be used
In:Determine the corresponding time parameter of the target data;
The filter submodule, specifically can be used for:According to the time parameter, object time parameter section is determined, it will
Stored in the N number of disk, data positioned at the section are determined as filtered data.
Optionally, the searching module can include:
Itself and the target data for being directed to each filtered data, are carried out similarity calculation, obtained by computational submodule
To its corresponding result of calculation;
Sorting sub-module, for according to the corresponding result of calculation of each filtered data, after each filtering
Data are ranked up;
Second determination sub-module, for determining the matched data of the target data in each filtered data.
Optionally, the computational submodule, specifically can be used for:
Cache each filtered data;
It is for each filtered data cached, each filtered data is similar to target data progress
Degree calculates, and obtains the corresponding result of calculation of each filtered data;
The sorting sub-module, specifically can be used for:
According to the corresponding result of calculation of each filtered data, using Binary ordering method, each filtering is determined
The position of data in the sequence afterwards;Wherein, the sequence is made of each filtered data, after each filtering in the sequence
Data are ranked up according to its corresponding result of calculation.
Optionally, the searching module, specifically can be used for:
Respectively described N number of disk distributes its corresponding second thread;
For each disk, using corresponding second thread of each disk, searched in each disk described in
The matched data of target data.
Optionally, the searching module, specifically can be used for:
Respectively in N number of disk target data described in parallel search matched data.
Optionally, the target data is image file;The matched data is believed for image file and its corresponding attribute
Breath;Described device can also include:
Second determining module, for determining matched data to be output in the matched data found;
3rd determining module, for the attribute information included in the matched data to be output to be determined as the number of targets
According to corresponding attribute information.
In order to achieve the above objectives, the embodiment of the invention also discloses a kind of data handling system, including:It management equipment and deposits
Node is stored up, wherein,
The management equipment, for obtaining data to be stored;According to default at least one dimension, to the data to be stored
Classify;For every a kind of data to be stored, N parts are divided into, wherein, the N is the number of identified memory node
Amount, the N are more than 1;The N parts of data to be stored is stored respectively into N number of memory node, wherein, every part is to be stored
The corresponding memory node of data is different.
Optionally, the management equipment is additionally operable to determine target data;Respectively institute is searched in N number of memory node
State the matched data of target data.
Using illustrated embodiment of the present invention, according to default at least one dimension, data to be stored is classified, for every
A kind of data to be stored is divided into N parts, wherein, N is the quantity of identified disk, and N is more than 1;Described N parts is waited to deposit
Storage data are stored respectively into N number of disk.It can be seen that in the present solution, storing of a sort data to N number of disk
In, the scattered storage of similar data is realized, when needing to carry out data search, the searching data in N number of disk, N number of disk
The read or write speed of permission is far longer than the read or write speed that a disk allows, therefore, it is possible to reduce the generation of I/O congestion situations.
Certainly, implement any of the products of the present invention or method must be not necessarily required to reach all the above excellent simultaneously
Point.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow diagram of date storage method provided in an embodiment of the present invention;
Fig. 2 is a kind of flow diagram of data search method provided in an embodiment of the present invention;
Fig. 3 is a kind of structure diagram of data storage device provided in an embodiment of the present invention;
Fig. 4 is a kind of structure diagram of data search device provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts
Embodiment belongs to the scope of protection of the invention.
In order to solve the above-mentioned technical problem, an embodiment of the present invention provides a kind of data storage, lookup method, device and numbers
According to processing system, the data storage, lookup method, device can be applied to the various electronic equipments such as computer, tablet computer, tool
Body is not construed as limiting.Date storage method provided in an embodiment of the present invention is described in detail first below.
Fig. 1 is a kind of flow diagram of date storage method provided in an embodiment of the present invention, including:
S101:Obtain data to be stored.
As a kind of embodiment, which can be image file and its corresponding attribute information, alternatively,
Can be other, such as audio file or audio file and its corresponding attribute information etc. do not limit herein.
When the data to be stored is image file and its corresponding attribute information, which can be binary system text
Part or, or the file of extended formatting does not limit herein.Assuming that the image content of the image file is adopted for image
The vehicle that collection equipment collects, then the attribute information can be the brand and type of the car plate of the vehicle, body color or vehicle
Number etc..
The binary file can be the file that the corresponding binary data of image itself is formed;Can also be directed to image into
Row modeling, the binary file is stored as by obtained model value;Characteristics extraction can also be carried out to image, by the spy of extraction
Value indicative is stored as the binary file etc., does not limit herein.
S102:According to default at least one dimension, classify to the data to be stored.
It, can be according to the time dimension of the data to be stored, by the data to be stored point as a kind of embodiment
For N classes.
As an example it is assumed that it is 20 days 8 October in 2016 that a storage time is included in data to be stored:00 data
(the first one's share of expenses for a joint undertaking data), a storage time are 21 days 10 October in 2016:00 data (the second one's share of expenses for a joint undertaking data), a storage
Time is 22 days 12 October in 2016:00 data (the 3rd one's share of expenses for a joint undertaking data) and a storage time are on October 21st, 2016
11:00 data (the 4th one's share of expenses for a joint undertaking data).
It can not can also specifically be limited using one week as period, etc. using one day as a period
It is fixed.It is assumed here that using one day as a period, each part data that storage time is located to the same period are divided into one kind
Data, that is to say, that the second one's share of expenses for a joint undertaking data are same class data with the 4th one's share of expenses for a joint undertaking data, other part of data are respectively a kind of data.
It, can be according to the characteristic dimension of the data to be stored, by the data to be stored point as a kind of embodiment
For N classes.
This feature dimension can be the type of data or other features of data.When the data to be stored is image file
And its during corresponding attribute information, this feature dimension can be the brand of the body color or vehicle included in the attribute information
Deng.
As an example it is assumed that this feature dimension is the type of data, it is TXT that a storage class is included in data to be stored
Data (the first one's share of expenses for a joint undertaking data), a storage class be JPG data (the second one's share of expenses for a joint undertaking data), portion storage class be WMV
Data (the 3rd one's share of expenses for a joint undertaking data) and data (the 4th one's share of expenses for a joint undertaking data) that a storage class is JPG.
That is, the second one's share of expenses for a joint undertaking data are same class data with the 4th one's share of expenses for a joint undertaking data, other part of data are respectively a kind of number
According to.
As a kind of embodiment, can be treated according to the time dimension and characteristic dimension of the data to be stored by described
Storage data are divided into N classes.
As an example it is assumed that this feature dimension is the type of data, it is TXT that a storage class is included in data to be stored
Data (the first one's share of expenses for a joint undertaking data), a storage class be JPG data (the second one's share of expenses for a joint undertaking data), portion storage class be WMV
Data (the 3rd one's share of expenses for a joint undertaking data) and data (the 4th one's share of expenses for a joint undertaking data) that a storage class is JPG, also, the first one's share of expenses for a joint undertaking data
Storage time be 20 days 8 October in 2016:00th, the storage time of the second one's share of expenses for a joint undertaking data is 21 days 10 October in 2016:00th,
The storage time of three one's share of expenses for a joint undertaking data is 22 days 12 October in 2016:00th, the 4th one's share of expenses for a joint undertaking time data memory is October 21 in 2016
Day 11:00.
In this case, storage time is located at the same period and the identical each part data of type is divided into a kind of number
According to, that is to say, that the second one's share of expenses for a joint undertaking data are same class data with the 4th one's share of expenses for a joint undertaking data, other part of data are respectively a kind of data.
S103:For every a kind of data to be stored, N parts are divided into.Wherein, the N is the number of identified disk
Amount, the N are more than 1.
In the present solution, to realize the scattered storage of similar data, that is, similar data are stored into multiple disks,
Firstly the need of the quantity of definite available disk, according to the quantity, similar data are divided.Here, per a kind of number to be stored
According to as similar data.
As a kind of embodiment, it can be directed to per a kind of data to be stored, be averaged and be divided into N parts.Alternatively, also may be used
Using random division as N parts, do not limit specifically.
It should be noted that " disk " in the present embodiment is it can be appreciated that each memory node in storage cluster.
S104:The N parts of data to be stored is stored respectively into N number of disk.Wherein, every part of data to be stored pair
The disk answered is different.
N is more than 1, it is assumed that N 4, this 4 disks are respectively disk A, disk B, disk C and disk D, by data to be stored
It is divided into 4 parts, it is assumed that this 4 parts of data to be stored are respectively A1, B1, C1 and D1, A1, B1, C1 and D1 are stored respectively to this 4
In disk.Specifically which disk is which part data to be stored is stored to and does not limit, describes for convenience, it is assumed here that by A1
It stores to disk A, B1 is stored to disk B, C1 is stored to disk C, D1 is stored to disk D.
Can be respectively that N number of disk distributes its corresponding first thread as a kind of embodiment;For each magnetic
Using its corresponding first thread, its corresponding data to be stored is stored for disk.
Continue above-mentioned example, it is assumed that there are 8 threads to be used for this 4 disk storage data, then by this 8 thread distribution
Give this 4 disks, it is assumed here that 2 threads are corresponded to for mean allocation, that is, each disk, it is of course also possible to which injustice is divided equally
Match somebody with somebody, do not limit specifically.Being assumed to be 2 threads that disk A is distributed is:Thread 1 and thread 2, it is assumed that for 2 of disk B distribution
Thread is:Thread 3 and thread 4, it is assumed that be that 2 threads that disk C is distributed are:Thread 5 and thread 6, it is assumed that for disk D distribution
2 threads are:Thread 7 and thread 8.
Then disk A stores A1 using thread 1 and thread 2, and disk B deposits B1 using thread 3 and thread 4
Storage, disk C store C1 using thread 5 and thread 6, and disk D stores D1 using thread 7 and thread 8.Specifically
, when disk stores data using multithreading, Handler objects can be called.
It is if right it should be noted that using the embodiment classified according to time dimension to data to be stored
When data are stored, the corresponding time dimension of the data can be stored together;If using according to feature dimensions
Spend the embodiment classified to data to be stored, then it, can be by the corresponding spy of the data when being stored to data
Sign dimension is stored together;If using the implementation classified according to time dimension and characteristic dimension to data to be stored
Mode when then being stored to data, can deposit the corresponding time dimension of the data and characteristic dimension together
Storage;In this way, when follow-up data is searched, data can be filtered for dimensional information, improve search efficiency.
Using embodiment illustrated in fig. 1 of the present invention, according to default at least one dimension, data to be stored is classified, pin
To every a kind of data to be stored, N parts are divided into, wherein, N is the quantity of identified disk, and N is more than 1;It will be N parts described
Data to be stored is stored respectively into N number of disk.It can be seen that in the present solution, storing of a sort data to N number of
In disk, the scattered storage of similar data is realized, when needing to carry out data search, the searching data in N number of disk is N number of
The read or write speed that disk allows is far longer than the read or write speed that a disk allows, therefore, it is possible to reduce the hair of I/O congestion situations
It is raw.
The embodiment of the present invention also provides a kind of data search method, as shown in Fig. 2, including:
S201:Determine target data.
The target data can be the image file that user uploads or or other data, such as voice data
Deng not limiting herein.The embodiment of the present invention carries out data search, as searches the matched data of target data.
S202:The matched data of the target data is searched in identified N number of disk respectively, wherein, the N is more than
1;The data date storage method that embodiment illustrated in fig. 1 provides according to the present invention in N number of disk is stored.
The date storage method that embodiment illustrated in fig. 1 provides according to the present invention distinguishes the same class data of target data
It stores into N number of disk, therefore, when carrying out data search, it is also desirable to the coupling number of target data is searched in this N number of disk
According to.
Specifically, can parameter of the target data under the filtering dimension be determined according to default filtering dimension;
According to the parameter, the data in N number of disk are filtered, obtain filtered data;In the filtered data
Search the matched data of the target data.
Corresponding with embodiment illustrated in fig. 1, which can be time dimension and/or characteristic dimension.
If the corresponding time dimension of data stored together in embodiment illustrated in fig. 1, the filtering dimension
It can be time dimension.
As an example it is assumed that target data is the image file that user uploads, the data stored in disk are image file
And its corresponding attribute information.Time parameter of the target data under time dimension, it is understood that for the target data
Creation time or acquisition time etc., it is assumed here that the time parameter is 21 days 10 October in 2016:00.
According to the time parameter, the data in 4 disks are filtered, specifically, can be according to the time parameter, really
Set the goal time parameter section, it is understood that determine the period where the time parameter.It can be using one day as one
Time parameter section can not also specifically limit, using one week as time parameter section, etc. moreover, shown in Fig. 2
" time parameter section " in embodiment may be the same or different with " period " in embodiment illustrated in fig. 1.Here it is false
If using one day as a time parameter section, the object time parameter section where the time parameter is October 21 in 2016
Day.The data that in this 4 disks, time dimension is located on October 21st, 2016 are determined as filtered data.Only filtering
The matched data of target data is searched in data afterwards, improves search efficiency.
If the corresponding characteristic dimension of data stored together in embodiment illustrated in fig. 1, the filtering dimension
Dimension can be characterized.
As an example it is assumed that target data is the image file that user uploads, the data stored in disk are image file
And its corresponding attribute information.Characteristic parameter of the target data under characteristic dimension, it is assumed here that be the class of the target data
Type, it is assumed here that the parameter is JPG.
According to this feature parameter, the data in 4 disks are filtered, that is, are by this 4 disks, type
The data of JPG are determined as filtered data.The matched data of target data is only searched in data after filtration, improves lookup effect
Rate.
Characteristic dimension or other features, do not limit specifically.
If the corresponding time dimension of data and characteristic dimension stored together in embodiment illustrated in fig. 1,
Then the filtering dimension can be time dimension and characteristic dimension.
As an example it is assumed that target data is the image file that user uploads, the data stored in disk are image file
And its corresponding attribute information.Time parameter of the target data under time dimension, it is understood that for the target data
Creation time or acquisition time etc., it is assumed here that the time parameter is 21 days 10 October in 2016:00.The target data exists
Characteristic parameter under characteristic dimension, it is assumed here that be the type of the target data, it is assumed here that this feature parameter is JPG.
According to the time parameter, the data in 4 disks are filtered, specifically, can be according to the time parameter, really
Set the goal time parameter section, it is understood that determine the period where the time parameter.It can be using one day as one
Time parameter section can not also specifically limit, using one week as time parameter section, etc. moreover, shown in Fig. 2
" time parameter section " in embodiment may be the same or different with " period " in embodiment illustrated in fig. 1.Here it is false
If using one day as a time parameter section, the object time parameter section where the time parameter is October 21 in 2016
Day.It is counted after the data that in this 4 disks, time dimension is located on October 21st, 2016 and type is JPG are determined as filtering
According to.The matched data of target data is only searched in data after filtration, improves search efficiency.
During being filtered to the data in disk, can Pattern be generated according to the parameter under filtering dimension
(model), and file filter is carried out by canonical mode, in this way, search efficiency can be further improved.It specifically, can be advance
According to filtering dimension generation configuration file, when being filtered to the data in disk, according to the filtering included in the configuration file
Dimension carries out parameter transmission, generation Pattern (model).As an example it is assumed that when the filtering dimension included in configuration file is
Between dimension, determine parameter of the target data under time dimension, which be transferred in the program set, the Program Generating
Pattern.In addition, being stored with the corresponding model data of image file in disk, the model data is carried out according to the Pattern
Filtering, just realizes the filtering to image file.
As a kind of embodiment, the matched data of the target data is searched in the filtered data, can be wrapped
It includes:For each filtered data, itself and the target data are subjected to similarity calculation, obtain its corresponding result of calculation;
According to the corresponding result of calculation of each filtered data, each filtered data is ranked up;It is tied according to sequence
Fruit determines the matched data of the target data in each filtered data.
In the present embodiment, filtered data and target data are subjected to similarity calculation and according to result of calculation pair
The detailed process that filtered data is ranked up can include:
Cache each filtered data;
It is for each filtered data cached, each filtered data is similar to target data progress
Degree calculates, and obtains the corresponding result of calculation of each filtered data;
According to the corresponding result of calculation of each filtered data, using Binary ordering method, each filtering is determined
The position of data in the sequence afterwards;Wherein, the sequence is made of each filtered data, after each filtering in the sequence
Data are ranked up according to its corresponding result of calculation.
In the present embodiment, multiple chained lists can be created, filtered data and its corresponding calculating are preserved in chained list
As a result.As an example it is assumed that above-mentioned similarity calculation result has 10 grades:0-10, grade 0 represent filtered data and target
Data differ completely, and grade 10 represents that filtered data is identical with target data.10 chained lists can be then created, each
The result of calculation of chained list one grade of preservation and its corresponding filtered data.In this way, when carrying out data search using multithreading,
Each thread searching data in different chained lists, compared to each thread, searching data can reduce performance in a chained list
Loss.Alternatively, other forms can also be used to preserve filtered data and its corresponding result of calculation, such as array etc., herein
It does not limit.
Specifically, using Binary ordering method, the position of each filtered data in the sequence is determined, can incite somebody to action first each
The corresponding similarity calculation result of filtered data meter corresponding with the filtered data on the head of existing sequence and afterbody respectively
Result is calculated to be compared respectively.
If the sequence with result of calculation it is descending be ranked sequentially each filtered data, the mistake on the head of sequence
The corresponding result of calculation of data is maximum after filter, and the corresponding result of calculation of filtered data of the afterbody of sequence is minimum.Alternatively, the sequence
Row with result of calculation it is ascending be ranked sequentially each filtered data, then the corresponding meter of the filtered data on the head of sequence
Result minimum is calculated, the corresponding result of calculation of filtered data of the afterbody of sequence is maximum.Here, suppose that the sequence is with result of calculation
Descending is ranked sequentially each filtered data.
If the corresponding similarity calculation result of filtered data is more than the corresponding meter of filtered data on the head of sequence
It calculates as a result, the filtered data then can be directly arranged in the head of the sequence;
If the corresponding similarity calculation result of filtered data is less than the corresponding meter of filtered data on the head of sequence
It calculates as a result, the filtered data then can be directly arranged in the afterbody of the sequence;
If filtered data of the corresponding similarity calculation result of filtered data both no more than the head of sequence corresponds to
Result of calculation, the corresponding result of calculation of filtered data nor less than the head of sequence, then by itself and sequence centre position
The corresponding result of calculation of filtered data be compared etc., until determining the position of the filtered data in the sequence.
If target data is the image file that user uploads, the data stored in disk are image file and its corresponding
Attribute information when then carrying out data search, can be modeled target data, in addition, also correspondence stores image in disk
The model data of file, the data obtained after being modeled to target data and the model data stored in disk carry out similarity meter
It calculates, according to result of calculation, determines the matched data of target data.
It should be noted that after being ranked up to each filtered data, it can be according to ranking results, in each filtering
The matched data of target data is determined in data afterwards.If according to the descending order of similarity calculation result to each filtering
Data are ranked up afterwards, then the default number of data for coming front can be determined as to the matched data of target data.It can be with
Identified matched data is exported, that is, shows user.
Target data be user upload image file, matched data be matched image file and its corresponding attribute
In the case of information, the attribute information included in the whole matched datas found can be determined as the corresponding category of target data
Property information, matched data to be output can also be determined in the matched data found, will only be wrapped in matched data to be output
The attribute information contained is determined as the corresponding attribute information of target data.Only the corresponding attribute information of target data can be showed
The attribute information and its corresponding image file can also be showed user together, do not limited specifically by user.
Can be respectively that N number of disk distributes its corresponding second thread as a kind of embodiment;For each magnetic
Disk using corresponding second thread of each disk, searches the matched data of the target data in each disk.
In the present embodiment, the thread applied when storing data is known as first thread, the line that by searching data when is applied
Journey is known as the second thread.
Continue above-mentioned example, N 4, this 4 disks are respectively disk A, disk B, disk C and disk D, target data
Same class data are respectively A1, B1, C1 and D1, and A1 is stored in disk A, and B1 is stored in disk B, and C1 is stored in disk C,
D1 is stored in disk D.
Assuming that there are 12 threads to be used for this 4 disk searching datas, then this 12 threads are distributed into this 4 magnetic
Disk, it is assumed here that 3 threads are corresponded to for mean allocation, that is, each disk, it is of course also possible to unequal distribution, specifically not
It limits.Being assumed to be 3 threads that disk A is distributed is:Thread 11, thread 12 and thread 13, it is assumed that for 3 of disk B distribution
Thread is:Thread 14, thread 15 and thread 16, it is assumed that be that 3 threads that disk C is distributed are:Thread 17, thread 18 and thread
19, it is assumed that be that 3 threads that disk D is distributed are:Thread 20, thread 21 and thread 22.
Then disk A carries out data search using thread 11, thread 12 and thread 13, and disk B utilizes thread 14,15 and of thread
Thread 16 carries out data search, and disk C carries out data search using thread 17, thread 18 and thread 19, and disk D utilizes thread
20th, thread 21 and thread 22 carry out data search.
Can be simultaneously it should be noted that when searching the matched data of target data in identified N number of disk respectively
Row is searched, in this way, search efficiency can be improved largely.It is of course also possible to serially search, serial search can also reduce I/
The generation of O congestion situations.
Using embodiment illustrated in fig. 2 of the present invention, of a sort data are stored into N number of disk, when progress data search
When, the searching data in N number of disk, the read or write speed that N number of disk allows is far longer than the read or write speed that a disk allows, because
This, can reduce the generation of I/O congestion situations.
Corresponding with above method embodiment, the embodiment of the present invention also provides a kind of data storage, searches device.
Fig. 3 is a kind of structure diagram of data storage device provided in an embodiment of the present invention, including:
Acquisition module 301, for obtaining data to be stored;
Sort module 302, for according to default at least one dimension, classifying to the data to be stored;
Division module 303 for being directed to every a kind of data to be stored, is divided into N parts, wherein, the N is is determined
Disk quantity, the N be more than 1;
Memory module 304, for being stored the N parts of data to be stored respectively into N number of disk, wherein, every part
The corresponding disk of data to be stored is different.
In the present embodiment, sort module 302 specifically can be used for:
According to the time dimension and/or characteristic dimension of the data to be stored, the data to be stored is divided into N classes.
In the present embodiment, division module 303 specifically can be used for:
For every a kind of data to be stored, it is averaged and is divided into N parts.
In the present embodiment, memory module 304 specifically can be used for:
Respectively described N number of disk distributes its corresponding first thread;
For each disk, using its corresponding first thread, its corresponding data to be stored is stored.
In the present embodiment, the data to be stored is image file and its corresponding attribute information.
Using embodiment illustrated in fig. 3 of the present invention, according to default at least one dimension, data to be stored is classified, pin
To every a kind of data to be stored, N parts are divided into, wherein, N is the quantity of identified disk, and N is more than 1;It will be N parts described
Data to be stored is stored respectively into N number of disk.It can be seen that in the present solution, storing of a sort data to N number of
In disk, the scattered storage of similar data is realized, when needing to carry out data search, the searching data in N number of disk is N number of
The read or write speed that disk allows is far longer than the read or write speed that a disk allows, therefore, it is possible to reduce the hair of I/O congestion situations
It is raw.
Fig. 4 is a kind of structure diagram of data search device provided in an embodiment of the present invention, including:
First determining module 401, for determining target data;
Searching module 402, for the matched data of the target data to be searched in identified N number of disk respectively,
In, the N is more than 1;Data in N number of disk using embodiment illustrated in fig. 3 of the present invention offer data storage device into
Row storage.
In the present embodiment, searching module 402 can include:First determination sub-module, filter submodule and lookup submodule
Block (not shown), wherein,
First determination sub-module, for according to default filtering dimension, determining the target data in the filtering dimension
Under parameter;
Filter submodule, for according to the parameter, being filtered to the data in N number of disk, after obtaining filtering
Data;
Submodule is searched, for searching the matched data of the target data in the filtered data.
In the present embodiment, in the case where the filtering dimension is time dimension, first determination sub-module, specifically
It can be used for:Determine the corresponding time parameter of the target data;
The filter submodule, specifically can be used for:According to the time parameter, object time parameter section is determined, it will
Stored in the N number of disk, data positioned at the section are determined as filtered data.
In the present embodiment, searching module 402 can include:Computational submodule, sorting sub-module and second determine submodule
Block (not shown), wherein,
Itself and the target data for being directed to each filtered data, are carried out similarity calculation, obtained by computational submodule
To its corresponding result of calculation;
Sorting sub-module, for according to the corresponding result of calculation of each filtered data, after each filtering
Data are ranked up;
Second determination sub-module, for determining the matched data of the target data in each filtered data.
In the present embodiment, the computational submodule, specifically can be used for:
Cache each filtered data;
It is for each filtered data cached, each filtered data is similar to target data progress
Degree calculates, and obtains the corresponding result of calculation of each filtered data;
The sorting sub-module, specifically can be used for:
According to the corresponding result of calculation of each filtered data, using Binary ordering method, each filtering is determined
The position of data in the sequence afterwards;Wherein, the sequence is made of each filtered data, after each filtering in the sequence
Data are ranked up according to its corresponding result of calculation.
In the present embodiment, searching module 402 specifically can be used for:
Respectively described N number of disk distributes its corresponding second thread;
For each disk, using corresponding second thread of each disk, searched in each disk described in
The matched data of target data.
In the present embodiment, searching module 402 specifically can be used for:
Respectively in N number of disk target data described in parallel search matched data.
In the present embodiment, the target data is image file;The matched data is image file and its corresponding
Attribute information;Described device can also include:Second determining module and the 3rd determining module (not shown), wherein,
Second determining module, for determining matched data to be output in the matched data found;
3rd determining module, for the attribute information included in the matched data to be output to be determined as the number of targets
According to corresponding attribute information.
Using embodiment illustrated in fig. 4 of the present invention, of a sort data are stored into N number of disk, when progress data search
When, the searching data in N number of disk, the read or write speed that N number of disk allows is far longer than the read or write speed that a disk allows, because
This, can reduce the generation of I/O congestion situations.
The embodiment of the present invention also provides a kind of data handling system, including:Management equipment and memory node, wherein,
The management equipment can be used for obtaining data to be stored;According to default at least one dimension, to described to be stored
Data are classified;For every a kind of data to be stored, N parts are divided into, wherein, the N is identified memory node
Quantity, the N be more than 1;The N parts of data to be stored is stored respectively into N number of memory node, wherein, every part is treated
It is different to store the corresponding memory node of data.
The above-mentioned process that data are stored for management equipment, in this process, management equipment can be according to the number to be stored
According to time dimension and/or characteristic dimension, the data to be stored is divided into N classes.
In this process, management equipment can be directed to per a kind of data to be stored, be averaged and be divided into N parts.
In this process, management equipment can be respectively that N number of memory node distributes its corresponding first thread;
For each memory node, using its corresponding first thread, its corresponding data to be stored is stored.
In this process, above-mentioned data to be stored is image file and its corresponding attribute information.
In the present embodiment, the management equipment can be also used for determining target data;Respectively in N number of storage section
The matched data of the target data is searched in point.
The above-mentioned process for management equipment searching data, in this process, management equipment can be tieed up according to default filtering
Degree determines parameter of the target data under the filtering dimension;According to the parameter, in N number of memory node
Data are filtered, and obtain filtered data;Respectively the target is searched in the filtered data in N number of memory node
The matched data of data.
In this process, in the case where filtering dimension is time dimension, management equipment can determine the target data
Corresponding time parameter;According to the time parameter, object time parameter section is determined, will be stored in N number of memory node
, the data positioned at the section be determined as filtered data.
In this process, management equipment can be directed to each filtered data, it is similar to target data progress
Degree calculates, and obtains its corresponding result of calculation;According to the corresponding result of calculation of each filtered data, to each mistake
Data are ranked up after filter;According to ranking results, the coupling number of the definite target data in each filtered data
According to.
In this process, management equipment can cache each filtered data;For each filtered data cached,
Each filtered data and the target data are subjected to similarity calculation, corresponding calculate of each filtered data is obtained and ties
Fruit;According to the corresponding result of calculation of each filtered data, using Binary ordering method, each filtered data is determined in the sequence
Position;Wherein, the sequence is made of each filtered data, and each filtered data in the sequence is corresponded to according to it
Result of calculation be ranked up.
In this process, management equipment can be respectively that N number of memory node distributes its corresponding second thread;For
Each memory node, using corresponding second thread of the memory node, that the target data is searched in the memory node
With data.
In this process, management equipment can the target data described in parallel search in N number of memory node respectively
Matched data.
In this process, above-mentioned target data is image file;The matched data is image file and its corresponding category
Property information;Management equipment can determine matched data to be output in the matched data found;By the matching to be output
The attribute information included in data is determined as the corresponding attribute information of the target data.
Using system embodiment shown in the present invention, according to default at least one dimension, data to be stored is classified, pin
To every a kind of data to be stored, N parts are divided into, wherein, N is the quantity of identified memory node, and N is more than 1;By described in
N parts of data to be stored are stored respectively into N number of memory node.It can be seen that in the present solution, of a sort data are deposited
Storage realizes the scattered storage of similar data, when needing to carry out data search, in N number of memory node into N number of memory node
Middle searching data, the read or write speed that N number of memory node allows are far longer than the read or write speed that a memory node allows, therefore,
The generation of I/O congestion situations can be reduced.
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any this actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to
Non-exclusive inclusion, so that process, method, article or equipment including a series of elements not only will including those
Element, but also including other elements that are not explicitly listed or further include as this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
Also there are other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is described using relevant mode, identical similar portion between each embodiment
Point just to refer each other, and the highlights of each of the examples are difference from other examples.It is real especially for device
For applying example, since it is substantially similar to embodiment of the method, so description is fairly simple, related part is referring to embodiment of the method
Part explanation.
One of ordinary skill in the art will appreciate that all or part of step in realization above method embodiment is can
Relevant hardware to be instructed to complete by program, the program can be stored in computer read/write memory medium,
The storage medium designated herein obtained, such as:ROM/RAM, magnetic disc, CD etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modifications, equivalent replacements and improvements are made within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (28)
1. a kind of date storage method, which is characterized in that including:
Obtain data to be stored;
According to default at least one dimension, classify to the data to be stored;
For every a kind of data to be stored, N parts are divided into, wherein, the N is the quantity of identified disk, and the N is big
In 1;
The N parts of data to be stored is stored respectively into N number of disk, wherein, the corresponding disk of every part of data to be stored
It is different.
2. according to the method described in claim 1, it is characterized in that, the basis presets at least one dimension, wait to deposit to described
The step of storage data are classified includes:
According to the time dimension and/or characteristic dimension of the data to be stored, the data to be stored is divided into N classes.
3. according to the method described in claim 1, it is characterized in that, described be directed to per a kind of data to be stored, N is divided into
The step of part, includes:
For every a kind of data to be stored, it is averaged and is divided into N parts.
4. according to the method described in claim 1, it is characterized in that, described store the N parts of data to be stored respectively to institute
The step stated in N number of disk includes:
Respectively described N number of disk distributes its corresponding first thread;
For each disk, using its corresponding first thread, its corresponding data to be stored is stored.
5. according to claim 1-4 any one of them methods, which is characterized in that the data to be stored for image file and its
Corresponding attribute information.
6. a kind of data search method, which is characterized in that including:
Determine target data;
The matched data of the target data is searched in identified N number of disk respectively, wherein, the N is more than 1;It is described N number of
Data date storage method according to claim 1 in disk is stored.
It is 7. according to the method described in claim 6, it is characterized in that, described respectively in identified N number of disk described in lookup
The step of matched data of target data, includes:
According to default filtering dimension, parameter of the target data under the filtering dimension is determined;
According to the parameter, the data in N number of disk are filtered, obtain filtered data;
The matched data of the target data is searched in the filtered data.
8. the method according to the description of claim 7 is characterized in that in the filtering dimension in the case of time dimension, institute
State according to default filtering dimension, determine the target data it is described filtering dimension under parameter the step of include:
Determine the corresponding time parameter of the target data;
It is described according to the parameter, the step of being filtered to the data stored in N number of disk, obtain filtered data, wraps
It includes:
According to the time parameter, determine object time parameter section, by it is being stored in N number of disk, positioned at the section
Data be determined as filtered data.
9. the method according to the description of claim 7 is characterized in that described search the number of targets in the filtered data
According to matched data the step of include:
For each filtered data, itself and the target data are subjected to similarity calculation, obtain its corresponding result of calculation;
According to the corresponding result of calculation of each filtered data, each filtered data is ranked up;
According to ranking results, the matched data of the definite target data in each filtered data.
10. according to the method described in claim 9, it is characterized in that, described be directed to each filtered data, by itself and the mesh
It marks data and carries out similarity calculation, obtain its corresponding result of calculation;Knot is calculated according to each filtered data is corresponding
Fruit, the step of being ranked up to each filtered data, include:
Cache each filtered data;
For each filtered data cached, each filtered data is subjected to similarity meter with the target data
It calculates, obtains the corresponding result of calculation of each filtered data;
According to the corresponding result of calculation of each filtered data, using Binary ordering method, number after each filtering is determined
According to position in the sequence;Wherein, the sequence is made of each filtered data, each filtered data in the sequence
It is ranked up according to its corresponding result of calculation.
It is 11. according to the method described in claim 6, it is characterized in that, described respectively in identified N number of disk described in lookup
The step of matched data of target data, includes:
Respectively described N number of disk distributes its corresponding second thread;
For each disk, using corresponding second thread of each disk, the target is searched in each disk
The matched data of data.
It is 12. according to the method described in claim 6, it is characterized in that, described respectively in identified N number of disk described in lookup
The step of matched data of target data, includes:
Respectively in N number of disk target data described in parallel search matched data.
13. according to claim 6-12 any one of them methods, which is characterized in that the target data is image file;Institute
Matched data is stated as image file and its corresponding attribute information;
After described the step of searching the matched data of the target data in identified N number of disk respectively, further include:
Matched data to be output is determined in the matched data found;
The attribute information included in the matched data to be output is determined as the corresponding attribute information of the target data.
14. a kind of data storage device, which is characterized in that including:
Acquisition module, for obtaining data to be stored;
Sort module, for according to default at least one dimension, classifying to the data to be stored;
Division module for being directed to every a kind of data to be stored, is divided into N parts, wherein, the N is identified disk
Quantity, the N be more than 1;
Memory module, for being stored the N parts of data to be stored respectively into N number of disk, wherein, every part of number to be stored
It is different according to corresponding disk.
15. device according to claim 14, which is characterized in that the sort module is specifically used for:
According to the time dimension and/or characteristic dimension of the data to be stored, the data to be stored is divided into N classes.
16. device according to claim 14, which is characterized in that the division module is specifically used for:
For every a kind of data to be stored, it is averaged and is divided into N parts.
17. device according to claim 14, which is characterized in that the memory module is specifically used for:
Respectively described N number of disk distributes its corresponding first thread;
For each disk, using its corresponding first thread, its corresponding data to be stored is stored.
18. according to claim 14-17 any one of them devices, which is characterized in that the data to be stored is image file
And its corresponding attribute information.
19. a kind of data search device, which is characterized in that including:
First determining module, for determining target data;
Searching module, for the matched data of the target data to be searched in identified N number of disk respectively, wherein, the N
More than 1;Data in N number of disk are stored using the data storage device described in claim 14.
20. device according to claim 19, which is characterized in that the searching module, including:
First determination sub-module, for according to default filtering dimension, determining the target data under the filtering dimension
Parameter;
Filter submodule, for according to the parameter, being filtered to the data in N number of disk, obtaining filtered data;
Submodule is searched, for searching the matched data of the target data in the filtered data.
21. device according to claim 20, which is characterized in that in the case where the filtering dimension is time dimension,
First determination sub-module, is specifically used for:Determine the corresponding time parameter of the target data;
The filter submodule, is specifically used for:According to the time parameter, object time parameter section is determined, by N number of magnetic
Data being stored in disk, positioned at the section are determined as filtered data.
22. device according to claim 20, which is characterized in that the searching module, including:
Itself and the target data for being directed to each filtered data, are carried out similarity calculation, obtain it by computational submodule
Corresponding result of calculation;
Sorting sub-module, for according to the corresponding result of calculation of each filtered data, to each filtered data
It is ranked up;
Second determination sub-module, for determining the matched data of the target data in each filtered data.
23. device according to claim 22, which is characterized in that the computational submodule is specifically used for:
Cache each filtered data;
For each filtered data cached, each filtered data is subjected to similarity meter with the target data
It calculates, obtains the corresponding result of calculation of each filtered data;
The sorting sub-module, is specifically used for:
According to the corresponding result of calculation of each filtered data, using Binary ordering method, number after each filtering is determined
According to position in the sequence;Wherein, the sequence is made of each filtered data, each filtered data in the sequence
It is ranked up according to its corresponding result of calculation.
24. device according to claim 19, which is characterized in that the searching module is specifically used for:
Respectively described N number of disk distributes its corresponding second thread;
For each disk, using corresponding second thread of each disk, the target is searched in each disk
The matched data of data.
25. device according to claim 19, which is characterized in that the searching module is specifically used for:
Respectively in N number of disk target data described in parallel search matched data.
26. according to claim 19-25 any one of them devices, which is characterized in that the target data is image file;Institute
Matched data is stated as image file and its corresponding attribute information;Described device further includes:
Second determining module, for determining matched data to be output in the matched data found;
3rd determining module, for the attribute information included in the matched data to be output to be determined as the target data pair
The attribute information answered.
27. a kind of data handling system, which is characterized in that including:Management equipment and memory node, wherein,
The management equipment, for obtaining data to be stored;According to default at least one dimension, the data to be stored is carried out
Classification;For every a kind of data to be stored, N part are divided into, wherein, the N is the quantity of identified memory node, institute
N is stated more than 1;The N parts of data to be stored is stored respectively into N number of memory node, wherein, every part of data to be stored pair
The memory node answered is different.
28. system according to claim 27, which is characterized in that
The management equipment is additionally operable to determine target data;Respectively the target data is searched in N number of memory node
Matched data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611037957.9A CN108073356B (en) | 2016-11-10 | 2016-11-10 | Data storage and search method, device and data processing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611037957.9A CN108073356B (en) | 2016-11-10 | 2016-11-10 | Data storage and search method, device and data processing system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108073356A true CN108073356A (en) | 2018-05-25 |
CN108073356B CN108073356B (en) | 2021-07-20 |
Family
ID=62161288
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611037957.9A Active CN108073356B (en) | 2016-11-10 | 2016-11-10 | Data storage and search method, device and data processing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108073356B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109271545A (en) * | 2018-08-02 | 2019-01-25 | 深圳市商汤科技有限公司 | A kind of characteristic key method and device, storage medium and computer equipment |
CN109783678A (en) * | 2018-12-29 | 2019-05-21 | 深圳云天励飞技术有限公司 | A kind of method and device of picture search |
CN114138469A (en) * | 2021-11-16 | 2022-03-04 | 苏州浪潮智能科技有限公司 | Thread self-adaptive allocation method and device for solid state disk and solid state disk |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101030222A (en) * | 2007-03-22 | 2007-09-05 | 华为技术有限公司 | Model searching device and method |
CN101038668A (en) * | 2006-03-18 | 2007-09-19 | 辽宁师范大学 | Image searching method based on important bit-plane |
US20140156613A1 (en) * | 2012-09-18 | 2014-06-05 | Squash Compression, LLC | Methods and Apparatus for Increasing the Efficiency of Electronic Data Storage and Transmission |
CN105868218A (en) * | 2015-01-23 | 2016-08-17 | ***通信集团河北有限公司 | Data processing method and electronic device |
CN106055706A (en) * | 2016-06-23 | 2016-10-26 | 杭州迪普科技有限公司 | Cache resource storage method and device |
-
2016
- 2016-11-10 CN CN201611037957.9A patent/CN108073356B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101038668A (en) * | 2006-03-18 | 2007-09-19 | 辽宁师范大学 | Image searching method based on important bit-plane |
CN101030222A (en) * | 2007-03-22 | 2007-09-05 | 华为技术有限公司 | Model searching device and method |
US20140156613A1 (en) * | 2012-09-18 | 2014-06-05 | Squash Compression, LLC | Methods and Apparatus for Increasing the Efficiency of Electronic Data Storage and Transmission |
CN105868218A (en) * | 2015-01-23 | 2016-08-17 | ***通信集团河北有限公司 | Data processing method and electronic device |
CN106055706A (en) * | 2016-06-23 | 2016-10-26 | 杭州迪普科技有限公司 | Cache resource storage method and device |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109271545A (en) * | 2018-08-02 | 2019-01-25 | 深圳市商汤科技有限公司 | A kind of characteristic key method and device, storage medium and computer equipment |
CN109271545B (en) * | 2018-08-02 | 2022-06-03 | 深圳市商汤科技有限公司 | Feature retrieval method and device, storage medium and computer equipment |
CN109783678A (en) * | 2018-12-29 | 2019-05-21 | 深圳云天励飞技术有限公司 | A kind of method and device of picture search |
CN109783678B (en) * | 2018-12-29 | 2021-07-20 | 深圳云天励飞技术有限公司 | Image searching method and device |
CN114138469A (en) * | 2021-11-16 | 2022-03-04 | 苏州浪潮智能科技有限公司 | Thread self-adaptive allocation method and device for solid state disk and solid state disk |
CN114138469B (en) * | 2021-11-16 | 2023-11-14 | 苏州浪潮智能科技有限公司 | Thread self-adaptive distribution method and device for solid state disk and solid state disk |
Also Published As
Publication number | Publication date |
---|---|
CN108073356B (en) | 2021-07-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106815263B (en) | The searching method and device of legal provision | |
KR101648204B1 (en) | Generating metadata for association with a collection of content items | |
CN109189991A (en) | Repeat video frequency identifying method, device, terminal and computer readable storage medium | |
CN105612516B (en) | Attribute redundancy removal | |
KR100380200B1 (en) | Multi-linearization data structure for image browsing | |
CN105917618A (en) | Data transfer monitoring system, data transfer monitoring method, and site system | |
CN106126558A (en) | A kind of public sentiment monitoring method and device | |
CN106294595A (en) | A kind of document storage, search method and device | |
CN108073356A (en) | A kind of data storage, lookup method, device and data handling system | |
EP1849099A2 (en) | Recommender system for identifying a new set of media items responsive to an input set of media items and knowledge base metrics | |
JP2013536491A (en) | Method and apparatus for managing video content | |
CN107180093A (en) | Information search method and device and ageing inquiry word recognition method and device | |
CN110019785A (en) | A kind of file classification method and device | |
Ou et al. | Non-transitive hashing with latent similarity components | |
CN103309984B (en) | The method and apparatus that data process | |
CN107528904A (en) | Method and apparatus for data distribution formula abnormality detection | |
Dang et al. | ComModeler: Topic Modeling Using Community Detection. | |
CN108241692A (en) | The querying method and device of data | |
CN110245297B (en) | Book keyword search-oriented user subject privacy protection method and system | |
CN108153781A (en) | The method and apparatus for extracting the keyword of business scope | |
CN110032721A (en) | A kind of judgement document's method for pushing and device | |
CN104166736A (en) | Storage method and device of reverse index files | |
Ahmed et al. | Visualisation and Analysis of Large and Complex Scale-free Networks. | |
Mousselly-Sergieh et al. | Tag relatedness using laplacian score feature selection and adapted jensen-shannon divergence | |
CN109117434A (en) | Judgement document's search method, device, storage medium and processor |
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 |