CN106250492A - The processing method and processing device of index - Google Patents
The processing method and processing device of index Download PDFInfo
- Publication number
- CN106250492A CN106250492A CN201610623529.8A CN201610623529A CN106250492A CN 106250492 A CN106250492 A CN 106250492A CN 201610623529 A CN201610623529 A CN 201610623529A CN 106250492 A CN106250492 A CN 106250492A
- Authority
- CN
- China
- Prior art keywords
- thread
- kernel
- index
- subindex
- reading
- 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
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
-
- 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
- G06F16/93—Document management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides the processing method and processing device of a kind of index, by all kernels of server are divided into two kernel set, limit reading thread and run on the kernel in the first kernel set, and limit and write thread and run on the kernel in the second kernel set, avoid and frequently revise the data in cache owing to writing thread, and cause for reading the problem that the data hit rate that stored in cache for thread is low, improve reading thread when needs read the first index, from cache, read the probability of the first index, thus improve search efficiency.
Description
Technical field
The present invention relates to Internet technical field, particularly relate to the processing method and processing device of a kind of index.
Background technology
Server can provide the user release news function and Query Information function.Generally, server is to pass through rope
The mode drawn carrys out management document, and wherein, index indicates the document corresponding to each key word.
In prior art, server uses different threads provide the user Query Information function respectively and release news
Function.Concrete, provide the user Query Information function by reading thread, provide the user, by writing thread, the merit that releases news
Energy.For Query Information function, reading thread needs to determine, according to index, the key word that the search sequence with user's input is mated,
And according to determined by key word to user return correspondence document, therefore read thread need index is carried out read operation.For
Release news function, writes thread it needs to be determined that key word in the document issued of user, closes in the index determined by write
Keyword and the corresponding relation of the document issued, therefore write thread needs and index carried out write operation.Wherein, thread is being read to rope
When introducing row read operation, first determine whether to run in the cache memory (cache) of the kernel of this reading thread and whether save
This index；The most directly from cache, read this index；Otherwise, from internal memory, read this index, also will read simultaneously
This index be stored in cache so that the later reading to this index can directly obtain from cache.Writing thread pair
Index is when carrying out write operation, and the data being written into are written to run in the cache of this kernel writing thread.
But, in prior art, there is the problem that search efficiency is relatively low.
Summary of the invention
The present invention provides the processing method and processing device of a kind of index, in order to solve relatively low the asking of search efficiency in prior art
Topic.
First aspect, the present invention provides the processing method of a kind of index, described method to be applied to server, described server
In all kernels be divided into two kernel set；Wherein, read thread and operate on the kernel in the first kernel set, write line
Journey operates on the kernel in the second kernel set, and described reading thread is used for providing the user Query Information function, described in write line
Journey is used for providing the user the function that releases news；Described method includes:
Described reading thread reads the first index, and for indicating, key word is corresponding with document identification to close described first index
System；
Described reading thread, according to described first index and the search sequence of user's input, determines corresponding with described search sequence
Document identification.
In a kind of possible design, described first index includes N number of subindex, and N is the integer more than 0；Described N number of son
Each subindex correspondence K in index reads thread, and K is the integer more than or equal to 0；
Accordingly, described reading thread reads the first index, including: first reads thread reads the first subindex；Wherein, described
First reading thread is the reading thread that described first subindex is corresponding.
In a kind of possible design, all kernels in described first kernel set are divided into N number of kernel subclass,
Each kernel subclass in described N number of kernel subclass and each subindex one_to_one corresponding described；Described each height
What index was corresponding reads on the kernel that thread operates in the kernel subclass that each subindex described is corresponding.
In a kind of possible design, before described reading index, also include:
Estimate the interior check figure needed for each subindex described and Thread Count；
According to the interior check figure needed for each subindex described, determine the kernel subset that each subindex described is corresponding
Close；
According to the Thread Count needed for each subindex described, determine the reading thread that each subindex described is corresponding.
In a kind of possible design, described method also includes: described in write the information to be released that thread inputs according to user,
Second index is updated.
Second aspect, the present invention provides the processing means of a kind of index, described device to be server, in described server
All kernels are divided into two kernel set；Wherein, read thread and operate on the kernel in the first kernel set, write thread fortune
On row kernel in the second kernel set, described reading thread is used for providing the user Query Information function, described in write thread and use
In providing the user the function that releases news；Described device includes:
First processing module of described reading thread, for reading the first index, described first index is used for indicating key word
Corresponding relation with document identification；
Second processing module of described reading thread, for the search sequence inputted according to described first index and user, really
The fixed document identification corresponding with described search sequence.
In a kind of possible design, described first index includes N number of subindex, and N is the integer more than 0；Described N number of son
Each subindex correspondence K in index reads thread, and K is the integer more than or equal to 0；
Accordingly, the processing module of described reading thread includes: first reads the process submodule of thread, for reading the first son
Index；Wherein, described first reading thread is the reading thread that described first subindex is corresponding.
In a kind of possible design, all kernels in described first kernel set are divided into N number of kernel subclass,
Each kernel subclass in described N number of kernel subclass and each subindex one_to_one corresponding described；Described each height
What index was corresponding reads on the kernel that thread operates in the kernel subclass that each subindex described is corresponding.
In a kind of possible design, described device also includes: estimates module and determines module；
Described estimate module, for estimating the interior check figure needed for each index described and Thread Count；
Described determine module for: according to the interior check figure needed for each index described, determine that each index described is right
The kernel subclass answered；According to the Thread Count needed for each index described, determine the reading thread that each index described is corresponding.
In a kind of possible design, described device also includes: described in write the processing module of thread, for defeated according to user
The information to be released entered, is updated the second index.
The processing method and processing device of the index that the present invention provides, by being divided into two kernels by all kernels of server
Set, limits reading thread and runs on the kernel in the first kernel set, and restriction is write thread and run in the second kernel set
Kernel, it is to avoid frequently revise data in cache owing to writing thread, and cause for thread, cache being stored for reading
The low problem of data hit rate, improve reading thread when needs read the first index, from cache, read the first index
Probability, thus improve search efficiency.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is this
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to
Other accompanying drawing is obtained according to these accompanying drawings.
Fig. 1 is the schematic diagram of the application scenarios of the processing method that the present invention indexes；
Fig. 2 is the flow chart of the processing method embodiment one that the present invention indexes；
Fig. 3 is the schematic diagram of inverted index of the present invention；
Fig. 4 is the flow chart of the processing method embodiment two that the present invention indexes；
Fig. 5 is the time-consuming comparison diagram of the inquiry under different modes of the present invention；
Fig. 6 is the schematic diagram that the first kernel set is divided by the present invention；
Fig. 7 is the structural representation of the processing means embodiment one that the present invention indexes.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
The a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under not making creative work premise, broadly falls into the scope of protection of the invention.
The method of the present invention can apply to any offer Query Information function and the server of the function that releases news, these clothes
Business device includes M processor, and M is the integer more than 0.Wherein, each processor in M processor can include one or many
Individual kernel.All kernels of server are divided into two kernel set, the respectively first kernel set and the second kernel set；
Wherein, (that is, described reading thread operates in the kernel in described first kernel set with described first kernel set binding to read thread
On), write thread and (with described second kernel set binding that is, write thread described in and operate in the kernel in described second kernel set
On), described reading thread is used for providing the user Query Information function, described in write thread for providing the user the merit that releases news
Energy.Such as, as it is shown in figure 1, server includes 2 processors, respectively processor 1 and processor 2；In processor 1 includes 4
Core, respectively kernel 1, kernel 2, kernel 3 and kernel 4；Processor 2 includes 2 kernels, respectively kernel 5 and kernel 6.Wherein,
Division for kernel set can be such as: kernel 1-kernel 5 is divided to the first kernel set, and kernel 6 is divided to
Two kernel set.It should be noted that read thread and the first kernel set and bind be in order to limit this reading thread run on this
Kernel in one kernel set；Write thread and the second kernel set binding is to write thread and run on this second kernel to limit this
Kernel in set.That is, the kernel in this first kernel set runs this reading thread and does not run this and write thread, this second kernel
Kernel in set runs this and writes thread and do not run this reading thread.
Wherein, for reading thread, it is when reading index, first determines whether to run in the cache of the kernel of this reading thread and is
No save this index；The most directly from cache, read this index；Otherwise, from internal memory, read this index, the most also
This index read is stored in cache so that the reading to this index later can directly obtain from cache.Right
In writing thread, its data that can be written into are written to run in the cache of this kernel writing thread.In prior art, by taking
The operating system consideration based on load balancing of business device, runs the kernel of this reading thread to reading thread distribution, and to writing thread
Distribution runs this kernel writing thread.Therefore, reading thread can be there is and write thread and run on the scene of same kernel.When reading thread
With write thread run on same kernel time, it may appear that frequently revise the data in the cache of kernel owing to writing thread, and cause right
Low in the data hit rate reading to be stored in cache for thread, read the number required for thread can not directly read from cache
According to, and need to read from internal memory, the time needed for reading data is longer, and then causes the problem that search efficiency is low.
Fig. 2 is the flow chart of the processing method embodiment one that the present invention indexes, and the method for the present embodiment can apply to appoint
What provides Query Information function and the server of the function that releases news；As in figure 2 it is shown, the method for the present embodiment may include that
Step 201, described reading thread read the first index.
In this step, described first index is for indicating the corresponding relation of key word and document identification.Described first index
Should be the index that Query Information function can be provided in described server.Described first index can be such as inverted index, its
In, described inverted index such as can be as shown in Figure 3.
Step 202, described reading thread, according to described first index and the search sequence of user's input, determine and described inquiry
The document identification that sequence pair is answered.
In this step, when described server is forum servers, described document identification is specifically as follows model mark
(ID, identification).Wherein, the search sequence of described user input can be by subscriber equipment (wherein, described user
Equipment can be such as mobile phone, panel computer etc.) sent to described server by inquiry request message.Step 202 specifically may be used
Think and by described search sequence, the key word in described first index is mated, according to mate with described search sequence
Key word, and the document identification corresponding with the key word that described search sequence is mated is defined as corresponding with described search sequence
Document identification.Such as, for the inverted index shown in Fig. 3, when the key word mated with described search sequence is that key word 1+ closes
During keyword 2, then corresponding with described search sequence document identification is document identification 3.
In the present embodiment, by all kernels of server being divided into two kernel set, limiting reading thread and running on
Kernel in first kernel set, and limit and write thread and run on the kernel in the second kernel set, it is to avoid owing to writing thread
The frequently data in amendment cache, and cause for reading the problem that the data hit rate that stored in cache for thread is low,
Improve reading thread when needs read the first index, from cache read first index probability, thus improve inquiry effect
Rate.
Fig. 4 is the flow chart of the processing method embodiment two that the present invention indexes, and the method for the present embodiment can apply to appoint
What provides Query Information function and the server of the function that releases news；As shown in Figure 4, the method for the present embodiment may include that
Step 401, described reading thread read the first index.
In this step, described first index is for indicating the corresponding relation of key word and document identification.Described first index
Should be the index that Query Information function can be provided in described server.Described reading thread reads the first index, is specifically as follows:
Described reading thread judges to run in the cache of the kernel of described reading thread whether save described first index；The most then from this
Cache reads described first index；Otherwise, from the internal memory of described server, read described first index, and by described the
One index stores to this cache.It should be noted that this kernel should be the kernel in described first kernel set.
Optionally, described first index can include N number of subindex, and N is the integer more than 0；In described N number of subindex
Each subindex correspondence K reads thread, and K is the integer more than or equal to 0.Accordingly, described reading thread reads the first index,
Including: first reads thread reads the first subindex；Wherein, described first reading thread is the reading line that described first subindex is corresponding
Journey.That is, the reading thread that subindex is corresponding this subindex is read out.N number of subindex is included by the first index, and each
Individual subindex correspondence K reads thread, so that multiple reading threads of corresponding different subindex can read the plurality of reading simultaneously
The subindex that thread is the most corresponding, thus further increase search efficiency.Fig. 5 gives the inquiry consumption under mode 1 and mode 2
Time comparison diagram.Wherein, the first index is divided into N number of subindex by mode 1, each subindex correspondence K reading thread, and according to
Above-mentioned mode of the prior art determines that operation is read thread and runs the kernel writing thread, therefore there is reading thread and writes thread fortune
Row is in the scene of same kernel；The difference of mode 2 and mode 1 essentially consists in and uses the method for the present invention will read thread in mode 2
With the first kernel set binding, write thread and the second kernel set binding, thus avoid reading thread and run on together with writing thread
The scene of one kernel.Such as Fig. 5 it can be seen that mode 2 time-consumingly greatly reduces compared to mode 1, inquiry.
Optionally, all kernels in described first kernel set are divided into N number of kernel subclass, described N number of kernel
Each kernel subclass in subclass and each subindex one_to_one corresponding described；The reading that each subindex described is corresponding
Thread operates on the kernel in the kernel subclass that each subindex described is corresponding.Division example for the first kernel set
As can be wherein assumed that the kernel in the first kernel set is kernel 1-kernel 5, N=5 as shown in Figure 6.Kernel subset in Fig. 6
Close the corresponding relation with subindex, such as, can be the corresponding subindex 1 of kernel subclass 1, the corresponding subindex 2 of kernel subclass 2,
The corresponding subindex 3 of kernel subclass 3, the corresponding subindex 4 of kernel subclass 4, the corresponding subindex 5 of kernel subclass 5.Need explanation
Be, it is assumed that the reading thread of subindex 1 correspondence is for reading thread 1 and reading thread 2, and the kernel subset of subindex 1 correspondence is combined into kernel
Subclass 1, then it represents that read thread 1 and reading thread 2 runs on the kernel (i.e. running on kernel 1) in kernel subclass 1.By inciting somebody to action
All kernels in first kernel set are divided into N number of kernel subclass, each kernel subclass and each subindex one
One is corresponding, and each subindex corresponding read on kernel that thread operates in kernel subclass that each index is corresponding,
Further define and run each scope of kernel reading thread, therefore further increase and the cache of kernel is stored
The hit rate of data, thus further increase search efficiency.
Optionally, before step 401, it is also possible to including: estimate the interior check figure needed for each subindex described and thread
Number；According to the interior check figure needed for each subindex described, determine the kernel subclass that each subindex described is corresponding；According to
Thread Count needed for each subindex described, determines the reading thread that each subindex described is corresponding.Optionally, estimate described in
Interior check figure needed for each subindex described, is specifically as follows: according to the kernel sum in described server and the first ratio
The example factor, (such as, the interior check figure=kernel needed for each subindex is total to determine the interior check figure needed for each subindex described
Number is multiplied by the first scale factor)；Wherein, in described first scale factor is previous time period, each subindex described takies
Kernel resources ratio in the kernel resources that described N number of subindex takies.Described estimate each subindex described needed for
Thread Count, is specifically as follows: according to the reading total number of threads in described server and the second scale factor, determine described each
Thread Count needed for subindex (such as, the Thread Count needed for each subindex=reading total number of threads be multiplied by the second ratio because of
Son)；Wherein, the thread resources that in described second scale factor is previous time period, each subindex described takies is at described N
Ratio in the reading thread resources that individual subindex takies.Needed for estimating each subindex described described in it should be noted that
Interior check figure and Thread Count, determine, according to the interior check figure needed for each subindex described, the kernel that each subindex described is corresponding
Subclass, and according to the Thread Count needed for each subindex described, determine the reading thread that each subindex described is corresponding,
Specifically can by described server except described reading thread with described write thread in addition to other threads perform.
Step 402, described reading thread, according to described first index and the search sequence of user's input, determine and described inquiry
The document identification that sequence pair is answered.
It should be noted that step 402 is similar with step 202, do not repeat them here.
Step 403, described reading thread, according to the document identification corresponding with described search sequence, obtain corresponding document.
In this step, described document can be such as model.After obtaining described document, can be disappeared by inquiry response
Described document is returned to described subscriber equipment, so that described document is shown to user by described subscriber equipment by breath.
Step 404, described in write the information to be released that thread inputs according to user, to second index be updated.
In this step, the information to be released of described user input can be sent extremely by issuing request message by subscriber equipment
Described server.Described second index is for indicating the corresponding relation of key word and document identification.Described second index should be institute
State the index that the service of releasing news can be provided in server.Optionally, described second index is different from above-mentioned first index
Index, or can also be identical index.Being indexed from above-mentioned first by the second index is different indexes, can avoid
Read thread and write thread and need to access same index and the access conflict problem that causes, thus further increasing search efficiency.
Optionally, when described second index is different index with described first index, described second index and above-mentioned each subindex
There is the life cycle of oneself；After the life cycle of the index (such as, index A) for the function that releases news arrives, rope
Draw A and be changed to the index for Query Information function, now can produce new index (such as, index B) and be used for releasing news
Function；After the life cycle of the index (such as, index C) for Query Information function arrives, can by index C and other
Certain merges for the index of Query Information function.
It should be noted that do not have the restriction of sequencing between step 404 and step 401-step 403.
Fig. 7 is the structural representation of the processing means embodiment one that the present invention indexes；The processing means of this index can be led to
Cross software, hardware or both be implemented in combination with become the some or all of of server.All kernels in described server
It is divided into two kernel set；Wherein, read thread and operate on the kernel in the first kernel set, write thread and operate in second
On kernel in kernel set, described reading thread is used for providing the user Query Information function, described in write thread for for user
Offer releases news function.As it is shown in fig. 7, the processing means of this index includes: the first processing module 701 of described reading thread,
Second processing module 702 of described reading thread.Wherein, the first processing module 701 of described reading thread, for reading the first rope
Drawing, described first index is for indicating the corresponding relation of key word and document identification；Second processing module of described reading thread
702, for described first index read according to the first processing module 701 of described reading thread and the inquiry sequence of user's input
Row, determine the document identification corresponding with described search sequence.
The device of the present embodiment, may be used for performing the technical scheme of embodiment of the method shown in Fig. 2, and it realizes principle and skill
Art effect is similar to, and here is omitted.
The processing means embodiment two of index
Optionally, on the basis of the processing means embodiment one of present invention index, further, described first index bag
Including N number of subindex, N is the integer more than 0；Each subindex correspondence K in described N number of subindex reads thread, and K is for being more than
Or the integer equal to 0；
Accordingly, the first processing module 701 of described reading thread includes: first reads the process submodule of thread, is used for reading
Take the first subindex；Wherein, described first reading thread is the reading thread that described first subindex is corresponding.
Optionally, all kernels in described first kernel set are divided into N number of kernel subclass, described N number of kernel
Each kernel subclass in subclass and each subindex one_to_one corresponding described；The reading that each subindex described is corresponding
Thread operates on the kernel in the kernel subclass that each subindex described is corresponding.
Optionally, the device of the present embodiment can also include: estimates module and determines module；Wherein, described in estimate module,
For estimating the interior check figure needed for each index described and Thread Count；Described determine module for: according to each rope described
Draw required interior check figure, determine the kernel subclass that each index described is corresponding；According to the line needed for each index described
Number of passes, determines the reading thread that each index described is corresponding.
Further alternative, the device of the present embodiment can also include: described in write the processing module of thread.Wherein, described
Write the processing module of thread, for the information to be released according to user's input, the second index is updated.
The device of the present embodiment, may be used for performing the technical scheme of embodiment of the method shown in Fig. 4, and it realizes principle and skill
Art effect is similar to, and here is omitted.
One of ordinary skill in the art will appreciate that: all or part of step realizing above-mentioned each method embodiment can be led to
The hardware crossing programmed instruction relevant completes.Aforesaid program can be stored in a computer read/write memory medium.This journey
Sequence upon execution, performs to include the step of above-mentioned each method embodiment；And aforesaid storage medium includes: ROM, RAM, magnetic disc or
The various media that can store program code such as person's CD.
Last it is noted that various embodiments above is only in order to illustrate technical scheme, it is not intended to limit；To the greatest extent
The present invention has been described in detail by pipe with reference to foregoing embodiments, it will be understood by those within the art that: it depends on
So the technical scheme described in foregoing embodiments can be modified, or the most some or all of technical characteristic is entered
Row equivalent；And these amendments or replacement, do not make the essence of appropriate technical solution depart from various embodiments of the present invention technology
The scope of scheme.
Claims (10)
1. the processing method of an index, it is characterised in that described method is applied to server, in all in described server
Core is divided into two kernel set；Wherein, read thread and operate on the kernel in the first kernel set, write thread and operate in the
On kernel in two kernel set, described reading thread is used for providing the user Query Information function, described in write thread for for use
Family provides the function that releases news；Described method includes:
Described reading thread reads the first index, and described first index is for indicating the corresponding relation of key word and document identification；
Described reading thread, according to described first index and the search sequence of user's input, determines the literary composition corresponding with described search sequence
Shelves mark.
Method the most according to claim 1, it is characterised in that described first index includes N number of subindex, and N is more than 0
Integer；Each subindex correspondence K in described N number of subindex reads thread, and K is the integer more than or equal to 0；
Accordingly, described reading thread reads the first index, including: first reads thread reads the first subindex；Wherein, described first
Reading thread is the reading thread that described first subindex is corresponding.
Method the most according to claim 2, it is characterised in that all kernels in described first kernel set are divided into
N number of kernel subclass, each the kernel subclass in described N number of kernel subclass and described each subindex one a pair
Should；The reading thread that each subindex described is corresponding operates in the kernel in the kernel subclass that each subindex described is corresponding
On.
Method the most according to claim 3, it is characterised in that before described reading index, described method also includes:
Estimate the interior check figure needed for each subindex described and Thread Count；
According to the interior check figure needed for each subindex described, determine the kernel subclass that each subindex described is corresponding；
According to the Thread Count needed for each subindex described, determine the reading thread that each subindex described is corresponding.
5. according to the method described in any one of claim 1-4, it is characterised in that described method also includes:
Described write the information to be released that thread inputs according to user, the second index is updated.
6. the processing means of an index, it is characterised in that described device is server or described device is integrated in described clothes
In business device, all kernels in described server are divided into two kernel set；Wherein, read thread and operate in the first kernel collection
On kernel in conjunction, writing on the kernel that thread operates in the second kernel set, described reading thread is used for providing the user inquiry
Informational function, described in write thread for providing the user the function that releases news；Described device includes:
First processing module of described reading thread, for reading the first index, described first index is used for indicating key word and literary composition
The corresponding relation of shelves mark；
Second processing module of described reading thread, for according to described first index and user input search sequence, determine with
The document identification that described search sequence is corresponding.
Device the most according to claim 6, it is characterised in that described first index includes N number of subindex, and N is more than 0
Integer；Each subindex correspondence K in described N number of subindex reads thread, and K is the integer more than or equal to 0；
Accordingly, the first processing module of described reading thread includes: first reads the process submodule of thread, for reading the first son
Index；Wherein, described first reading thread is the reading thread that described first subindex is corresponding.
Device the most according to claim 7, it is characterised in that all kernels in described first kernel set are divided into
N number of kernel subclass, each the kernel subclass in described N number of kernel subclass and described each subindex one a pair
Should；The reading thread that each subindex described is corresponding operates in the kernel in the kernel subclass that each subindex described is corresponding
On.
Device the most according to claim 8, it is characterised in that described device also includes: estimate module and determine module；
Described estimate module, for estimating the interior check figure needed for each index described and Thread Count；
Described determine module for: according to the interior check figure needed for each index described, determine that each index described is corresponding
Kernel subclass；According to the Thread Count needed for each index described, determine the reading thread that each index described is corresponding.
10. according to the device described in any one of claim 6-9, it is characterised in that described device also includes:
The described processing module writing thread, for the information to be released according to user's input, is updated the second index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610623529.8A CN106250492B (en) | 2016-07-28 | 2016-07-28 | The processing method and processing device of index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610623529.8A CN106250492B (en) | 2016-07-28 | 2016-07-28 | The processing method and processing device of index |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106250492A true CN106250492A (en) | 2016-12-21 |
| CN106250492B CN106250492B (en) | 2019-11-19 |
Family
ID=57606919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610623529.8A Active CN106250492B (en) | 2016-07-28 | 2016-07-28 | The processing method and processing device of index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106250492B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408613A (en) * | 2018-08-14 | 2019-03-01 | 广东神马搜索科技有限公司 | Index structure operating method, device and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1193600A2 (en) * | 1995-02-20 | 2002-04-03 | Hitachi, Ltd. | Memory control apparatus and its control method |
| US20030182507A1 (en) * | 2002-03-20 | 2003-09-25 | Naffziger Samuel David | Methods and apparatus for control of asynchronous cache |
| CN102171695A (en) * | 2008-10-05 | 2011-08-31 | 微软公司 | Efficient large-scale joining for querying of column based data encoded structures |
| CN102346714A (en) * | 2011-10-09 | 2012-02-08 | 西安交通大学 | Consistency maintenance device for multi-kernel processor and consistency interaction method |
| CN103207907A (en) * | 2013-03-28 | 2013-07-17 | 新浪网技术(中国)有限公司 | Method and device for combining index files |
| CN104484131A (en) * | 2014-12-04 | 2015-04-01 | 珠海金山网络游戏科技有限公司 | Device and corresponding method for processing data of multi-disk servers |
| CN105740164A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Multi-core processor supporting cache consistency, reading and writing methods and apparatuses as well as device |
-
2016
- 2016-07-28 CN CN201610623529.8A patent/CN106250492B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1193600A2 (en) * | 1995-02-20 | 2002-04-03 | Hitachi, Ltd. | Memory control apparatus and its control method |
| US20030182507A1 (en) * | 2002-03-20 | 2003-09-25 | Naffziger Samuel David | Methods and apparatus for control of asynchronous cache |
| CN102171695A (en) * | 2008-10-05 | 2011-08-31 | 微软公司 | Efficient large-scale joining for querying of column based data encoded structures |
| CN102346714A (en) * | 2011-10-09 | 2012-02-08 | 西安交通大学 | Consistency maintenance device for multi-kernel processor and consistency interaction method |
| CN103207907A (en) * | 2013-03-28 | 2013-07-17 | 新浪网技术(中国)有限公司 | Method and device for combining index files |
| CN104484131A (en) * | 2014-12-04 | 2015-04-01 | 珠海金山网络游戏科技有限公司 | Device and corresponding method for processing data of multi-disk servers |
| CN105740164A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Multi-core processor supporting cache consistency, reading and writing methods and apparatuses as well as device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408613A (en) * | 2018-08-14 | 2019-03-01 | 广东神马搜索科技有限公司 | Index structure operating method, device and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106250492B (en) | 2019-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8224813B2 (en) | Cost based analysis of direct I/O access | |
| CN101458705A (en) | Data collating method between different utility systems, apparatus and system | |
| CN105550270B (en) | Data base query method and device | |
| US20170212930A1 (en) | Hybrid architecture for processing graph-based queries | |
| CN111752944B (en) | Data allocation method, device, computer equipment and storage medium | |
| CN109325055A (en) | The screening of business association tables of data and checking method, device, electronic equipment | |
| Hao et al. | PatMat: a distributed pattern matching engine with cypher | |
| Zhang et al. | Oceanrt: Real-time analytics over large temporal data | |
| CN111949315A (en) | Management device and method for block chain account book data | |
| CN109885642B (en) | Hierarchical storage method and device for full-text retrieval | |
| CN112182004A (en) | Method and device for viewing data in real time, computer equipment and storage medium | |
| CN109271247A (en) | Memory Optimize Method, device, computer installation and storage medium | |
| CN111813517B (en) | Task queue allocation method and device, computer equipment and medium | |
| CN107861878A (en) | The method, apparatus and equipment of java application performance issue positioning | |
| CN109614411B (en) | Data storage method, device and storage medium | |
| US8332595B2 (en) | Techniques for improving parallel scan operations | |
| US20210149746A1 (en) | Method, System, Computer Readable Medium, and Device for Scheduling Computational Operation Based on Graph Data | |
| CN106250492A (en) | The processing method and processing device of index | |
| CN112052330B (en) | Application keyword distribution method and device | |
| CN114969189A (en) | Method and device for determining connection in database connection pool | |
| CN111143328A (en) | Agile business intelligent data construction method, system, equipment and storage medium | |
| CN107291371A (en) | The implementation method and device of a kind of Read-Write Locks | |
| CN114334075A (en) | Data shape confidence | |
| CN112667682A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| Zhao et al. | Speeding Up Subgraph Matching Queries with Schema Guided Index |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |