CN106484875A - Data processing method based on MOLAP and device - Google Patents
Data processing method based on MOLAP and device Download PDFInfo
- Publication number
- CN106484875A CN106484875A CN201610893549.7A CN201610893549A CN106484875A CN 106484875 A CN106484875 A CN 106484875A CN 201610893549 A CN201610893549 A CN 201610893549A CN 106484875 A CN106484875 A CN 106484875A
- Authority
- CN
- China
- Prior art keywords
- data
- dimension
- precomputation
- result
- postgresql database
- 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/283—Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
Abstract
The invention discloses a kind of data processing method based on MOLAP and device.This data processing method includes:Data cube is created according to true table and dimension table;Based on the data of record in described data cube, data precomputation may be carried out combination to the whole of dimension;Precomputation result is preserved to PostgreSQL database, to calculate result on the estimation in inquiry to determine Query Result.Available data query scheme can be optimized using said method so that non-technical personnel can also realize the inquiry based on mass data.
Description
Technical field
The present embodiments relate to technical field of data processing, the more particularly, to data processing method based on MOLAP and dress
Put.
Background technology
On-line analytical processing (Online Analytical Processing, OLAP) system is that data warehouse is the most main
The application wanted, dedicated for supporting complicated analysis operation, stresses the decision support to decision-maker and senior management staff.
OLAP can carry out the complex query processing of big data quantity according to the requirement of analysis personnel, and intuitively form will be looked into one kind
Ask result and be supplied to decision-maker, so that they accurately grasp the management state of enterprise (company), understand the demand of object, formulate
Correct scheme.
OLAP system according to the data memory format of its memorizer can be divided into relational OLAP (RelationalOLAP,
ROLAP), multidimensional OLAP (MultidimensionalOLAP, MOLAP) and three kinds of mixed type OLAP (HybridOLAP, HOLAP)
Type.Wherein, the multidimensional data used in olap analysis is physically stored as the form of Multidimensional numerical by MOLAP, formed " cube
The structure of body ".
Traditional MOLAP engine is limited by software and hardware resources, be only capable of process gigabit or<The number of 10 terabyte ranks
According to, and, when calculating the data of multi-dimension data cube, server configures are required higher.Meanwhile, magnanimity is directed to based on MOLAP
During data real-time query, frequently with SQL (the SQL on Hadoop) side based on distributed system architecture
Case, on the one hand, its time delay is up to several seconds, tens of second or even several tens minutes, on the other hand, for some columnar database, generally
Can only be good for according to row and carry out quick search, and the inquiry of row rank is then only capable of using in ad hoc inquiry scene.Additionally, in inquiry
When, need to be related to writing so that non-technical personnel cannot be carried out inquiring about of SQL statement.
Content of the invention
In view of this, the embodiment of the present invention provides a kind of data processing method based on MOLAP and device, existing to optimize
Data query scheme is so that non-technical personnel can also realize the inquiry based on mass data.
In a first aspect, embodiments providing a kind of data processing method based on MOLAP, including:
Data cube is created according to true table and dimension table;
Based on the data of record in described data cube, data precomputation may be carried out combination to the whole of dimension;
Precomputation result is preserved to PostgreSQL database, to calculate result on the estimation to determine inquiry knot in inquiry
Really.
Second aspect, the embodiment of the present invention additionally provides a kind of data processing equipment based on MOLAP, including:
Cube creation module, for creating data cube according to true table and dimension table;
Precalculation module, for based in described data cube record data, to dimension whole may combine into
Line number is it is expected that calculate;
Preserving module, for preserving precomputation result to PostgreSQL database, to calculate knot on the estimation in inquiry
Fruit determines Query Result.
Data processing method based on MOLAP provided in an embodiment of the present invention and device, according to true table and dimension table wound
Build data cube, and data precomputation carried out to all possible dimension combination according to the data of record in data cube,
And precomputation result is saved in PostgreSQL database it is achieved that when user carries out data query it is only necessary in client
Pull dimension and tolerance in the page, server just can determine Query Result according to corresponding precomputation result, do not need to use
SQL statement is write at family.Meanwhile, make full use of big data component characteristic and MOLAP characteristic, simplify data query process, improve
Inquiry response speed.
Brief description
By reading the detailed description that non-limiting example is made made with reference to the following drawings, other of the present invention
Feature, objects and advantages will become more apparent upon:
A kind of flow chart of data processing method based on MOLAP that Fig. 1 provides for the embodiment of the present invention one;
A kind of flow chart of data processing method based on MOLAP that Fig. 2 provides for the embodiment of the present invention two;
Fig. 3 is that the whole of the cubical dimension of data may groups hierarchical relationship schematic diagram when being combined in data precomputation;
A kind of flow chart of the creation method of PostgreSQL database table that Fig. 4 provides for the embodiment of the present invention two;
A kind of flow chart of data query method that Fig. 5 provides for the embodiment of the present invention two;
Fig. 6 is user interface schematic diagram during client query;
A kind of structural representation of data processing equipment based on MOLAP that Fig. 7 provides for the embodiment of the present invention three.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that, in order to just
Part related to the present invention rather than full content is illustrate only in description, accompanying drawing.
Embodiment one
A kind of flow chart of data processing method based on MOLAP that Fig. 1 provides for the embodiment of the present invention one.The present embodiment
The method providing can be executed by the data processing equipment based on MOLAP, and this device can be by the side of software and/or hardware
Formula realize, and integrated in the server.With reference to Fig. 1, the method that the present embodiment provides can include:
S110, data cube is created according to true table and dimension table.
Wherein, true table and dimension table are stored in the data warehouse of big data platform.Dimension table stores institute under this dimension
There are the value of attribute and the ID of each record.Such as sell region dimension to include:Region ID, the affiliated province in this region, affiliated
The information such as city, affiliated county, taking the value that Tianhe district of Guangzhou is under this sale region dimension as a example, then with regard to the Milky Way in dimension table
Area is recorded as:01 (ID being worth), Guangdong Province, Guangzhou, Tianhe District.Factual data is store, including each dimension in true table
The ID of value of degree, sales volume, consumption sum etc..The field such as selling true table includes each sales record:Sell
Region ID (the corresponding ID selling region dimension table), product IDs (ID of corresponding product dimension table), selling time ID (corresponding time
The ID of dimension table), sales volume, consumption sum etc., according to the corresponding each ID of a certain sales volume, just it is known that this sale number
Amount is sales volume within which section time for which kind of product of which region.
Optionally, it is pre-created true table and dimension table, wherein, specific establishment rule, for example, select which dimension is created
Build dimension table, and create several dimension tables, can be set according to practical situation.Under normal circumstances, the number of dimension table
It is not construed as limiting, the number of true table is one.
Typically, after creating true table and dimension table, obtain from the external data base that current server can access
In data, and the fact that be all directed into corresponding table and dimension table.
Specifically, data cube is referred to as multi-dimension data cube, and it can comprise at least one dimension setting.Than
As a data cube can include:(value includes region dimension:Beijing, Tianjin, Shanghai and Guangdong), time dimension (value bag
Include:The first quarter, for the second quarter, the third quater and the fourth quater), (value includes commodity dimension:Gym suit, kettle, sport shoess, parasols with
And sunbonnet) and (the value inclusion of age of user dimension:0-18 year, 19-40 year, 41-60 year and more than 60 years old), wherein each dimension
In the fact that corresponding factual data (including consumption sum and sales volume) is stored in association table, that is, this data cube includes
Four dimension tables and a true table, and be associated by the ID of the dimension table intermediate value of record in dimension table and true table.Root
Just can determine factual data according to the data cube determining.As Beijing area just be can determine according to above-mentioned data cube
The consumption sum of first quarter gym suit, for another example, it may also be determined that the consumption sum of Efficiency in Buildings in Tianjin Area kettle, and for example, it may also be determined that
The year District of Shanghai fourth quater 19-40 consumer movementss' footwear sales volume.
Optionally, at least one data cube can be created with the dimension table associating according to true table, wherein, specifically
Create rule to be set according to practical situation.
The whole of dimension combination may be carried out data precomputation by S120, the data based on record in data cube.
Wherein, different dimensions can be included in data cube, to can be in any combination between different dimensions.With
For the data cube of citing in step S110, comprise four dimensions altogether, then the whole of dimension combination may have 16, tool
The calculation of body can beWithAs a example, corresponding dimension combination includes (region
Dimension, time dimension and commodity dimension), (region dimension, commodity dimension and age of user dimension), (region dimension, time dimension
With age of user dimension) and (time dimension, commodity dimension and age of user dimension).Certainly, when data cube only comprises
During one dimension, all may combine combination and the empty set only comprising this dimension.
Further, after determining the whole possible combination of data cube, line number is entered to the corresponding factual data of each combination
It is expected that calculating, to obtain the precomputation result of each combination.Optionally, data precomputation is carried out according to the tolerance determining.In this reality
Apply in example, how tolerance calculates or requirement that result of calculation sets it is expected that calculating if being logarithm, preferably include polymeric type (polymeric type
Type includes being polymerized maximum and/or polymerization minima).Assume that dimension combination includes region dimension and time dimension, polymeric type
For be polymerized maximum, then to during dimension data splitting precomputation be respectively determine (region dimension, time dimension), (region tie up
Degree), (time dimension) andThe polymerization maximum of corresponding factual data.Taking (region dimension) as a example, data precomputation
It is the polymerization maximum obtaining Beijing in region dimension, Tianjin, Shanghai and factual data in the fact that the corresponding table of Guangdong.
Accordingly, all possible dimension combination corresponding to data cube data precomputation can be carried out.
S130, precomputation result is preserved to PostgreSQL database, look into calculate result on the estimation in inquiry and to determine
Ask result.
In the present embodiment, PostgreSQL database is preferably Hbase.Hbase is distributed memory system, can using Hbase
To build extensive destructuring storage cluster in the middle of common server.
Further, when user is inquired about it is only necessary to selecting, in client, the dimension combination needing and measuring, service
Device just can find corresponding precomputation result in Hbase, and this precomputation result is fed back to client.
The technical scheme that the present embodiment provides, creates data cube by true table with the dimension table associating, to data
In cube, the whole of dimension combination may carry out data precomputation, and precomputation result is saved in PostgreSQL database, can
Calculated with the big data realized based on MOLAP and store.Meanwhile, when user is inquired about, SQL statement need not be write it is only necessary to
To pull dimension and the tolerance of needs in the page of client, server just can determine according to corresponding precomputation result to be looked into
Ask result, simplify data query process, improve inquiry response speed.
Embodiment two
A kind of flow chart of data processing method based on MOLAP that Fig. 2 provides for the embodiment of the present invention two.The present embodiment
The data processing method providing is on the basis of the data processing method that above-described embodiment provides, and is optimized.With reference to Fig. 2,
The data processing method that the present embodiment provides, specifically includes:
S210, according to the list item demand to true table and dimension table in default Data Analysis Model, create corresponding thing
Real table and dimension table.
Specifically, default Data Analysis Model includes the list item demand to true table and dimension table, and wherein, list item needs
Ask the dimension values that can include dimension table and dimension hierarchy, and the fact that true table data attribute (as consumption sum etc.).Separately
Outward, the tolerance (as polymerization maximum etc.) during data precomputation can also be included in Data Analysis Model.In Data Analysis Model
Particular content can be set according to practical situation.In general, Data Analysis Model can be regarded as number in the present embodiment
According to the plan of processing method, that is, in the present embodiment, data processing method executes according to Data Analysis Model.
Exemplary, according to list item demand, create true table and the dimension table associating with true table.
S220, according to true table and dimension table list item demand, by the data in external data base be directed into true table and
In dimension table.
Specifically, external data base is accessed according to list item demand, and find corresponding data from external data base and import
In true table and dimension table.Wherein, external data base can include data base and the enterprise external of related each business management system
Database.
When further, to true table and dimension table importing data, it is specific that data warehouse may determine that whether user sets
Data form, if set specific data form, original form of data is converted into specific data form, and imports
In true table and dimension table, if not setting specific data form, according to original form of data, import data to the fact
In table and dimension table.
S230, according to the metadata in Data Analysis Model, create data cube using true table and dimension table.
Exemplary, Data Analysis Model includes metadata, and wherein, metadata is used for indicating the cubical attribute of data
Parameter and establishment rule.Optionally, metadata can include following at least one:The creation time of data cube, data are stood
The fact that the establishment position of cube, the title of data cube, data cube, the dimension of data cube and its order, number
According to cubical tolerance, polymeric type configuration information corresponding with programming model configuration information corresponding with PostgreSQL database,
Storage information of pre-calculation time and precomputation result etc..Wherein, configuration information corresponding with programming model is to start volume
During journey model, inform the information of the memory size that programming model data cube needs to take.Corresponding with PostgreSQL database join
Confidence breath is the information of data cube precomputation result and the corresponding relation of PostgreSQL database table.The storage letter of precomputation result
Breath includes precomputation result be saved in multiple dimensions codings that row major key in PostgreSQL database is related to order (if any dimension A with
Dimension B, wherein, dimension A includes three values, respectively 1,2 and 3, and corresponding precomputation result is being stored to data of increasing income
During storehouse, to 1,2 and 3 orders being encoded, and the order to dimension A and dimension B coding).Optionally, with PostgreSQL database
Data in the storage information of corresponding configuration information, pre-calculation time and precomputation result can complete to store precomputation
It is updated after result.
The essential information when creating for the data cube, such as creation time, establishment just be can determine according to above-mentioned metadata
Position, title, the fact, dimension and its order and tolerance etc..Optionally, the essential information of above-mentioned metadata is stored in Hbase
In.That is, when needing the essential information of data cube, Hbase inquiry can be accessed.
S240, according to the metadata in Data Analysis Model, start precomputation programming model task, and read and stand with data
The corresponding whole dimension tables of cube and the data of true table.
Specifically, create and start the task of corresponding precomputation programming model according to metadata.In the present embodiment it is contemplated that
Calculation programming model is MapReduce, and it can realize the concurrent operation of large-scale dataset.
Optionally, by precomputation programming model read data platform in whole dimension tables corresponding with data cube and
The data of true table.
S250, the dimension of whole dimension tables is carried out permutation and combination, obtain including empty set whole may combination.
S260, according to set polymeric rule, aminated polyepichlorohydrin is carried out to the combination comprising whole dimensions, obtains polymerizing value.
For example, if data cube includes 4 dimensions, represented with A, B, C and D respectively, then comprise the combination of whole dimensions
For (A, B, C, D).
Optionally, polymeric rule can include being polymerized maximum and/or polymerization minima.Taking be polymerized maximum as a example, then
Aminated polyepichlorohydrin is the maximum MAX (M) returning (A, B, C, D) corresponding factual data using aggregate function, and using MAX (M) as
Polymerizing value.
Optionally, corresponding value the whole dimensions that comprise after dictionary encoding, and calculation code can be carried out to each dimension
The polymerizing value of combination.
S270, using comprise whole dimensions combination as precomputation programming model key value input, using polymerizing value as
The keyword input of precomputation programming model.
Specific it is contemplated that calculating programming model when calculating by the way of step-by-step calculation.Wherein, when calculating, will comprise
All the combination of dimension is as the key value input of precomputation programming model, i.e. key input.Polymerizing value is programmed as precomputation
The keyword input of model, i.e. value input.
S280, obtain new dimension combination using precomputation programming model and combine corresponding polymerization with new dimension
Value.
Because by the way of step-by-step calculation, then calculated new dimension is combined as comprising the combination of whole dimensions
Higher level's dimension combines, and polymerizing value is the polymerizing value of new dimension combination.
For example, Fig. 3 be the cubical dimension of data whole may groups when being combined in data precomputation hierarchical relationship show
It is intended to.With reference to Fig. 3 it may be determined that current data cube comprises four dimensions value, respectively A, B, C and D.Wherein, bottom
Level for calculating the dimension combination of polymerizing value at first, that is, comprises the combination of whole dimensions, according to the hierarchical relationship in Fig. 3, step by step to
Upper calculating, until calculating to empty set.
S290, judge whether to complete to the data precomputation that all may combine.If so, then execute S2120, otherwise, hold
Row S2100.
Wherein it is possible to by judging whether to complete the data precomputation to empty set, it is confirmed whether to complete to all may tieing up
The data precomputation of degree combination.
S2100, using new dimension combination as precomputation programming model key value input, new dimension combination is corresponding
Polymerizing value inputs as the keyword of precomputation programming model.
Exemplary, if not completing to the data precomputation that all may combine, can be by precomputation programming model
As key input, corresponding polymerizing value inputs as value for the last calculated new dimension combination, with continue into
Line number is it is expected that calculate.
S2110, obtain new dimension combination using precomputation programming model and combine corresponding polymerization with new dimension
Value.Return execution S290.
Further, after obtaining precomputation result, pre-calculation time can be preserved in the metadata.
The data precomputation process of a certain data cube of explanation exemplary below.
Set this data cube and comprise 4 dimensions, the value of respectively A, B, C and D, wherein dimension A is 1, dimension B
It is worth for 2, the value of dimension C is (3,4), the value of dimension D is (4,6,3,8).I.e. this data cube comprises 4 dimension tables and 1 altogether
Individual fact table.
Further, according to metadata in Data Analysis Model, start precomputation programmed tasks model, and determine this data
The whole of cubical dimension combination may have 16.The polymeric rule setting is as polymerization maximum and polymerization minima.Right
After the corresponding value of each dimension carries out dictionary encoding, data precomputation is carried out to dimension combination (A, B, C, D), obtained polymerization maximum
Value MAX (M) and polymerization minimum value MIN (N), that is, complete the first order and calculate.
Table 1
Can determine the polymerization result of calculation of (A, B, C, D) dimension combination by table 1.Wherein, by (A, B, C, D) conduct
The key input of MapReduce, using MAX (M) and MIN (N) as the value input of MapReduce, then can obtain new dimension
Degree combination, including (A, B, C), (A, B, D), (A, C, D) and (B, C, D), and each new dimension combines corresponding polymerizing value, that is,
Complete the second level to calculate.
What table 2 was merely exemplary lists the data precomputation result of dimension combination (A, B, C).
Table 2
A | B | C | MAX(M) | MIN(N) |
1 | 2 | 3 | 300 | 50 |
1 | 2 | 4 | 500 | 20 |
Further, the new dimension combination obtaining when MapReduce being calculated in the second level, will used as key input
(A, B, C), (A, B, D), (A, C, D) and (B, C, D) inputs as key, and each dimension is combined corresponding MAX (M) and MIN (N)
As value input, just can obtain new dimension combination, and each new dimension combines corresponding polymerizing value, that is, complete
The third level calculates.Wherein, the new dimension combination obtaining includes (A, B), (A, C), (A, D), (B, C), (B, D) and (C, D).
What table 3 was merely exemplary lists the data precomputation result of dimension combination (A, B).
Table 3
A | B | MAX(M) | MIN(N) |
1 | 2 | 500 | 20 |
Further, new dimension combination (A, B) that MapReduce obtained when the third level calculates, (A, C), (A,
D), (B, C), (B, D) and (C, D) inputs as key, each dimension is combined corresponding MAX (M) and MIN (N) defeated as value
Enter, just can obtain new dimension combination and each new dimension combines corresponding polymerizing value, that is, complete the fourth stage and calculate.Its
In, this grade of calculated new dimension combination includes:(A), (B), (C) and (D).
What table 4 was merely exemplary lists the data precomputation result of dimension combination (A).
Table 4
A | MAX(M) | MIN(N) |
1 | 500 | 20 |
Further, new dimension combination (A), (B), (C) and (D) MapReduce being obtained in fourth stage calculating
As key input, each dimension is combined corresponding MAX (M) and MIN (N) and inputs as value, just can obtain new dimension
Combination and each new dimension combine corresponding polymerizing value, that is, complete the calculating of afterbody.Wherein, this grade calculated
New dimension is combined as
Dimension combination listed by table 5Data precomputation result.After obtaining the data precomputation result of empty set, confirm
Complete the precomputation to this data cube.
Table 5
S2120, create PostgreSQL database table for storing precomputation result.
During due to being stored in precomputation result in PostgreSQL database, it is that precomputation result is stored in data of increasing income in logic
In the table of storehouse.Therefore, it is pre-created the PostgreSQL database table for storing precomputation result.Wherein, different pieces of information is cubical pre-
The different PostgreSQL database table of result of calculation correspondence.
Wherein, with reference to Fig. 4, this step can include:
S2121, determine in PostgreSQL database for PostgreSQL database table set memory block capacity.
Wherein, memory block is memory block in logic.In PostgreSQL database, inquiry data is by inquiring about data of increasing income
Storehouse table is realized.If the data volume being stored in PostgreSQL database table constantly increases, the speed of inquiry data will be slack-off.This
When it is possible to subregion is carried out to PostgreSQL database table, that is, be divided into multiple memory blocks in logic.After tables of data of increasing income subregion, table
Existing form remains a complete table, the memory block simply when inquiring about data, in inquiry PostgreSQL database table.
Typically, PostgreSQL database can preset the capacity of each PostgreSQL database table each memory block in subregion.Generally,
This capacity keeps constant after setting as far as possible.As set the capacity of memory block as 256M, that is, each memory block at most can store
The data of 256M.
The size of S2122, on the estimation calculation result, determines that precomputation result is stored in required depositing during PostgreSQL database table
The quantitative value of storage area.
For example it is contemplated that the size calculating result is 3G, the capacity of memory block is 256M, then obtain 12 with 3G divided by 256M (3 ×
1024 ÷ 256=12), that is, determine that the quantitative value needing memory block is 12.
Optionally, when calculating the memory block quantitative value that precomputation result needs, if integer amount value can not be obtained, right
The result obtaining rounds on carrying out, and such as it is contemplated that the size calculating result is 3.1G, the capacity of each memory block is 256M, then use
3.1G obtains 12.4 divided by 256M, obtains 13 to rounding on 12.4.I.e. the quantitative value of this corresponding memory block of precomputation result is
13.
S2123, quantitative value is sent to PostgreSQL database, so that PostgreSQL database creates for storing according to quantitative value
The PostgreSQL database table of precomputation result.
Further, after determining the quantitative value of memory block that precomputation result needs, this quantitative value is sent to increasing income
In data base, that is, notify the size of the PostgreSQL database table of PostgreSQL database storage current precomputation result needs.
Specifically, after PostgreSQL database receives quantitative value, corresponding PostgreSQL database table, this PostgreSQL database table bag are created
Memory block containing this quantitative value.
For example, determine that the memory block that current precomputation result needs is 12, then PostgreSQL database creates PostgreSQL database
Table, wherein, this PostgreSQL database table actually includes 12 memory blocks.
S2130, startup storage programming model task, using precomputation result as the input storing programming model task.
Specifically, storage programming model task is started according to metadata.In the present embodiment, storage programming model is
MapReduce.Wherein it is possible to using precomputation result as MapReduce input.
S2140, utilization storage programming model generate corresponding binary format file.
Further, the binary format file that the result of storage programming model output is given tacit consent to for PostgreSQL database, that is,
The file of HFile form.Advantage of this is that, precomputation result can be avoided to insert the property brought in PostgreSQL database one by one
Can impact.
S2150, using PostgreSQL database BulkLoad, binary format file is directed into described PostgreSQL database table
In, to realize storing described precomputation result in PostgreSQL database.
Specifically, when binary format file being directed into PostgreSQL database table, using the BulkLoad of PostgreSQL database,
Can realize binary format file is simultaneously directed to the corresponding memory block of PostgreSQL database table.
For example, binary format file includes the precomputation result of 1G, and corresponding PostgreSQL database table comprises 4 storages
Area, then when utilizing BulkLoad, it is possible to achieve front 256M precomputation result imports the 1st memory block, following 256M precomputation knot
Phenolphthalein enters the 2nd memory block, next 256M precomputation result imports the 3rd memory block and the precomputation result of last 256M is led
The importing process entering the 4th memory block is carried out simultaneously.
In the present embodiment, the benefit carrying out partitioned storage to PostgreSQL database table is:On the one hand, precomputation result is being led
Fashionable, multiple nodes can be had simultaneously to enter the operation of row write, accelerate the speed of data write by the principle of load balancing;Another
Query capability, when inquiring about data, can be distributed to each destination node (i.e. memory block), be effectively prevented from data and incline by aspect
Tiltedly, accelerate the speed of data query.
S2160, preservation precomputation result and the corresponding relation extremely described Data Analysis Model of described PostgreSQL database table
In metadata.
Specifically, when preserving precomputation result, the corresponding relation of precomputation result and PostgreSQL database table is saved in
In the configuration information corresponding with PostgreSQL database of metadata.When inquiring about data, just can be determined estimated according to corresponding relation
Calculate result corresponding PostgreSQL database table.
Optionally, said method can also include:Interval setting time, management data cube and Data Analysis Model
In metadata.
Wherein it is possible to interval setting time, the total data cube in management platform and corresponding metadata.Logarithm
Can include according to cubical management:Modification, inquiry, calculating and deletion etc..
For example, it is spaced setting time, the new recorded data in external data base is imported the fact that data cube is corresponding
In table and dimension table, and combine the new data modification data cube importing, corresponding precomputation result and precomputation result
Storage location.And, amended data cube relevant information is preserved to corresponding metadata, with complete paired data
Cubical modification and calculating.
Optionally, interlude can set according to time situation.As when business dull season, can be every fortnight
Complete paired data cube and the management of metadata.When the business busy season, every a week complete paired data cube with
And the management of metadata.
Above-mentioned steps are inquiry set-up procedure, and here is specific query script.The number providing with reference to Fig. 5, the present embodiment
Specifically include according to querying method:
S510, when getting the inquiry request of client transmission, parsing inquiry request is simultaneously converted into PostgreSQL database and looks into
Ask sentence.
Fig. 6 is user interface schematic diagram during client query.User can select to need the dimension of inquiry by user interface
Degree and tolerance.As shown in fig. 6, dimension A is encoded 1, dimension C1 level 1, dimension D1 level 1 and tolerance MAX1 by user, and (polymerization is
Big value) it is drawn to the row as inquiry in row major key 61, certainly, user can arbitrarily pull user's assets according to the actual requirements and divide
The dimension of analysis model 1 and index.Specifically, client generates inquiry request and sends to server according to the selection of user.Can
Choosing, user can also input filtercondition by the filter column 62 in Fig. 6, to filter to inquiry request.
Further, server, when getting inquiry request, this inquiry request is converted into PostgreSQL database and can recognize that
Query statement.
S520, PostgreSQL database query statement is sent to PostgreSQL database, with according to PostgreSQL database query statement
Transfer precomputation result, and form Query Result.
Specifically, when PostgreSQL database receives PostgreSQL database query statement, result and data of increasing income are calculated on the estimation
The corresponding relation of storehouse table determines this PostgreSQL database query statement corresponding PostgreSQL database table, inquires about in PostgreSQL database table
With this PostgreSQL database query statement corresponding precomputation result, and form Query Result.
Further, when inquiring about PostgreSQL database table, can be only to the head data of storage and mantissa's evidence in each memory block
Inquired about, if confirming the corresponding result of PostgreSQL database query statement in the scope of data that certain memory block stores,
This memory block carries out data query, to determine PostgreSQL database query statement corresponding precomputation result.
S530, Query Result is encapsulated and is back to client, so that client end response Query Result.
Specifically, client, when receiving Query Result, parses this Query Result, and the graphical interface of user in Fig. 6
Viewing area 63 show this Query Result, wherein, the display format form of acquiescence.Optionally, user can also pass through
In Fig. 6, display type module 64 determines the display type of Query Result, inputs show bar number by bar digital-to-analogue block 65.
The technical scheme that the present embodiment provides, by the combination of MOLAP and big data it is achieved that the inquiry of mass data,
And improve inquiry velocity.Meanwhile, in inquiry, SQL statement need not be write so that non-technical personnel user can also execute looks into
Ask work, improve the experience of user.
Embodiment three
A kind of structural representation of data processing equipment based on MOLAP that Fig. 7 provides for the embodiment of the present invention three.Reference
Fig. 7, the data processing equipment that the present embodiment provides specifically includes:Cube creation module 701, precalculation module 702 and preservation
Module 703.
Wherein, cube creation module 701, for creating data cube according to true table and dimension table;Precomputation mould
The whole of dimension, for the data based on record in data cube, combination may be carried out data precomputation by block 702;Preserve
Module 703, for preserving precomputation result to PostgreSQL database, to calculate result on the estimation in inquiry to determine inquiry
Result.
The technical scheme that the present embodiment provides, creates data cube by true table with the dimension table associating, to data
In cube, the whole of dimension combination may carry out data precomputation, and precomputation result is saved in PostgreSQL database, can
Calculated with the big data realized based on MOLAP and store.Meanwhile, when user is inquired about, SQL statement need not be write it is only necessary to
To pull dimension and tolerance in the page of client, server just can determine inquiry knot according to corresponding precomputation result
Really, simplify data query process, improve inquiry response speed.
On the basis of above-described embodiment, cube creation module 701 includes:Tables of data creating unit, for according to pre-
If Data Analysis Model in list item demand to true table and dimension table, the fact that create corresponding table and dimension table;Data is led
Enter unit, for the list item demand according to true table and dimension table, the data in external data base is directed into true table and dimension
In degree table;Cube creating unit, for according to the metadata in Data Analysis Model, creating number using true table and dimension table
According to cube, wherein, metadata is used for indicating the cubical property parameters of data and creates rule.
Include it is contemplated that calculating module 702 on the basis of above-described embodiment:Precomputation task start unit, for according to number
According to the metadata in analysis model, start precomputation programming model task, and read whole dimensions corresponding with data cube
Table and the data of true table;Assembled unit, for the dimension of whole dimension tables is carried out permutation and combination, obtains including empty set
Whole may combination;Polymerized unit, for according to the polymeric rule setting, carrying out polymerization fortune to the combination comprising whole dimensions
Calculate, obtain polymerizing value;Input value determining unit, for using comprise whole dimensions combination as precomputation programming model key
Value input, polymerizing value is inputted as the keyword of precomputation programming model;Result signal generating unit, for being programmed using precomputation
Model obtains new dimension combination and combines corresponding polymerizing value with new dimension;Cycling element, for successively by new dimension
Degree combination is combined corresponding polymerizing value and is programmed mould as precomputation as the key value input of precomputation programming model, new dimension
The keyword input of type, and obtain new dimension combination and corresponding poly- with new dimension combination using precomputation programming model
Conjunction value, till the polymerizing value that all may combine until obtaining and all may combine.
On the basis of above-described embodiment, preserving module 703 includes:Build table unit, be used for storing precomputation for creating
The PostgreSQL database table of result;Store tasks set up unit, for start storage programming model task, using precomputation result as
The input of storage programming model task;File generating unit, for generating corresponding binary format using storage programming model
File;Result import unit, for the BulkLoad using PostgreSQL database, binary format file is directed into data of increasing income
In the table of storehouse, to realize storing precomputation result in PostgreSQL database;Relation memory element, for preserving precomputation result and opening
The corresponding relation of source database table is to the metadata of Data Analysis Model.
On the basis of above-described embodiment, build table unit and include:Capacity determination subelement, for determining in PostgreSQL database
The capacity of the memory block setting for PostgreSQL database table;Quantitative value determination subelement, for calculating the size of result on the estimation, really
Determine the quantitative value that precomputation result is stored in required memory block during PostgreSQL database table;Create subelement, for by quantitative value
Send to PostgreSQL database, so that PostgreSQL database creates the PostgreSQL database for storing precomputation result according to quantitative value
Table.
On the basis of above-described embodiment, also include:Management module, for being spaced setting time, manages data cube
And metadata in Data Analysis Model.
On the basis of above-described embodiment, also include:Sentence acquisition module, in the inquiry getting client transmission
During request, parsing inquiry request is simultaneously converted into PostgreSQL database query statement;Result queries module, for looking into PostgreSQL database
Ask sentence to send to PostgreSQL database, so that precomputation result is transferred according to PostgreSQL database query statement, and form inquiry knot
Really;Result returns module, for Query Result being encapsulated and being back to client, so that client end response Query Result.
On the basis of above-described embodiment, metadata include following at least one:The creation time of data cube, data are stood
The fact that the establishment position of cube, the title of data cube, data cube, the dimension of data cube and its order, number
According to cubical tolerance, polymeric type configuration information corresponding with programming model configuration information corresponding with PostgreSQL database,
Pre-calculation time and the storage information of result of calculation.
The base that above-mentioned any embodiment offer is provided based on the data processing equipment of MOLAP provided in an embodiment of the present invention
In the data processing method of MOLAP, possess corresponding function and beneficial effect.
Note, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore although being carried out to the present invention by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other Equivalent embodiments more can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of data processing method based on MOLAP is it is characterised in that include:
Data cube is created according to true table and dimension table;
Based on the data of record in described data cube, data precomputation may be carried out combination to the whole of dimension;
Precomputation result is preserved to PostgreSQL database, to calculate result on the estimation in inquiry to determine Query Result.
2. data processing method according to claim 1 is stood it is characterised in that creating data according to true table and dimension table
Cube includes:
According to the list item demand to true table and dimension table in default Data Analysis Model, create the fact that correspond to table and dimension
Table;
According to the list item demand of described fact table and dimension table, the data in external data base is directed into true table and dimension table
In;
According to the metadata in described Data Analysis Model, create data cube using described fact table and described dimension table,
Wherein, described metadata is used for indicating the property parameters of described data cube and creates rule.
3. data processing method according to claim 1 it is characterised in that based in described data cube record number
According to, data precomputation may be carried out combination to the whole of dimension, including:
According to the metadata in Data Analysis Model, start precomputation programming model task, and read and described data cube
Corresponding whole dimension table and the data of true table;
The dimension of described whole dimension tables is carried out permutation and combination, obtains the whole possible combination including empty set;
According to the polymeric rule setting, aminated polyepichlorohydrin is carried out to the combination comprising whole dimensions, obtains polymerizing value;
The combination comprising whole dimensions is inputted as the key value of precomputation programming model, using described polymerizing value as precomputation
The keyword input of programming model;
Obtain new dimension combination using described precomputation programming model and combine corresponding polymerizing value with new dimension;
Successively new dimension combination is inputted as the key value of precomputation programming model, new dimension combines corresponding polymerizing value
As the keyword input of precomputation programming model, and using described precomputation programming model obtain new dimension combination and with
Till new dimension combines corresponding polymerizing value, the polymerizing value that all may combine until obtaining and all may combine.
4. data processing method according to claim 1 is it is characterised in that preserve precomputation result to PostgreSQL database
Include:
Create the PostgreSQL database table for storing precomputation result;
Start storage programming model task, using precomputation result as the described input storing programming model task;
Generate corresponding binary format file using described storage programming model;
Using the BulkLoad of PostgreSQL database, binary format file is directed in described PostgreSQL database table, to realize
Described precomputation result is stored in PostgreSQL database;
Preserve the metadata of described precomputation result and the corresponding relation extremely described Data Analysis Model of described PostgreSQL database table
In.
5. data processing method according to claim 4 is it is characterised in that create for storing increasing income of precomputation result
Database table includes:
Determine the capacity of the memory block setting in PostgreSQL database for PostgreSQL database table;
Calculate the size of result on the estimation, determine that described precomputation result is stored in required memory block during PostgreSQL database table
Quantitative value;
Described quantitative value is sent to described PostgreSQL database, so that described PostgreSQL database creates according to described quantitative value using
PostgreSQL database table in storage precomputation result.
6. data processing method according to claim 1 is it is characterised in that also include:
Interval setting time, manages metadata in described data cube and described Data Analysis Model.
7. data processing method according to claim 1 is it is characterised in that preserve precomputation result to PostgreSQL database
In after, also include:
When getting the inquiry request of client transmission, parse described inquiry request and be converted into PostgreSQL database inquiry language
Sentence;
Described PostgreSQL database query statement is sent to PostgreSQL database, pre- to be transferred according to PostgreSQL database query statement
Result of calculation, and form Query Result;
Described Query Result is encapsulated and is back to described client, so that Query Result described in described client end response.
8. the data processing method according to any one of claim 2-6 it is characterised in that described metadata include following extremely
One item missing:
The creation time of data cube, the establishment position of data cube, the title of data cube, the thing of data cube
In fact, the dimension of data cube and its order, the tolerance of data cube, polymeric type configuration corresponding with programming model letter
Cease the storage information of configuration information corresponding with PostgreSQL database, pre-calculation time and precomputation result.
9. a kind of data processing equipment based on MOLAP is it is characterised in that include:
Cube creation module, for creating data cube according to true table and dimension table;
The whole of dimension, for the data based on record in described data cube, may be combined into line number by precalculation module
It is expected that calculating;
Preserving module, for preserving precomputation result to PostgreSQL database, so that it is true to calculate result on the estimation in inquiry
Determine Query Result.
10. data processing equipment according to claim 9 is it is characterised in that described cube creation module includes:
Tables of data creating unit, for according to the list item demand to true table and dimension table in default Data Analysis Model, creating
The fact that build corresponding table and dimension table;
Data import unit, for the list item demand according to described fact table and dimension table, the data in external data base is led
Enter to true table and dimension table;
Cube creating unit, for according to the metadata in described Data Analysis Model, using described fact table and described dimension
Degree table creates data cube, and wherein, described metadata is used for indicating the property parameters of described data cube and creates rule.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610893549.7A CN106484875B (en) | 2016-10-13 | 2016-10-13 | MOLAP-based data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610893549.7A CN106484875B (en) | 2016-10-13 | 2016-10-13 | MOLAP-based data processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106484875A true CN106484875A (en) | 2017-03-08 |
CN106484875B CN106484875B (en) | 2019-12-31 |
Family
ID=58270539
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610893549.7A Active CN106484875B (en) | 2016-10-13 | 2016-10-13 | MOLAP-based data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106484875B (en) |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106997386A (en) * | 2017-03-28 | 2017-08-01 | 上海跬智信息技术有限公司 | A kind of OLAP precomputations model, method for automatic modeling and automatic modeling system |
CN107229730A (en) * | 2017-06-08 | 2017-10-03 | 北京奇虎科技有限公司 | Data query method and device |
CN107729399A (en) * | 2017-09-21 | 2018-02-23 | 北京京东尚科信息技术有限公司 | The method and apparatus of data processing |
CN108280046A (en) * | 2017-11-30 | 2018-07-13 | 深圳市科列技术股份有限公司 | A kind of method, battery data server and the user terminal of battery data processing |
CN108334554A (en) * | 2017-12-29 | 2018-07-27 | 上海跬智信息技术有限公司 | A kind of novel OLAP precomputations model and construction method |
CN108875008A (en) * | 2018-06-15 | 2018-11-23 | 湖北德普电气股份有限公司 | A kind of Large Volume Data analysis method and device |
CN109213829A (en) * | 2017-06-30 | 2019-01-15 | 北京国双科技有限公司 | Data query method and device |
CN109241159A (en) * | 2018-08-07 | 2019-01-18 | 威富通科技有限公司 | A kind of subregion querying method, system and the terminal device of data cube |
CN109308301A (en) * | 2018-09-28 | 2019-02-05 | 中国银行股份有限公司 | The preparation method and device of test data |
CN109710859A (en) * | 2019-01-21 | 2019-05-03 | 北京字节跳动网络技术有限公司 | Data query method and apparatus |
CN109885609A (en) * | 2019-01-31 | 2019-06-14 | 平安科技(深圳)有限公司 | Based on combined data area control method, device and storage medium |
CN109976910A (en) * | 2019-03-20 | 2019-07-05 | 跬云(上海)信息科技有限公司 | Querying method and device based on precomputation OLAP model |
CN109992594A (en) * | 2019-03-22 | 2019-07-09 | 跬云(上海)信息科技有限公司 | Distributed based on precomputation optimization executes optimization method and system |
CN110008239A (en) * | 2019-03-22 | 2019-07-12 | 跬云(上海)信息科技有限公司 | Logic based on precomputation optimization executes optimization method and system |
CN110110165A (en) * | 2019-04-01 | 2019-08-09 | 跬云(上海)信息科技有限公司 | Dynamic routing method and device for query engine in precomputation system |
CN110222124A (en) * | 2019-05-08 | 2019-09-10 | 跬云(上海)信息科技有限公司 | Multidimensional data processing method and system based on OLAP |
CN110275920A (en) * | 2019-06-27 | 2019-09-24 | 中国石油集团东方地球物理勘探有限责任公司 | Data query method, apparatus, electronic equipment and computer readable storage medium |
CN110674117A (en) * | 2019-09-26 | 2020-01-10 | 京东数字科技控股有限公司 | Data modeling method and device, computer readable medium and electronic equipment |
CN110704514A (en) * | 2019-10-25 | 2020-01-17 | 南京录信软件技术有限公司 | Precomputation method based on Lucene |
CN110781228A (en) * | 2018-07-13 | 2020-02-11 | 马上消费金融股份有限公司 | Data query method, system and device based on data warehouse and storage medium |
CN111061758A (en) * | 2018-10-16 | 2020-04-24 | 杭州海康威视数字技术股份有限公司 | Data storage method, device and storage medium |
CN111125264A (en) * | 2019-12-12 | 2020-05-08 | 跬云(上海)信息科技有限公司 | Extra-large set analysis method and device based on extended OLAP model |
CN111125266A (en) * | 2019-12-24 | 2020-05-08 | 中国建设银行股份有限公司 | Data processing method, device, equipment and storage medium |
CN111563123A (en) * | 2020-05-07 | 2020-08-21 | 北京首汽智行科技有限公司 | Live warehouse metadata real-time synchronization method |
CN112100177A (en) * | 2020-09-04 | 2020-12-18 | 北京三快在线科技有限公司 | Data storage method and device, computer equipment and storage medium |
CN112835966A (en) * | 2019-11-22 | 2021-05-25 | 北京金山云网络技术有限公司 | Data query method and device and electronic equipment |
CN110019357B (en) * | 2017-09-29 | 2021-06-29 | 北京国双科技有限公司 | Database query script generation method and device |
CN113094409A (en) * | 2021-04-08 | 2021-07-09 | 国网电子商务有限公司 | Service data processing method and device and computer storage medium |
CN113220719A (en) * | 2021-06-04 | 2021-08-06 | 上海天旦网络科技发展有限公司 | Mass dimension data association query optimization method and system |
CN113407587A (en) * | 2021-07-19 | 2021-09-17 | 北京百度网讯科技有限公司 | Data processing method, device and equipment for online analysis processing engine |
WO2023273183A1 (en) * | 2021-06-30 | 2023-01-05 | 北京邮电大学 | Hybrid engine-based multidimensional data query method and apparatus |
WO2023155273A1 (en) * | 2022-02-15 | 2023-08-24 | 上海跬智信息技术有限公司 | Correlation coefficient calculation method and apparatus, device, and storage medium |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103164222A (en) * | 2013-02-25 | 2013-06-19 | 用友软件股份有限公司 | Multidimensional modeling system and multidimensional modeling method |
-
2016
- 2016-10-13 CN CN201610893549.7A patent/CN106484875B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103164222A (en) * | 2013-02-25 | 2013-06-19 | 用友软件股份有限公司 | Multidimensional modeling system and multidimensional modeling method |
Non-Patent Citations (2)
Title |
---|
郭文君: "联机分析处理的研究与应用", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
高伟: "基于MapReduce的数据聚集运算算法研究与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (45)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106997386B (en) * | 2017-03-28 | 2019-12-27 | 上海跬智信息技术有限公司 | OLAP pre-calculation model, automatic modeling method and automatic modeling system |
WO2018176623A1 (en) * | 2017-03-28 | 2018-10-04 | 上海跬智信息技术有限公司 | Olap precomputed model, automatic modeling method, and automatic modeling system |
CN106997386A (en) * | 2017-03-28 | 2017-08-01 | 上海跬智信息技术有限公司 | A kind of OLAP precomputations model, method for automatic modeling and automatic modeling system |
CN107229730A (en) * | 2017-06-08 | 2017-10-03 | 北京奇虎科技有限公司 | Data query method and device |
CN109213829A (en) * | 2017-06-30 | 2019-01-15 | 北京国双科技有限公司 | Data query method and device |
CN107729399A (en) * | 2017-09-21 | 2018-02-23 | 北京京东尚科信息技术有限公司 | The method and apparatus of data processing |
CN107729399B (en) * | 2017-09-21 | 2020-06-05 | 北京京东尚科信息技术有限公司 | Data processing method and device |
CN110019357B (en) * | 2017-09-29 | 2021-06-29 | 北京国双科技有限公司 | Database query script generation method and device |
CN108280046A (en) * | 2017-11-30 | 2018-07-13 | 深圳市科列技术股份有限公司 | A kind of method, battery data server and the user terminal of battery data processing |
CN108334554A (en) * | 2017-12-29 | 2018-07-27 | 上海跬智信息技术有限公司 | A kind of novel OLAP precomputations model and construction method |
CN108334554B (en) * | 2017-12-29 | 2021-10-01 | 上海跬智信息技术有限公司 | Novel OLAP pre-calculation model and construction method |
CN108875008A (en) * | 2018-06-15 | 2018-11-23 | 湖北德普电气股份有限公司 | A kind of Large Volume Data analysis method and device |
CN110781228B (en) * | 2018-07-13 | 2021-07-13 | 马上消费金融股份有限公司 | Data query method, system and device based on data warehouse and storage medium |
CN110781228A (en) * | 2018-07-13 | 2020-02-11 | 马上消费金融股份有限公司 | Data query method, system and device based on data warehouse and storage medium |
CN109241159B (en) * | 2018-08-07 | 2021-07-23 | 威富通科技有限公司 | Partition query method and system for data cube and terminal equipment |
CN109241159A (en) * | 2018-08-07 | 2019-01-18 | 威富通科技有限公司 | A kind of subregion querying method, system and the terminal device of data cube |
CN109308301A (en) * | 2018-09-28 | 2019-02-05 | 中国银行股份有限公司 | The preparation method and device of test data |
CN111061758A (en) * | 2018-10-16 | 2020-04-24 | 杭州海康威视数字技术股份有限公司 | Data storage method, device and storage medium |
CN111061758B (en) * | 2018-10-16 | 2023-10-20 | 杭州海康威视数字技术股份有限公司 | Data storage method, device and storage medium |
CN109710859A (en) * | 2019-01-21 | 2019-05-03 | 北京字节跳动网络技术有限公司 | Data query method and apparatus |
CN109885609A (en) * | 2019-01-31 | 2019-06-14 | 平安科技(深圳)有限公司 | Based on combined data area control method, device and storage medium |
CN109976910A (en) * | 2019-03-20 | 2019-07-05 | 跬云(上海)信息科技有限公司 | Querying method and device based on precomputation OLAP model |
CN109992594A (en) * | 2019-03-22 | 2019-07-09 | 跬云(上海)信息科技有限公司 | Distributed based on precomputation optimization executes optimization method and system |
CN110008239A (en) * | 2019-03-22 | 2019-07-12 | 跬云(上海)信息科技有限公司 | Logic based on precomputation optimization executes optimization method and system |
CN110110165A (en) * | 2019-04-01 | 2019-08-09 | 跬云(上海)信息科技有限公司 | Dynamic routing method and device for query engine in precomputation system |
WO2020199832A1 (en) * | 2019-04-01 | 2020-10-08 | 跬云(上海)信息科技有限公司 | Dynamic routing method and apparatus for query engine in pre-computation system |
US11397734B2 (en) * | 2019-04-01 | 2022-07-26 | Kuyun (Shanghai) Information Technology Co., Ltd. | Dynamic routing method and apparatus for query engine in pre-computing system |
CN110222124A (en) * | 2019-05-08 | 2019-09-10 | 跬云(上海)信息科技有限公司 | Multidimensional data processing method and system based on OLAP |
CN110275920A (en) * | 2019-06-27 | 2019-09-24 | 中国石油集团东方地球物理勘探有限责任公司 | Data query method, apparatus, electronic equipment and computer readable storage medium |
CN110275920B (en) * | 2019-06-27 | 2021-08-03 | 中国石油集团东方地球物理勘探有限责任公司 | Data query method and device, electronic equipment and computer readable storage medium |
CN110674117A (en) * | 2019-09-26 | 2020-01-10 | 京东数字科技控股有限公司 | Data modeling method and device, computer readable medium and electronic equipment |
CN110704514A (en) * | 2019-10-25 | 2020-01-17 | 南京录信软件技术有限公司 | Precomputation method based on Lucene |
CN112835966A (en) * | 2019-11-22 | 2021-05-25 | 北京金山云网络技术有限公司 | Data query method and device and electronic equipment |
CN111125264A (en) * | 2019-12-12 | 2020-05-08 | 跬云(上海)信息科技有限公司 | Extra-large set analysis method and device based on extended OLAP model |
CN111125266A (en) * | 2019-12-24 | 2020-05-08 | 中国建设银行股份有限公司 | Data processing method, device, equipment and storage medium |
CN111125266B (en) * | 2019-12-24 | 2024-01-12 | 建信金融科技有限责任公司 | Data processing method, device, equipment and storage medium |
CN111563123B (en) * | 2020-05-07 | 2023-08-22 | 北京首汽智行科技有限公司 | Real-time synchronization method for hive warehouse metadata |
CN111563123A (en) * | 2020-05-07 | 2020-08-21 | 北京首汽智行科技有限公司 | Live warehouse metadata real-time synchronization method |
CN112100177A (en) * | 2020-09-04 | 2020-12-18 | 北京三快在线科技有限公司 | Data storage method and device, computer equipment and storage medium |
CN113094409A (en) * | 2021-04-08 | 2021-07-09 | 国网电子商务有限公司 | Service data processing method and device and computer storage medium |
CN113220719A (en) * | 2021-06-04 | 2021-08-06 | 上海天旦网络科技发展有限公司 | Mass dimension data association query optimization method and system |
WO2023273183A1 (en) * | 2021-06-30 | 2023-01-05 | 北京邮电大学 | Hybrid engine-based multidimensional data query method and apparatus |
CN113407587A (en) * | 2021-07-19 | 2021-09-17 | 北京百度网讯科技有限公司 | Data processing method, device and equipment for online analysis processing engine |
CN113407587B (en) * | 2021-07-19 | 2023-10-27 | 北京百度网讯科技有限公司 | Data processing method, device and equipment for online analysis processing engine |
WO2023155273A1 (en) * | 2022-02-15 | 2023-08-24 | 上海跬智信息技术有限公司 | Correlation coefficient calculation method and apparatus, device, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106484875B (en) | 2019-12-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106484875A (en) | Data processing method based on MOLAP and device | |
CN104598647B (en) | A kind of tree graph search and the method for matching article | |
US20140025533A1 (en) | Method and Apparatus of Recommending Clothing Products | |
CN106528787B (en) | query method and device based on multidimensional analysis of mass data | |
EP2778960A1 (en) | Generating insightfull connections between graph entities | |
CN102270232B (en) | Semantic data query system with optimized storage | |
CN107016001A (en) | A kind of data query method and device | |
CN107943952A (en) | A kind of implementation method that full-text search is carried out based on Spark frames | |
CN102411591A (en) | Method and equipment for processing information | |
CN104866474A (en) | Personalized data searching method and device | |
KR102654884B1 (en) | Method for providing interactive information service and apparatus therefor | |
CN110909182A (en) | Multimedia resource searching method and device, computer equipment and storage medium | |
CN103377232A (en) | Headline keyword recommendation method and system | |
CN102298650B (en) | Distributed recommendation method of massive digital information | |
CN102737123B (en) | A kind of multidimensional data distribution method | |
CN102662986A (en) | System and method for microblog message retrieval | |
CN103488672B (en) | Optimization standard of comparison is created in associative storage | |
CN100530192C (en) | Text searching method and device | |
CN106484889A (en) | The flooding method and apparatus of Internet resources | |
Rodríguez-Arauz et al. | Design of a multimedia data management system that uses horizontal fragmentation to optimize content-based queries | |
CN111400301B (en) | Data query method, device and equipment | |
US8825698B1 (en) | Showing prominent users for information retrieval requests | |
CN103164407A (en) | Information searching method and system | |
CN110959157A (en) | Accelerating large-scale similarity calculations | |
CN111143582B (en) | Multimedia resource recommendation method and device for updating association words in double indexes in real time |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 |