CN103853772B - High-efficiency reverse index organizing method - Google Patents

High-efficiency reverse index organizing method Download PDF

Info

Publication number
CN103853772B
CN103853772B CN201210511478.1A CN201210511478A CN103853772B CN 103853772 B CN103853772 B CN 103853772B CN 201210511478 A CN201210511478 A CN 201210511478A CN 103853772 B CN103853772 B CN 103853772B
Authority
CN
China
Prior art keywords
item
attribute
record
row
arrange
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.)
Active
Application number
CN201210511478.1A
Other languages
Chinese (zh)
Other versions
CN103853772A (en
Inventor
王弘蔚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
TOLS INFORMATION TECHNOLOGY Co.,Ltd.
Original Assignee
BEIJING TRS INFORMATION TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by BEIJING TRS INFORMATION TECHNOLOGY Co Ltd filed Critical BEIJING TRS INFORMATION TECHNOLOGY Co Ltd
Priority to CN201210511478.1A priority Critical patent/CN103853772B/en
Publication of CN103853772A publication Critical patent/CN103853772A/en
Application granted granted Critical
Publication of CN103853772B publication Critical patent/CN103853772B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/22Indexing; Data structures therefor; Storage structures
    • G06F16/2228Indexing structures
    • G06F16/2272Management thereof

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (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 a high-efficiency reverse index organizing method. A recording reverse item and an attribute reverse item are separately stored, so that information redundancy is reduced; reverse indexes are stored in a partitioning mode; the optimization of a database index is taken as an entry point and the advantages of a chain table and a centralized positioning table are comprehensively utilized to manage the reverse blocks, so that the dynamic maintenance of indexes is benefited and the random access under the SKIP technique is convenient; the compression technique is adopted according to inter-domain incidence relation, so that much information is stored in less space; according to the high-efficiency reverse index structure and the organizing method provided by the invention, the storage space is reduced, the access time is optimized, the problem of reverse index optimization is solved to some extent and the effective management for mass data is realized.

Description

A kind of efficient reverse index organization method
Technical field
The present invention relates to information retrieval technique, specifically, it is related to efficient inverted index used in text retrieval system Method for organizing.
Background technology
In the network environment in big data epoch, quantity of information and customer volume are in explosive growth, to extensive information retrieval system The service of system precise and high efficiency brings pressure and challenge.Inverted index is the core of information retrieval, its storage organization and organizer Formula has a significant impact to the lifting of information retrieval performance.While improving searching algorithm, people also are making great efforts to optimize the row's of falling rope Guiding structure and its organizational form.
It is directed to the research that inverted index optimizes both at home and abroad at present, mainly launch in terms of following three:(1) pass through compress technique Reduce volume on external memory for the index;(2) organizational form of inverted list content is optimized, reduces the inverted list needing to access Content;(3) the disk storage structure of inverted list is managed, reduces IO number of disk as far as possible.But existing achievement includes Compress technique, inverted index organizational form and storage organization are also immature, are difficult to adapt to big data epoch information retrieval system letter Breath amount is big, quick response is smart and the demand of standard, and therefore the present invention proposes a kind of efficient inverted index merging compress technique Structure and method for organizing, to solve current inverted index optimization problem, realize effective management of mass data.
Content of the invention
The technical problem to be solved in the present invention is a kind of offer optimization memory space, simplifies the inverted index knot of read access time Structure and method for organizing.
For solve current information retrieval system contain much information, quick response essence and the technical problem of quasi need aspect, this Bright provide a kind of efficient inverted index structure, this structure includes:
S101 inverted index is made up of the row's of falling item, and row's item includes recording the row's of falling item and attribute falls to arrange item two types, it Be stored separately to reduce information redundancy, logically inverted index is also divided into recording inverted index and attribute and falls to arrange simultaneously Index;
S102 record inverted index is formed by recording down row's item, and " record falls to arrange item " is individually formed by " recording mechanism ", Including " recording mechanism " and " linker ";
S103 " recording mechanism " is the record being located when word in a field value or full-text data occurs in data base;
The particular content of S104 " linker " depends on different organizational forms, can be address pointer, side-play amount or Attribute falls to arrange number of item etc., and its function is all properties item that can quickly position and read a word in this record;
S105 attribute inverted index falls to arrange item by attribute and constitutes, and " attribute arranges item " is by " position attribution " and " form belongs to Property " composition;
106 " position attributions " are one of positions that the word in full-text data occurs in current record, and position attribution divides three Individual level " paragraph, sentence, position ";
S107 " attribute of a configuration " is word in full-text data other format informations that position has (as Native XML indexes);
The present invention also provides a kind of efficient reverse index organization method, and the method includes:
201 inverted indexs carry out tissue storage hereof by the way of piecemeal, and row's block comprehensively adopts chained list and collection Middle positioning table two ways was managed so that inverted index had not only been easy to Dynamic Maintenance but also had been conducive to the random visit under SKIP technology Ask;For improving data loading efficiency, be conducive to data random access simultaneously, to be selected as critical point using the optimization of database index Select down the way to manage of row's block:Inverted index after optimization is managed using concentration positioning table, and the new number loading after optimizing Still managed with chained list according to corresponding inverted index, until the optimization of database index next time;
S202, in whole inverted index, will concentrate positioning table to regard a special chained list node as, using flag bit Mechanism is conventional chained list node to distinguish, or concentrates positioning table;Attribute falls to arrange item data block and is placed on record to fall to arrange item data After block, combination is stored in the same row of the falling item data block of identical file;
The S203 table block of falling row chain is made up of " linker " and " arranging item data block ", and the basic role of " linker " is instruction Next table block of falling row chain or the first address concentrating positioning table, wherein contain 1 flag bit and indicate that what next address pointed to is down row chain Table block still concentrates positioning table;
S204 concentrates positioning table to be made up of " gauge outfit " and " list item ", because the row's of falling section is piecemeal storage, the master of " gauge outfit " Want the address that function is the concentration positioning table providing next row's of falling section;Last row's of falling section is public, is used for depositing precise results Collection and the searching algorithm of non-precision result set, therefore, the gauge outfit of each row's of falling section is both needed to deposit the concentration of last row's of falling section The address of positioning table, so that the searching algorithm of precise results collection can quickly locate this section;
The row's of falling block that S205 concentration positioning table manages is all through optimizing, and a list item arranges block, institute with respect to one Take up space very little, directly uses natural each domain of data types to express list item;As the table block of falling row chain " linker ", " Location " is indispensable domain for each list item, and " dominant record number " and " smallest record number " domain can not be omitted simultaneously, so permissible Avoid obtaining " dominant record number " and " smallest record number " from the row's of falling item data block, to reduce unnecessary operation;Additionally, collection Record in middle positioning table falls to arrange item and carries out Fragmentation by recording mechanism, and therefore, each " list item " also needs to a recording mechanism page The length of code table, is easy to quickly position, search recording mechanism, is the initial address of recording mechanism offset table after page number table, " page number table 1 bit flag position in length " domain, distinguishes the row's of falling block whether paging, during not paging, low level is used for representing attribute excursion amount, otherwise then Represent recording mechanism side-play amount;
In the S206 table block of falling row chain, the row's of falling block is different on storage mode with the row's of the falling block concentrating positioning table to be managed, Using the Fragmentation mode that have passed through optimization, the former still adopts simplest storage mode to the latter, thus the two is corresponding The logical structure that record falls to arrange item also differs;
In the S207 table block of falling row chain, record falls to arrange item and is made up of " recording mechanism " and " linker ".All genus of one record Property fall row item and record between attribute fall arrange item attribute fall arrange item data block in all continuously deposit, " linker " Effect is to indicate that all properties of this record fall to arrange item and fall to arrange the original position in item data block in attribute;
S208 concentrates in positioning table, and row's item is divided into " recording mechanism page number table " list item in storage to record and " recording mechanism is inclined Move table " list item.Each " recording mechanism page number table " stores the recording mechanism page number of a page and corresponding recording mechanism offset table Location information." recording mechanism offset table " is made up of " recording mechanism skew " and " attribute excursion amount " two domains, is concentrating positioning table " table It is marked tissue in " page number table length " domain of item ";
S209 attribute fall arrange item contained by subitem is many, some subitems are optional, subitem number is not fixed, have 8 kinds of specific shapes Formula, respectively:<Position>、<Sentence, position>、<Paragraph, position>、<Paragraph, sentence, position>、<Position, form>、<Sentence, Position, form>、<Paragraph, position, form>、<Paragraph, sentence, position, form>;" position attribution " domain adopts 2 table of bytes Show;Attribute falls to arrange item and can comprise the attribute of a configuration, and the attribute of a configuration, to Narive XML index, is the knot of an XML data Period, with four byte representations, in order to support retrieval weighting, one of byte storage weight (be could support up 256 kinds of power Weight), remaining three bytes are used for storing original node number.
Efficient inverted index structure and method for organizing that technical solution of the present invention provides, the row's of falling item in inverted index is divided into Record falls to arrange item and attribute falls to arrange item and stored, to reduce information redundancy;Inverted index organizes storage in the way of piecemeal, comprehensive The advantage closed chained list and concentrate positioning table, is managed so that inverted index had not only been easy to Dynamic Maintenance but also had been conducive to the row's of falling block Random access under SKIP technology, employs certain compress technique simultaneously in data storage, to realize in as far as possible few sky Between the as far as possible many information of storage;Decrease data space, improve data directory efficiency.
Brief description
Fig. 1 is a kind of efficient inverted index structure schematic diagram provided in an embodiment of the present invention.
Fig. 2 is a kind of high-efficient inverted row index organization provided in an embodiment of the present invention frame diagram.
Fig. 3 is the organisational chart concentrating the row's of falling block in positioning table provided in an embodiment of the present invention.
Specific embodiment
For solving magnanimity information epoch data directory, optimization problem, efficient inverted index knot provided in an embodiment of the present invention Structure and method for organizing, are optimized to inverted index in terms of memory space and read access time two, will arrange item and be divided into record to fall Row's item and attribute arrange item, reduce information redundancy, and inverted index carries out tissue storage with partitioned mode, comprehensively adopt chained list and collection The advantage of middle positioning table is managed, and has not only been easy to Dynamic Maintenance but also has been conducive to random access, it is possible to resolve inverted index optimization problem.
Purpose, technical method and advantage for making the embodiment of the present invention become more apparent, below in conjunction with accompanying drawing to this The technical scheme that bright embodiment provides is described in detail.
It is illustrated in figure 1 the efficient inverted index structure schematic diagram in the embodiment of the present invention, searching system is positioned by vocabulary Record falls to arrange item, comprises " recording mechanism " and " linker ", arrange item by " linker " positioning properties, give in " record falls to arrange item " Go out this word all properties information in record, attribute falls to arrange all properties information that item is also recorded between record, " attribute simultaneously The row's of falling item " comprises " position attribution " and " attribute of a configuration ", and wherein " attribute of a configuration " is optional content, and " position attribution " comprises again " paragraph ", " sentence ", " position ", wherein " paragraph ", " sentence " are optional contents, thus achieve dividing between record and attribute From so that each record only needs a record to arrange item can complete the index to it, decrease information redundancy, reduce number While according to memory space, improve data directory efficiency.
Inverted index organization need hereof considers both sides factor:Random under Dynamic Maintenance and SKIP technology Access.Need for inverted index to carry out piecemeal storage, a kind of high-efficient inverted row's rope providing in the embodiment of the present invention is provided Draw organization framework figure.Attribute is fallen to arrange item data block is placed on record to arrange after item data block, and combination is stored in identical file Same fall row item data block in, constitute fall arrange block.
Management to the row's of falling block is divided into chained list and concentrates positioning table two ways.Divide according to the pluses and minuses of two ways Analysis, is optimized for point of penetration with database index, and the inverted index after optimization adopts concentrates positioning table management, new loading after optimization Data still adopts chained list to manage.Consider the row's of falling block size to IO performance impact problem, the row's of the falling block after optimizing presses record simultaneously Number carry out paging, the organisational chart of the row's of falling block in the concentration positioning table provide in the embodiment of the present invention is provided.Set The maximum capacity of one page is 256, if recording mechanism is 32, least-significant byte is recording mechanism skew, and high 24 is then recording mechanism page Code.
Fall to arrange item according to vocabulary positioning record, locking falls to arrange item data block, according to " linker " letter in the row's of falling item data block Breath and the organizational form of recording mechanism descending, search respective attributes information and next record information, therefore chained list " link The logical structure that record before device " and optimization falls to arrange item may be designed as:
The logical structure of the table block of falling row chain " linker ":
Record before optimization falls to arrange the logical structure of item:
Domain name Byte number
Recording mechanism 4 bytes.
Attribute space size 2 bytes.
Due to concentrating positioning table to store by the way of paging, if next record information is with " concentrating positioning table " mode pipe Reason, then need to obtain minimum and maximum recording mechanism and the page number table length of current page, therefore concentrates list item and the table of positioning table The logical structure of head can be designed as:
Concentrate the logical structure of positioning table " list item ":
Concentrate the logical structure of positioning table " gauge outfit ":
The logical structure that record after optimization falls to arrange item can be designed as:
Domain name Byte number
Recording mechanism offsets 1 byte.
Attribute excursion amount 3 bytes.
The logical structure of recording mechanism page number table list item:
Domain name Byte number
The recording mechanism page number 3 bytes.
Record number in page 1 byte.
Attribute falls to arrange item and is divided into 8 kinds of concrete combining forms, and the logical structure of each combining form can be designed specifically to:
Attribute falls to arrange " position " logical structure of item:
Domain name Byte number
Position 2 bytes.
Attribute falls to arrange " sentence-position " logical structure of item:
Domain name Byte number
Sentence 2 bytes.
Position 2 bytes.
Attribute falls to arrange " paragraph-position " logical structure of item:
Domain name Byte number
Paragraph 2 bytes.
Position 2 bytes.
Attribute falls to arrange " paragraph-sentence-position " logical structure of item:
Domain name Byte number
Paragraph 2 bytes.
Sentence 2 bytes.
Position 2 bytes.
Attribute falls to arrange " position-form " logical structure of item:
Domain name Byte number
Position 2 bytes.
Form 4 bytes.
Attribute falls to arrange " sentence-position-form " logical structure of item:
Domain name Byte number
Sentence 2 bytes.
Position 2 bytes.
Form 4 bytes.
Attribute falls to arrange " paragraph-position-form " logical structure of item:
Domain name Byte number
Paragraph 2 bytes.
Position 2 bytes.
Form 4 bytes.
Attribute falls to arrange " paragraph-sentence-position-form " logical structure of item:
Domain name Byte number
Paragraph 2 bytes.
Sentence 2 bytes.
Position 2 bytes.
Form 4 bytes.

Claims (9)

1. it is characterised in that inverted index is made up of the row's of falling item, row's item is split into a kind of efficient reverse index organization method Record falls to arrange item and attribute arranges item, and attribute falls to arrange item data block and is placed on record to fall to arrange after item data block, and combination is stored in In the same row of the falling item data block of identical file, constitute the row's of falling block, record inverted index and attribute are divided into by this inverted index Inverted index, wherein, record inverted index is formed by recording down row's item, and " record falls to arrange item " is individually formed by " recording mechanism ", bag Include " recording mechanism " and " linker ";" recording mechanism " is that the word in a field value or full-text database when institute in data base Record;The particular content of " linker " depends on different organizational forms, and its function is quickly to position and read one All properties item in this record for the word;Attribute inverted index falls to arrange item by attribute and constitutes, and " attribute arranges item " is by " position belongs to Property " and " attribute of a configuration " composition;" position attribution " is one of position that the word in full-text data occurs in current record, position Attribute is divided into three levels " paragraph, sentence, position ";" attribute of a configuration " is word in full-text data to be had position Other format informations;Therefore, inverted index carries out tissue storage using partitioned mode, arranges block and is comprehensively determined with concentrating using chained list Position table is managed.
2. efficient reverse index organization method according to claim 1 is it is characterised in that being optimized for database index Cut-off point selection falls to arrange the way to manage of block, concentrates positioning table to be regarded as a special chained list node, using flag bit machine Make and to distinguish.
3. efficient reverse index organization method according to claim 1 and 2 it is characterised in that " linker " and " arrange item Data block " forms the table block of falling row chain, and the basic role of " linker " is next row's of falling block of instruction or the first address concentrating positioning table, Wherein contain that 1 flag bit indicates that next address points to is the table of falling row chain or concentrates positioning table.
4. efficient reverse index organization method according to claim 1 and 2 is it is characterised in that concentrate positioning table by " table Head " and " list item " composition, row's section piecemeal storage, list item carries out paging by recording mechanism.
5. efficient reverse index organization method according to claim 4 is it is characterised in that concentrate a list item in positioning table The row's of falling block with respect to one, each row's of falling block need to contain " arranging block address ", " dominant record
Number ", " smallest record number " and " page number table length ".
6. efficient reverse index organization method according to claim 5 is it is characterised in that page number table length is used for storage one The length of the page number table of individual recording mechanism, is the initial address of recording mechanism offset table after page number table, 1 in " page number table length " domain Flag bit, distinguishes the row's of falling block whether paging, during not paging, low level is used for representing attribute excursion amount, otherwise then represents that recording mechanism is inclined Shifting amount.
7. efficient reverse index organization method according to claim 3 is it is characterised in that the row's of falling item data block is divided into record The row's of falling item data block and attribute fall to arrange item data block, and attribute falls to arrange subitem contained by item and includes:Paragraph, sentence, position, form, no Same compound mode can form specific form in 8, is followed successively by:<Position>、<Sentence, position>、<Paragraph, position>、<Paragraph, Sentence, position>、<Position, form>、<Sentence, position, form>、<Paragraph, position, form>、<Paragraph, sentence, position, form >.
8. efficient reverse index organization method according to claim 7 is it is characterised in that in the table of falling row chain, record falls Row's item is made up of " recording mechanism " and " linker ", and the attribute between all properties of a record fall to arrange item and record arranges item Fall to arrange in item data block in attribute and all continuously deposit, the effect of " linker " is that all properties indicating this record arrange item Fall to arrange the original position in item data block in attribute.
9. efficient reverse index organization method according to claim 8 is it is characterised in that in concentrating positioning table, record The row's of falling item is divided into " recording mechanism page number table " list item and " recording mechanism offset table " list item in storage, and each " recording mechanism page number table " is deposited The storage recording mechanism page number of one page and the location information of corresponding recording mechanism offset table, " recording mechanism offset table " is by " record Number side-play amount " and " attribute excursion amount " two domains compositions, enter rower in " the page number table length " domain concentrating positioning table " list item " Note tissue.
CN201210511478.1A 2012-12-04 2012-12-04 High-efficiency reverse index organizing method Active CN103853772B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210511478.1A CN103853772B (en) 2012-12-04 2012-12-04 High-efficiency reverse index organizing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210511478.1A CN103853772B (en) 2012-12-04 2012-12-04 High-efficiency reverse index organizing method

Publications (2)

Publication Number Publication Date
CN103853772A CN103853772A (en) 2014-06-11
CN103853772B true CN103853772B (en) 2017-02-08

Family

ID=50861438

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210511478.1A Active CN103853772B (en) 2012-12-04 2012-12-04 High-efficiency reverse index organizing method

Country Status (1)

Country Link
CN (1) CN103853772B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106575294B (en) * 2014-07-31 2020-01-21 华为技术有限公司 Track data query method and device
CN106909623B (en) * 2017-01-19 2019-11-26 中国科学院信息工程研究所 A kind of data set and date storage method for supporting efficient mass data to analyze and retrieve
CN108509625A (en) * 2018-04-08 2018-09-07 中国人民解放军63601部队 A kind of data inactivity table location positioning method
CN110162662B (en) * 2019-04-18 2023-02-28 创新先进技术有限公司 Verification method, device and equipment for data records in block chain type account book

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6578026B1 (en) * 1999-07-29 2003-06-10 International Business Machines Corporation Method and system for conducting reverse index scans
CN102201001A (en) * 2011-04-29 2011-09-28 西安交通大学 Fast retrieval method based on inverted technology
CN102063446B (en) * 2009-11-13 2012-07-11 中国移动通信集团四川有限公司 Method for creating inverted index and inverted indexing device
CN102722553A (en) * 2012-05-24 2012-10-10 浙江大学 Distributed type reverse index organization method based on user log analysis

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8176021B2 (en) * 2008-06-02 2012-05-08 Microsoft Corporation Optimized reverse key indexes

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6578026B1 (en) * 1999-07-29 2003-06-10 International Business Machines Corporation Method and system for conducting reverse index scans
CN102063446B (en) * 2009-11-13 2012-07-11 中国移动通信集团四川有限公司 Method for creating inverted index and inverted indexing device
CN102201001A (en) * 2011-04-29 2011-09-28 西安交通大学 Fast retrieval method based on inverted technology
CN102722553A (en) * 2012-05-24 2012-10-10 浙江大学 Distributed type reverse index organization method based on user log analysis

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"基于全文数据库的全文检索模型研究";郭琦娟;《中国优秀硕士学位论文全文数据库 信息科技辑》;20080315;论文正文第40页第3段-第44页第3段、第57页第1段-第59页第4段,附图4.7-4.11 *
"数字图书馆中基于结构和内容的文本信息检索技术的研究";阚忠良;《中国优秀硕士学位论文全文数据库 信息科技辑》;20030315;论文正文第15页第2段-第34页第13段,附图2.6-2.11 *

Also Published As

Publication number Publication date
CN103853772A (en) 2014-06-11

Similar Documents

Publication Publication Date Title
CN103020204B (en) A kind of method and its system carrying out multi-dimensional interval query to distributed sequence list
CN102999519B (en) Read-write method and system for database
CN107045531A (en) A kind of system and method for optimization HDFS small documents access
CN103853772B (en) High-efficiency reverse index organizing method
CN102629269B (en) Searching and storing method for embedded database
CN102880615B (en) A kind of date storage method and device
CN102024047B (en) Data searching method and device thereof
CN102156711B (en) Cloud storage based power full text retrieval method and system
CN106055621A (en) Log retrieval method and device
CN102402617A (en) Easily compressed database index storage system using fragments and sparse bitmap, and corresponding construction, scheduling and query processing methods
CN106844555A (en) A kind of time series data storage method for power network WAMS systems
CN102446184B (en) Industrial data storage and index method based on time series
CN101799808A (en) Data processing method and system thereof
CN102819547A (en) Method and device for configuring field attribute
CN103488709A (en) Method and system for building indexes and method and system for retrieving indexes
CN103412889A (en) Data storage and query method of intelligent electric meter and system thereof
CN107577436A (en) A kind of date storage method and device
CN101551807A (en) Multilevel index technology for file database
CN102622434A (en) Data storage method, data searching method and device
CN101751475B (en) Method for compressing section records and device therefor
CN104536700B (en) Quick storage/the read method and system of a kind of bit stream data
CN102169499A (en) Repeated ticket removing method
CN107423321A (en) It is applicable the method and its device of high-volume small documents cloud storage
CN100585597C (en) File organization method
CN107273443B (en) Mixed indexing method based on metadata of big data model

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: 100088 Beijing city Haidian District No. 6 Zhichun Road Jinqiu International Building 14 floor 14B04

Patentee after: TOLS INFORMATION TECHNOLOGY Co.,Ltd.

Address before: 100088 Beijing city Haidian District No. 6 Zhichun Road Jinqiu International Building 14 floor 14B04

Patentee before: BEIJING TRS INFORMATION TECHNOLOGY Co.,Ltd.