CN104516906A - Adaptive indexing method and device - Google Patents
Adaptive indexing method and device Download PDFInfo
- Publication number
- CN104516906A CN104516906A CN201310456838.7A CN201310456838A CN104516906A CN 104516906 A CN104516906 A CN 104516906A CN 201310456838 A CN201310456838 A CN 201310456838A CN 104516906 A CN104516906 A CN 104516906A
- Authority
- CN
- China
- Prior art keywords
- result data
- block
- data block
- blocks
- data
- 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.)
- Pending
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an adaptive indexing method and an adaptive indexing device, and belongs to the database technology field. The adaptive indexing method includes: receiving an query request which carries range query conditions; obtaining at least one first result data block corresponding to the query request according to the range query conditions; sorting data in the at least one first result data block corresponding to the range query conditions, which is disordered and meets a preset price condition; updating a result data block set according to the result data blocks which are sorted and the other result data blocks which are not sorted, and updating an index of the result data block set. By using the adaptive indexing method and the adaptive indexing device, convergence rate of a database is improved on the premise that resource consumption is reduced during the early query process of the database.
Description
Technical field
The present invention relates to database technical field, particularly a kind of method and apparatus of adaptive index.
Background technology
Along with the development of database technology, adaptive index technology is widely used.Adaptive index technology is in a database, sets up index by query script, and by the continuous renolation index of continuous query script, to improve the technology of search efficiency.The type of inquiring about in database can comprise range data inquiry, fuzzy data inquiry etc., generally in adaptive index technology, index be range data inquiry process in set up and perfect.
Article Merging What ' s Cracked, Cracking What ' s Merged:adaptive indexing inmain memory column-stores, VLDB, the 2011 HCC(Hyper Crack Crack proposed, divide fast) algorithm is a kind of adaptive index technology.Based on HCC algorithm, at the building database initial stage, can be multiple initial data block by Data Placement by the rule pre-established, composition primary data set of blocks.When receiving the inquiry request of carrying range query condition, adopting crack(division) method inquires about, the data meeting range query condition are proposed from corresponding initial data block, simultaneously to data line division process remaining in this initial data block, the data being about to be greater than range query condition and the data being less than range query condition are placed in different data blocks respectively, obtain two new initial data block.Then the data meeting range query condition in each initial data block crack method inquired are merged into a result data block, the result data block composition result data set of blocks of each inquiry.Result data block in result data set of blocks is set up and has index, as Adelson-Velskii-Landis tree (a kind of binary tree is named with the name abbreviation of presenter) etc., the data area of each result data block can be recorded in the index.In follow-up query script, can inquire about respectively in result data set of blocks and primary data set of blocks, when carrying out range query in result data set of blocks, also adopt crack method to inquire about.
Article Merging What ' s Cracked, Cracking What ' s Merged:adaptive indexing inmain memory column-stores, VLDB, the 2011 HCS(Hyper Crack Sort proposed, divide sequence fast) algorithm is also a kind of adaptive index technology.With HCC technology type seemingly, at the building database initial stage, can be multiple initial data block by Data Placement by the rule pre-established, composition primary data set of blocks.In HCS algorithm, same employing crack method is inquired about in primary data set of blocks, then the data that crack method inquires are merged into a result data block, unlike, in HCS algorithm, first data sorting in block is carried out to the result data block inquired, then put into result data set of blocks.In follow-up query script, can inquire about respectively in result data set of blocks and primary data set of blocks, in HCS algorithm, because data are orderly in result data block, so when inquiring about in result data set of blocks, split can be adopted to inquire about.
In HCC algorithm and HCS algorithm, according to inquiring about the result data block obtained, result data set of blocks can be upgraded, concrete, if do not comprise the data that certain inquires about the result data block obtained in result data set of blocks, then this result data block is added in result data set of blocks.Like this, based on HCC algorithm and HCS algorithm, along with database is constantly used, result data set of blocks can constantly increase, index is constantly updated with the renewal of result data set of blocks, and ratio data that can be indexed in database constantly promotes, so index can be more and more perfect, corresponding search efficiency also can constantly promote, and this process can be called the convergence process of database.
Realizing in process of the present invention, inventor finds that prior art at least exists following problem:
Adopt the database of HCC algorithm, inquired about by range data, constantly update result data set of blocks, constantly update the index of result data set of blocks simultaneously, to promote the search efficiency of database gradually, but, although by the data area of each result data block recorded in index, the efficiency of positioning result data block can be improved, but, after navigating to result data block by index, when carrying out data query further in block, the efficiency of inquiry is still comparatively low, this causes adopting the speed of convergence of the database of HCC algorithm slower, often can not meet the demand of application.In addition, adopt the database of HCS algorithm, although speed of convergence is very fast, in the query script that database carries out in earlier stage, because need to sort to result data block, and cause the resource of query processing (as time, process resource etc.) to expend serious.
Summary of the invention
In order to solve the problem of prior art, embodiments providing a kind of method and apparatus of adaptive index, with while reducing the resource cost in database query script in early stage, improving the speed of convergence of database.Described technical scheme is as follows:
First aspect, provides a kind of method of adaptive index, and the result data set of blocks of database is set up index, and described method comprises:
Receive the inquiry request carrying range query condition;
According to described range query condition, obtain at least one first result data block that described inquiry request is corresponding;
In the first result data block that described inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block;
According to the first result data block through sequence, and without the first result data block sorted, result data set of blocks is upgraded, and upgrades the index of described result data set of blocks.
Second aspect, provide a kind of data enquire method adopting the method for adaptive index as above, described method comprises:
Data in the first corresponding for described inquiry request result data block are fed back as Query Result.
The third aspect, provides a kind of device of adaptive index, and the result data set of blocks of database is set up index, and described device comprises:
Receiver module, for receiving the inquiry request carrying range query condition;
Acquisition module, for according to described range query condition, obtains at least one first result data block that described inquiry request is corresponding;
Order module, in the first result data block that described inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block;
Update module, for the first result data block according to process sequence, and without the first result data block sorted, upgrades result data set of blocks, and upgrades the index of described result data set of blocks.
Fourth aspect, provides a kind of data query arrangement, and described data query arrangement comprises the device of adaptive index as above, and described data query arrangement also comprises:
Feedback module, for feeding back the data in the first corresponding for described inquiry request result data block as Query Result.
The beneficial effect that the technical scheme that the embodiment of the present invention provides is brought is:
In the embodiment of the present invention, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the method flow diagram of the adaptive index that the embodiment of the present invention provides;
Fig. 2 is the schematic diagram of the processing procedure of the application example that the embodiment of the present invention provides;
Fig. 3 is the apparatus structure schematic diagram of the adaptive index that the embodiment of the present invention provides.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiment of the present invention is described further in detail.
Embodiment one
Embodiments provide a kind of method of adaptive index, the result data set of blocks of database is set up index, and as shown in Figure 1, the treatment scheme volume of the method can comprise following step:
Step 101, receives the inquiry request carrying range query condition.
Step 102, according to this range query condition, obtains at least one first result data block that inquiry request is corresponding.
Step 103, in the first result data block that inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block.
Step 104, according to the first result data block through sequence, and without the first result data block sorted, upgrades result data set of blocks, and upgrades the index of described result data set of blocks.
In inventive embodiments, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
Embodiment two
Embodiments provide a kind of method of adaptive index, the result data set of blocks of database is set up index, the data area of each result data block in result data set of blocks can be recorded in index, preferably, the index of Adelson-Velskii-Landis tree data block set as a result can be adopted, the data area of each result data block of each leaf node corresponding record of Adelson-Velskii-Landis tree.The executive agent of the method can be set up the server that there is database or terminal device etc.
Below in conjunction with concrete embodiment, be explained in detail the treatment scheme shown in Fig. 1, content can be as follows:
Step 101, receives the inquiry request carrying range query condition.
Wherein, range query condition is the querying condition of the data in a certain data area of inquiry, and such as, range query condition can be greater than a to be less than b, also can be greater than c, etc.
In force, when carrying out large-scale data and analyzing, can to data building database to be analyzed, analyst can arrange range query condition according to the demand of data analysis, and send corresponding inquiry request to database, such as, analyst can total data in Query Database within the scope of 1-100, corresponding range query condition can be [1,100].
Or, certain application program can at terminal local building database, user is in the process using this application program, range data inquiry can be carried out in terminal local, send the inquiry request carrying certain range query condition to terminal, such as, user is when using the application of certain instant messaging, can inquire about the good friend of age within 20-30 year, corresponding range query condition can be [20,30].
Step 102, according to this range query condition, obtains at least one first result data block that inquiry request is corresponding.
Wherein, first result data block corresponding to inquiry request is the data block of the data composition of the range query condition meeting this inquiry request, this step can get a first result data block, also multiple first result data block can be got, inquire about in each first result data block obtained, original result data block (i.e. the direct result data block obtained in result data set of blocks) can be comprised, also can comprise newly-generated result data block.
Concrete, according to range query condition, data query can be carried out respectively in the primary data set of blocks and/or result data set of blocks of database, obtain the first result data block that inquiry request is corresponding.At the Database initial stage, also do not carry out inquiry, in database, only had primary data set of blocks, data block of coming to nothing set, at this moment, can only inquire about in primary data set of blocks.In the mid-term that database runs, both comprised primary data set of blocks in database, comprised again result data set of blocks, all can inquire about in primary data set of blocks and result data set of blocks.In the later stage that database runs, inquire about through a large amount of range data, data in primary data set of blocks have all been transferred in the conjunction of result set of data blocks by range data query script, result data set of blocks is only had in database, there is no primary data set of blocks, at this moment, can only inquire about in result data set of blocks.Along with the increase of result data set of blocks proportion in a database, the search efficiency of database can promote gradually.
There is not common factor in the data area of primary data set of blocks and result data set of blocks, the data area of result data set of blocks can be recorded, according to the data area of result data set of blocks and the relation (query context refers to the data area of range query condition) of query context, determine whether to inquire about in result data set of blocks, and determine whether to inquire about in primary data set of blocks, accordingly, when performing step 102, different process can be carried out in the following several ways:
Process one, if the data area of result data set of blocks comprises query context completely, then according to the data meeting range query condition in result data set of blocks, determines at least one first result data block that inquiry request is corresponding.
Wherein, result data set of blocks comprises the result data block obtained when carrying out range data inquiry to database, and result data set of blocks can upgrade in the process of range data inquiry.Result data set of blocks can set up the index of corresponding wherein each result data block, can record the data area of each result data block in index.Preferably, the index of Adelson-Velskii-Landis tree data block set as a result can be adopted, in Adelson-Velskii-Landis tree, the corresponding result data block of each leaf node, each leaf node carries out arranging (according to the foundation of result data set of blocks and the feature of renewal process, the data area of each result data block there will not be overlap) according to the order of the data area of corresponding result data block.
In force, the data area of result data set of blocks and query context can be compared, if this query context is completely in the data area of result data set of blocks, then illustrate that the data that will inquire about are all in result data set of blocks, the data that primary data set of blocks will not inquired about, so, can process according to process one, only result data set of blocks be inquired about.The data area of result data set of blocks can obtain in the index of result data set of blocks (as Adelson-Velskii-Landis tree).
Concrete, different according to data area of result data block each in result data set of blocks and the relation of query context, can obtain the first result data block that inquiry request is corresponding in different ways, specifically can be as follows:
Situation 1, for each result data block in result data set of blocks, a part of data in if block are in query context, then in decision block, whether data are orderly, in if block, data are unordered, then use division crack method in block, inquire about the data meeting range query condition, the first result data block that composition inquiry request is corresponding, in if block, data are orderly, then according to the order of data in block, inquiry meets the data of range query condition, the first result data block that composition inquiry request is corresponding.
In force, by the data area of result data block compared with query context, if the latter comprises the former part, then can judge that a part of data of this result data block are in query context.The data area of result data block can obtain in the index of result data set of blocks, have recorded the data area of each result data block in result data set of blocks in this index.
Such as, result data set of blocks comprises scope for [1,10], [11,20], the result data block of [21,30], [31,40], range query condition is [5,35], so, scope is [1,10], the result data block of [31,40] belongs to situation 1, and scope is [11,20], the result data block of [21,30] belongs to situation 2 below.
For this kind of result data block of situation 1, because both comprised the data meeting range query condition in result data block, comprised again the data not meeting range query condition, so, need in block, carry out inquiring about the data determining to meet range query condition.
The determination methods that in block, whether data orderly can have multiple, preferably, can in the block of tense marker this result data block that carries out result data block sorting data orderly, the result data block not having this to mark then can determine that in its block, data are unordered.
When using crack method to carry out inquiring about in block to certain result data block, the data meeting range query condition in block are proposed from result data block, division process is carried out to remaining data in this result data block simultaneously, the data being about to be greater than range query condition are placed in different data blocks with the data being less than range query condition respectively, obtain the result data block that two (or) is new, the result data block of the data composition inquiry request of the range query condition that meets of proposition.Like this, can obtain two or three result data blocks, one of them is the first result data block of corresponding inquiry request.
The method meeting the data of range query condition according to the sequential query of data in block can have multiple, preferably, split can be adopted to inquire about, concrete, first, the position of boundary value in block of split query context querying condition can be adopted, then, according to the order of data in the position of boundary value and block, the data meeting range query condition in block can be determined.
Such as, the data area of result data block is [0,12], in block, data sequence is for increasing progressively, and the data area of range query condition is [8,40], the intermediate value 6 first can getting 0 and 12 compares with 8,6 are less than 8, then continue 6 with 12 intermediate value 9 compare with 8, the rest may be inferred, until determine 8 positions in this result data block, because data sequence is for increasing progressively, so to get from the position of 8 all data backward, as the data meeting range query condition in this result data block.
Situation 2, for each result data block in result data set of blocks, the total data in if block all in query context, then the first result data block that it can be used as inquiry request corresponding.
In force, by the data area of result data block compared with query context, if the latter comprises the former completely, then can judge that the total data of this result data block is all in query context.
For this kind of result data block of situation 2, because its data are all in query context, so, data query can not be carried out in block, obtain the first result data block that total data composition inquiry request in block is corresponding, namely direct using this type of result data block as the first result data block corresponding to inquiry request.
Process two, if the data area of result data set of blocks does not comprise query context completely, then according to the data meeting range query condition in primary data set of blocks, determines at least one first result data block that inquiry request is corresponding.
Wherein, comprise one or more initial data block in primary data set of blocks, when Database, each initial data block is carried out grouping by raw data by the rule pre-established and is obtained.In process two, according to the data meeting range query condition inquired in primary data set of blocks, the number of the result data block of the inquiry request determined is preferably one.
In force, the data area of result data set of blocks and query context can be compared, if both do not occur simultaneously, then illustrate in result data set of blocks and do not comprise the data meeting range query condition, and the data meeting range query condition in primary data set of blocks, may be comprised, so, can process by process two, only primary data set of blocks be inquired about.
Concrete, in process two, in primary data set of blocks, the inquiry of crack method can be used to meet the data of range query condition, and the data inquired merged, obtain the first result data block that inquiry request is corresponding.
The inquiry of crack method can be carried out to each initial data block, the data meeting range query condition are proposed from corresponding initial data block, simultaneously to data line division process remaining in this initial data block, the data being about to be greater than range query condition are placed in different data blocks respectively with the data being less than range query condition, obtain the initial data block that two (or) is new.Then, all data meeting range query condition proposed are merged, obtain the first result data block of a corresponding inquiry request in primary data set of blocks.
Process three, if the data area of result data set of blocks comprises a part for query context, then according to the data meeting range query condition in result data set of blocks, determine at least one first result data block that inquiry request is corresponding, and, according to the data meeting range query condition in primary data set of blocks, determine at least one first result data block that inquiry request is corresponding, by at least one corresponding for the inquiry request determined in result data set of blocks first result data block, at least one first result data block corresponding with the inquiry request determined in primary data set of blocks, jointly as the first result data block that inquiry request is corresponding.
In force, the data area of result data set of blocks and query context can be compared, if the former comprises a part of scope of the latter, then illustrate in result data set of blocks and comprise the data meeting range query condition, also the data meeting range query condition may be comprised in primary data set of blocks, so, can process by process three, inquire about primary data set of blocks and result data set of blocks respectively, concrete processing procedure can distinguish the content in reference process one and process two.Then, by each first result data block determined in result data set of blocks and each first result data block determined in primary data set of blocks, all as the first result data block that inquiry request is corresponding.
Step 103, in the first result data block that inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block.
Wherein, the cost condition preset can be the requirement of the resource consumption to current queries and/or subsequent query, and this resource can be time, process resource (resource such as processor, internal memory) etc.
Concrete, the process of step 103 can be as follows:
First, in the first result data block that inquiry request is corresponding, the result data block that data in block are unordered is chosen, as the second result data block.
According to the cost condition judgment preset, data sorting in block is carried out to those the first result data blocks in this step 103, so the result data block can choosing data in block unordered carries out subsequent treatment, for the result data block that data are orderly, then sort without the need to judging whether again.The method that in decision block, whether data are orderly with reference to the content in above-mentioned process one, can be not repeated at this.
Then, current income corresponding to each second result data block and follow-up income is obtained.
Wherein, the current income that result data block is corresponding be when current do not carry out data sorting in block this inquiry stock number of saving, also be, this inquiry does not carry out data sorting in block to this result data block, with carry out in block compared with data sorting, the stock number can saved in this query script.The follow-up income that result data block is corresponding be when current carry out data sorting in block subsequent query (can be the query processing in follow-up one section of preset duration, as one month, one week etc.) stock number of saving, also be, this inquiry carries out data sorting in block to this result data block, with do not carry out in block compared with data sorting, the stock number altogether can saved in the query processing in follow-up one section of preset duration.
The relation of cost function u=F (T) for the stock number and processing time that represent consumption can be set, wherein, T is the processing time, u is cost value, for representing the stock number of consumption, this resource can be time or process resource etc., and this function can think linear function, i.e. F (T)=a*T ... (1)
Wherein, a is fixed coefficient (can by experiment data calculate), if this resource is the time, and so F (T)=T.
According to the definition of current income and follow-up income, they can be used respectively F (T) to be expressed as follows:
Current income: F (T
0+ T
c+ T
s)-F (T
0+ T
c) ... (2)
Follow-up income: p*t* (F (T
1+ T
c)-F (T
1+ T
b)) ... (3)
Wherein, T
cfor crack method query processing duration, namely data block is carried out to the processing time of crack method inquiry, T
bfor split query processing duration, namely data block is carried out to the processing time of split inquiry, T
sfor sequence handling duration, namely to the processing time that data block sorts, T
0for the processing time of the work in current queries process except the inquiry of crack method and sequence, T1 is the processing time of the work in follow-up each query script except the inquiry of crack method and split inquiry, p is the data query frequency of this result data block, t is for calculating later stage income and default duration, as one month or three months etc., follow-up income can think the stock number of saving in follow-up a period of time.The frequency that the data query frequency of result data block is queried for the data represented in this result data block, data query frequency is an estimated value.
Visible, F (T
0+ T
c+ T
s) can represent, for certain result data block, if carry out data sorting in block, the stock number that this query script consumes; F (T
0+ T
c) can represent, for this result data block, if do not carry out data sorting in block, the stock number that this query script consumes.They subtract each other, and represent, if this inquiry does not carry out data sorting in block to this result data block, then and the stock number that can save of this query script, the current income that namely this result data block is corresponding.
F (T
1+ T
c) can approximate representation, for certain result data block, if do not carry out data sorting in block, the stock number that follow-up each query script consumes; F (T
1+ T
b) can approximate representation, for this result data block, if carry out data sorting in block, the stock number that follow-up each query script consumes.They subtract each other, and represent, if carry out data sorting in block to this result data block, then and the stock number that can save of follow-up each query script.Again this difference is multiplied with p, t, namely obtains the follow-up income that this result data block is corresponding.
Concrete, the process obtaining current income corresponding to each second result data block can be as follows:
Obtain the sequence handling duration that each second result data block is corresponding; The sequence handling duration corresponding according to each second result data block, determines the current income that each second result data block is corresponding.
From formula (1), (2), current income can be expressed as a*T
s, a is fixed coefficient again, so when obtaining current income corresponding to certain result data block, and T that can be corresponding according to this result data block
sdetermine corresponding current income.
Concrete, the process obtaining follow-up income corresponding to each second result data block can be as follows:
Obtain the query processing duration that each second result data block is corresponding; Obtain the data query frequency of each second result data block; The query processing duration corresponding according to each second result data block, and the data query frequency of each second result data block, determine the follow-up income that each second result data block is corresponding.
Wherein, query processing duration can comprise crack method query processing duration and split query processing duration.
From formula (1), (3), follow-up income can be expressed as p*t*a* (T
c-T
b), a is fixed coefficient again, and t is default value, so when obtaining follow-up income corresponding to certain result data block, and T that can be corresponding according to this result data block
c, T
bcorresponding follow-up income is determined with p.
Preferably, in the process above, according to the data amount check (n) of each second result data block, the sequence handling duration that each second result data block is corresponding can be determined; According to the data amount check of each second result data block, determine the query processing duration that each second result data block is corresponding, can be specifically, according to the data amount check of each second result data block, determine the crack method query processing duration that each second result data block is corresponding, according to the data amount check of each second result data block, determine the split query processing duration that each second result data block is corresponding.
Because be the result data block of n for data amount check, the computation complexity carrying out the inquiry of crack method is directly proportional to n, carries out computation complexity and the lg of split inquiry
2n is directly proportional, and carries out the computation complexity that sorts and n*lg
2n is directly proportional, and computation complexity was directly proportional to the processing time again, so, following relational expression can be drawn:
T
C=A*n、T
B=B*lg
2n、T
S=C*n*lg
2n……………………（4）
Wherein, A, B, C are fixed coefficients, are determined by the hardware environment of database.The value of A, B, C can be determined by experiment, respectively the inquiry of crack method, split inquiry and sequence three kinds of processing procedures be tested, n, T then experimentally in data
c, T
b, T
sdetermine the value of A, B, C.
Preferably, the mode obtaining the data query frequency of result data block can have multiple, and such as, can adopt default default value (as 0.5), this default value can be an empirical value; Or the frequency that the inquiry relating to data within the scope of this result data blocks of data in the inquiry before can adding up occurs, by the data query frequency of this frequency data block as a result; Or the mean value of the frequency that each data are queried in this result data block in the inquiry before can adding up, by the data query frequency of this mean value data block as a result.Preferably, at the Database initial stage, historical query number of times is less, can adopt the data query frequency of default value data block as a result, when the inquiry times of database reaches some, the data query frequency of the method determination result data block of statistics can be adopted.
According to various above, current income and follow-up income can be expressed as follows respectively:
Current income is a*C*n*lg
2n, follow-up income is p*t*a*(A*n-B*lg
2n).
Again, calculate the preset relation between current income corresponding to each second result data block and follow-up income, and in all second result data blocks, choose preset relation and meet pre-conditioned result data block, as the 3rd result data block.
Wherein, this cost condition choosing in process is the requirement to the preset relation between current income corresponding to result data block and follow-up income.Preset relation can be the calculated relationship arbitrarily such as proportionate relationship, difference relationship or multiplication relationship, and such as, preset relation can be
preset relation can be arranged according to the actual requirements.
In force, can arrange pre-conditioned according to the actual requirements, can represent pre-conditioned by the form such as equation or inequality, such as, continue the example using above-mentioned preset relation, when comparing the follow-up income of emphasis, can arrange and require that this preset relation meets as lower inequality:
Again such as, when comparing emphasis delayed credits, can arrange and require that this preset relation meets as lower inequality:
Wherein, α, β can be default values, and span is greater than 0 and is less than 1, can arrange arbitrarily according to demand, and it is higher that the numerical value of α is arranged, and represents higher to the requirement of follow-up income, and it is lower that the numerical value of β is arranged, and represents higher to the requirement of delayed credits.
By the expression formula of the above-mentioned current income that draws and follow-up income, namely current income is a*C*n*lg
2n, follow-up income is p*t*a*(A*n-B*lg
2n), substitute into equation or the inequality of pre-conditioned correspondence, substitute into n and p of certain result data block again, whether can set up according to equation or inequality, judge whether the preset relation between the current income that this result data block is corresponding and follow-up income meets pre-conditioned.
To substitute into (5) formula, can obtain as lower inequality:
I.e. (1-α) * p*t* (A*n-B*lg
2n) > α * C*n*lg
2n ... (7)
If inequality (7) is set up, then relation can meet the pre-conditioned of (5) formula between current income corresponding to determination result data block and follow-up income.Pre-conditioned for other, similar method can be adopted to calculate and determine, the embodiment of the present invention is not repeated.
Finally, data sorting in block is carried out to each 3rd result data block.
The method of data sorting has a variety of, and arbitrary method can be adopted here according to demand to carry out data sorting, and the embodiment of the present invention does not limit this.
Preferably, carrying out in block after data sorting to the 3rd result data block, can record the 3rd result data block is the result data block that in block, data are orderly.
Process through above-mentioned sequence, in each first result data block that inquiry request is corresponding, may comprise through the result data block of sequence and/or the result data block without sequence.
Step 104, according to the first result data block through sequence, and without the first result data block sorted, upgrades result data set of blocks, and upgrades the index of described result data set of blocks.
Concrete, in each first result data block that inquiry request is corresponding, for the first result data block carrying out data sorting in block in this query script, if there is the result data block before the sequence that this first result data block is corresponding in result data set of blocks, then replace the result data block before this sequence with this first result data block, if there is no the result data block before the sequence that this first result data block is corresponding in result data set of blocks, then this first result data block is added in result data set of blocks; In each first result data block that inquiry request is corresponding, for the first result data block not carrying out data sorting in block in this query script, if do not comprise the data of this first result data block in result data set of blocks, then this first result data block is added in result data set of blocks.
When upgrading result data set of blocks, can based on the result data set of blocks upgraded, the index of result data set of blocks is upgraded, the index entry of corresponding newly-increased result data block can be increased in the index, record the data area of this newly-increased result data block.This index preferably can use Adelson-Velskii-Landis tree, and can carry out according to a conventional method the renewal process of Adelson-Velskii-Landis tree, the embodiment of the present invention does not limit this.
As shown in Figure 2, be an application example that the embodiment of the present invention provides.During building database, by the division rule pre-established, raw data is divided, obtain initial data block A1, A2, A3, A4, composition primary data set of blocks.
Inquiry one be Database after first time range data inquiry, its processing procedure can be as follows:
Step one, receives the first inquiry request carrying range query condition.This range query condition is " c to i ".
Step 2, uses crack method in primary data set of blocks, inquire about the data meeting this range query condition, and the data inquired is carried out merging and obtain result data block B1.
Step 3, through judging that determination result data block B1 does not meet the cost condition preset, so do not carry out data sorting in block to it.
Step 4, adds to result data block B1 in result data set of blocks.
The processing procedure of inquiry two can be as follows:
Step one, receives the second inquiry request carrying range query condition.This range query condition is " e to l ".
Step 2, use crack method in result data set of blocks, inquire about the data meeting this range query condition, obtain result data block B3, divide result data block B2, use crack method in primary data set of blocks, inquire about the data meeting this range query condition, the data inquired are carried out merging and obtain result data block B4.
Step 3, through judge determination result data block B3, B4 all meet preset cost condition and in block data unordered, so carry out data sorting in block to result data block B3, B4, obtain result data block B3 ', B4 '.
Step 4, adds to result data block B3 ', B4 ' in result data set of blocks.
In inventive embodiments, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
The embodiment of the present invention additionally provides a kind of data enquire method, the method of adaptive index as above is adopted in this data enquire method, except the process of above-mentioned flow process, this data enquire method also comprises following process: the data in the first corresponding for inquiry request result data block fed back as Query Result.
The step of this result feedback can perform after obtaining the first result data block corresponding to inquiry request, also can perform after result data set of blocks upgrades, can also perform after the index upgrade of result data set of blocks.
In inventive embodiments, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, the efficiency of inquiry can be improved simultaneously, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so, the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
Embodiment three
Based on identical technical conceive, the embodiment of the present invention additionally provides a kind of device of adaptive index, and the result data set of blocks of database is set up index, and as shown in Figure 3, described device comprises:
Receiver module 310, for receiving the inquiry request carrying range query condition;
Acquisition module 320, for according to described range query condition, obtains at least one first result data block that described inquiry request is corresponding;
Order module 330, in the first result data block that described inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block;
Update module 340, for the first result data block according to process sequence, and without the first result data block sorted, upgrades result data set of blocks, and upgrades the index of described result data set of blocks.
Preferably, described acquisition module 320, for:
If the data area of described result data set of blocks comprises query context completely, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding; Wherein, described query context is the data area of described range query condition;
If the data area of described result data set of blocks does not comprise described query context completely, then according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding;
If the data area of described result data set of blocks comprises a part for described query context, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding, and, according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding, by at least one corresponding for the described inquiry request determined in described result data set of blocks first result data block, at least one first result data block corresponding with the described inquiry request determined in described primary data set of blocks, jointly as the first result data block that described inquiry request is corresponding.
Preferably, described acquisition module 320, for:
For each result data block in described result data set of blocks, a part of data in if block are in described query context, then in decision block, whether data are orderly, in if block, data are unordered, then use division crack method in block, inquire about the data meeting described range query condition, form the first result data block that described inquiry request is corresponding, in if block, data are orderly, then according to the order of data in block, inquiry meets the data of described range query condition, forms the first result data block that described inquiry request is corresponding;
For each result data block in described result data set of blocks, the total data in if block all in described query context, then the first result data block that it can be used as described inquiry request corresponding.
Preferably, described acquisition module 320, for:
In primary data set of blocks, use the inquiry of crack method to meet the data of described range query condition, and the data inquired are merged, obtain the first result data block that described inquiry request is corresponding.
Preferably, described order module 330, for:
In the first result data block that described inquiry request is corresponding, choose the result data block that data in block are unordered, as the second result data block;
Obtain current income corresponding to each second result data block and follow-up income, calculate the preset relation between current income corresponding to each second result data block and follow-up income, and in all second result data blocks, choose preset relation and meet pre-conditioned result data block, as the 3rd result data block; Wherein, the current income that second result data block is corresponding be when current do not carry out data sorting in block this inquiry stock number of saving, the follow-up income that the second result data block is corresponding be when current carry out data sorting in block the subsequent query stock number of saving;
Data sorting in block is carried out to each 3rd result data block.
Preferably, described order module 330, for:
According to the data amount check of each second result data block, determine the sequence handling duration that each second result data block is corresponding;
The sequence handling duration corresponding according to described each second result data block, determines the current income that each second result data block is corresponding.
Preferably, described order module 330, for:
According to the data amount check of each second result data block, determine the query processing duration that each second result data block is corresponding;
Obtain the data query frequency of each second result data block;
The query processing duration corresponding according to described each second result data block, and the data query frequency of described each second result data block, determine the follow-up income that each second result data block is corresponding.
In inventive embodiments, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
Additionally provide a kind of data query arrangement in the embodiment of the present invention, described data query arrangement comprises the device of adaptive index as above, and described data query arrangement also comprises:
Feedback module, for feeding back the data in the first corresponding for described inquiry request result data block as Query Result.
In inventive embodiments, by the setting of cost condition, can only sort to the satisfactory result data block of cost of sequence process, at database in earlier stage, the resource cost in query script can be reduced, the efficiency of inquiry can be improved simultaneously, and, by sorting to the result data block meeting certain cost condition, progressively can make the result data block ordering in result data set of blocks, because a minor sort can reach the convergence effect dividing for several times and just can reach, can more effective lifting search efficiency, so, the speed of convergence of database can be improved.Thus, while reducing the resource cost in database query script in early stage, the speed of convergence of database can be improved.
It should be noted that: the device of the adaptive index that above-described embodiment provides is when upgrading index, only be illustrated with the division of above-mentioned each functional module, in practical application, can distribute as required and by above-mentioned functions and be completed by different functional modules, inner structure by device is divided into different functional modules, to complete all or part of function described above.In addition, the device of the adaptive index that above-described embodiment provides and the embodiment of the method for adaptive index belong to same design, and its specific implementation process refers to embodiment of the method, repeats no more here.
The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.
One of ordinary skill in the art will appreciate that all or part of step realizing above-described embodiment can have been come by hardware, the hardware that also can carry out instruction relevant by program completes, described program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium mentioned can be ROM (read-only memory), disk or CD etc.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (16)
1. a method for adaptive index, the result data set of blocks of database is set up index, and it is characterized in that, described method comprises:
Receive the inquiry request carrying range query condition;
According to described range query condition, obtain at least one first result data block that described inquiry request is corresponding;
In the first result data block that described inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block;
According to the first result data block through sequence, and without the first result data block sorted, result data set of blocks is upgraded, and upgrades the index of described result data set of blocks.
2. method according to claim 1, is characterized in that, described according to described range query condition, obtains at least one first result data block that described inquiry request is corresponding, comprising:
If the data area of described result data set of blocks comprises query context completely, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding; Wherein, described query context is the data area of described range query condition;
If the data area of described result data set of blocks does not comprise described query context completely, then according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding;
If the data area of described result data set of blocks comprises a part for described query context, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding, and, according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding, by at least one corresponding for the described inquiry request determined in described result data set of blocks first result data block, at least one first result data block corresponding with the described inquiry request determined in described primary data set of blocks, jointly as the first result data block that described inquiry request is corresponding.
3. method according to claim 2, is characterized in that, the described data according to meeting described range query condition in described result data set of blocks, determines to comprise at least one first result data block that described inquiry request is corresponding:
For each result data block in described result data set of blocks, a part of data in if block are in described query context, then in decision block, whether data are orderly, in if block, data are unordered, then use division crack method in block, inquire about the data meeting described range query condition, form the first result data block that described inquiry request is corresponding, in if block, data are orderly, then according to the order of data in block, inquiry meets the data of described range query condition, forms the first result data block that described inquiry request is corresponding;
For each result data block in described result data set of blocks, the total data in if block all in described query context, then the first result data block that it can be used as described inquiry request corresponding.
4. method according to claim 2, is characterized in that, the described data according to meeting described range query condition in primary data set of blocks, determines to comprise at least one first result data block that described inquiry request is corresponding:
In primary data set of blocks, use the inquiry of crack method to meet the data of described range query condition, and the data inquired are merged, obtain the first result data block that described inquiry request is corresponding.
5. method according to claim 1, is characterized in that, in the described first result data block corresponding in described inquiry request, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block, comprising:
In the first result data block that described inquiry request is corresponding, choose the result data block that data in block are unordered, as the second result data block;
Obtain current income corresponding to each second result data block and follow-up income, calculate the preset relation between current income corresponding to each second result data block and follow-up income, and in all second result data blocks, choose preset relation and meet pre-conditioned result data block, as the 3rd result data block; Wherein, the current income that second result data block is corresponding be when current do not carry out data sorting in block this inquiry stock number of saving, the follow-up income that the second result data block is corresponding be when current carry out data sorting in block the subsequent query stock number of saving;
Data sorting in block is carried out to each 3rd result data block.
6. method according to claim 5, is characterized in that, obtains the current income that each second result data block is corresponding, comprising:
According to the data amount check of each second result data block, determine the sequence handling duration that each second result data block is corresponding;
The sequence handling duration corresponding according to described each second result data block, determines the current income that each second result data block is corresponding.
7. method according to claim 5, is characterized in that, obtains the follow-up income that each second result data block is corresponding, comprising:
According to the data amount check of each second result data block, determine the query processing duration that each second result data block is corresponding;
Obtain the data query frequency of each second result data block;
The query processing duration corresponding according to described each second result data block, and the data query frequency of described each second result data block, determine the follow-up income that each second result data block is corresponding.
8. adopt a data enquire method for the method for the adaptive index as described in any one of claim 1-7, it is characterized in that, described method comprises:
Data in the first corresponding for described inquiry request result data block are fed back as Query Result.
9. a device for adaptive index, the result data set of blocks of database is set up index, and it is characterized in that, described device comprises:
Receiver module, for receiving the inquiry request carrying range query condition;
Acquisition module, for according to described range query condition, obtains at least one first result data block that described inquiry request is corresponding;
Order module, in the first result data block that described inquiry request is corresponding, unordered to data in block and the result data block of satisfied default cost condition carries out data sorting in block;
Update module, for the first result data block according to process sequence, and without the first result data block sorted, upgrades result data set of blocks, and upgrades the index of described result data set of blocks.
10. device according to claim 9, is characterized in that, described acquisition module, for:
If the data area of described result data set of blocks comprises query context completely, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding; Wherein, described query context is the data area of described range query condition;
If the data area of described result data set of blocks does not comprise described query context completely, then according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding;
If the data area of described result data set of blocks comprises a part for described query context, then according to the data meeting described range query condition in described result data set of blocks, determine at least one first result data block that described inquiry request is corresponding, and, according to the data meeting described range query condition in primary data set of blocks, determine at least one first result data block that described inquiry request is corresponding, by at least one corresponding for the described inquiry request determined in described result data set of blocks first result data block, at least one first result data block corresponding with the described inquiry request determined in described primary data set of blocks, jointly as the first result data block that described inquiry request is corresponding.
11. devices according to claim 10, is characterized in that, described acquisition module, for:
For each result data block in described result data set of blocks, a part of data in if block are in described query context, then in decision block, whether data are orderly, in if block, data are unordered, then use division crack method in block, inquire about the data meeting described range query condition, form the first result data block that described inquiry request is corresponding, in if block, data are orderly, then according to the order of data in block, inquiry meets the data of described range query condition, forms the first result data block that described inquiry request is corresponding;
For each result data block in described result data set of blocks, the total data in if block all in described query context, then the first result data block that it can be used as described inquiry request corresponding.
12. devices according to claim 10, is characterized in that, described acquisition module, for:
In primary data set of blocks, use the inquiry of crack method to meet the data of described range query condition, and the data inquired are merged, obtain the first result data block that described inquiry request is corresponding.
13. devices according to claim 9, is characterized in that, described order module, for:
In the first result data block that described inquiry request is corresponding, choose the result data block that data in block are unordered, as the second result data block;
Obtain current income corresponding to each second result data block and follow-up income, calculate the preset relation between current income corresponding to each second result data block and follow-up income, and in all second result data blocks, choose preset relation and meet pre-conditioned result data block, as the 3rd result data block; Wherein, the current income that second result data block is corresponding be when current do not carry out data sorting in block this inquiry stock number of saving, the follow-up income that the second result data block is corresponding be when current carry out data sorting in block the subsequent query stock number of saving;
Data sorting in block is carried out to each 3rd result data block.
14. devices according to claim 13, is characterized in that, described order module, for:
According to the data amount check of each second result data block, determine the sequence handling duration that each second result data block is corresponding;
The sequence handling duration corresponding according to described each second result data block, determines the current income that each second result data block is corresponding.
15. devices according to claim 13, is characterized in that, described order module, for:
According to the data amount check of each second result data block, determine the query processing duration that each second result data block is corresponding;
Obtain the data query frequency of each second result data block;
The query processing duration corresponding according to described each second result data block, and the data query frequency of described each second result data block, determine the follow-up income that each second result data block is corresponding.
16. 1 kinds of data query arrangement, it is characterized in that, described data query arrangement comprises the device of the adaptive index as described in any one of claim 9-15, described data query arrangement also comprises:
Feedback module, for feeding back the data in the first corresponding for described inquiry request result data block as Query Result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310456838.7A CN104516906A (en) | 2013-09-29 | 2013-09-29 | Adaptive indexing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310456838.7A CN104516906A (en) | 2013-09-29 | 2013-09-29 | Adaptive indexing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104516906A true CN104516906A (en) | 2015-04-15 |
Family
ID=52792217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310456838.7A Pending CN104516906A (en) | 2013-09-29 | 2013-09-29 | Adaptive indexing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104516906A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416027A (en) * | 2018-03-09 | 2018-08-17 | 广西师范大学 | A kind of bottom-up merging data fragmentation optimization method based on range query boundary set |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101379463A (en) * | 2005-12-20 | 2009-03-04 | 贝克曼库尔特公司 | Adaptable database system |
| US20100281013A1 (en) * | 2009-04-30 | 2010-11-04 | Hewlett-Packard Development Company, L.P. | Adaptive merging in database indexes |
-
2013
- 2013-09-29 CN CN201310456838.7A patent/CN104516906A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101379463A (en) * | 2005-12-20 | 2009-03-04 | 贝克曼库尔特公司 | Adaptable database system |
| US20100281013A1 (en) * | 2009-04-30 | 2010-11-04 | Hewlett-Packard Development Company, L.P. | Adaptive merging in database indexes |
Non-Patent Citations (2)
| Title |
|---|
| ZHONGBIN XUE等: "Optimized Adaptive Hybrid Indexing for In-memory Column Stores", 《DATABASE SYSTEMS FOR ADVANCED APPLICATIONS,18TH INTERNATIONAL CONFERENCE, DASFAA 2013》 * |
| 房俊华等: "一种具有自适应机制的闪存数据库索引结构", 《计算机应用》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416027A (en) * | 2018-03-09 | 2018-08-17 | 广西师范大学 | A kind of bottom-up merging data fragmentation optimization method based on range query boundary set |
| CN108416027B (en) * | 2018-03-09 | 2021-07-20 | 广西师范大学 | Merged data fragmentation optimization method based on range query boundary set |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110418200A (en) | A kind of video recommendation method, device and terminal device | |
| CN103559300B (en) | The querying method and inquiry unit of data | |
| CN105550225B (en) | Index structuring method, querying method and device | |
| CN107229730A (en) | Data query method and device | |
| CN107357846A (en) | The methods of exhibiting and device of relation map | |
| CN103309894B (en) | Based on search implementation method and the system of user property | |
| CN103942220A (en) | Method used for intelligently linking work orders with knowledge of knowledge base and suitable for IT operation and maintenance system | |
| CN110008199A (en) | A kind of Data Migration dispositions method based on access temperature | |
| WO2021128523A1 (en) | Technology readiness level determination method and system based on science and technology big data | |
| CN103455560A (en) | Data query method and system | |
| CN110727690B (en) | Data updating method | |
| CN111538901A (en) | Article recommendation method and device, server and storage medium | |
| CN105069117A (en) | Data flow efficiency improving method based on storage process | |
| CN102915344B (en) | SQL (structured query language) statement processing method and device | |
| US10963963B2 (en) | Rule based hierarchical configuration | |
| CN112365070A (en) | Power load prediction method, device, equipment and readable storage medium | |
| CN110825526B (en) | Distributed scheduling method and device based on ER relationship, equipment and storage medium | |
| CN102567536A (en) | Key performance target analyzing method based on data statistics | |
| CN112419124B (en) | Method and device for quickly identifying low-efficiency industrial land and storage medium thereof | |
| CN108255852A (en) | SQL performs method and device | |
| US9135302B2 (en) | Query rewrite with a nested materialized view | |
| CN104516906A (en) | Adaptive indexing method and device | |
| CN104778253B (en) | A kind of method and apparatus that data are provided | |
| CN106547774B (en) | Website content detection method and device | |
| CN104008178A (en) | Processing method and system for dynamic loading of data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150415 |