CN109726305A - A kind of complex_relation data storage and search method based on graph structure - Google Patents
A kind of complex_relation data storage and search method based on graph structure Download PDFInfo
- Publication number
- CN109726305A CN109726305A CN201811643795.2A CN201811643795A CN109726305A CN 109726305 A CN109726305 A CN 109726305A CN 201811643795 A CN201811643795 A CN 201811643795A CN 109726305 A CN109726305 A CN 109726305A
- Authority
- CN
- China
- Prior art keywords
- node
- relationship
- pointer
- complex
- attribute
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of complex_relation data storage and search method based on graph structure, including constructing complex_relation data storage model, there are multiple nodes in the model, have side to be connected with each other between different nodes, wherein the node table shows entity object, relationship between the side presentation-entity, each node have the multiple attributes of itself, and the mode that the relationship and the attribute are all made of doubly linked list is stored.The present invention combines the storage method of the node structure based on entity relationship, the date storage method of side structure based on graph structure, and combine the building hybrid index structure of node structure and side structure, support the quick indexing of large-scale graph data, and scalability is supported well, using a pair of of relationship of relational database memory node and side, the disadvantage for overcoming chart database committed memory excessive realizes the quick search of complex_relation data.
Description
Technical field
The application belongs to database technology and diagram data management domain, is suitable for specifically, being related to one kind based on graph structure
Complex_relation data storage and search method.
Background technique
Complex_relation data is a kind of important kind of big data, such as social network data, scientific and technological information data.Very much
Typical case aiming at complex_relation data analysis and excavation, such as community discovery, scientific research relationship inquiry.It is closed for complexity
The storage and retrieval of coefficient evidence is the basis for realizing above-mentioned function.Therefore, research complex_relation data method for quickly retrieving has
Important meaning.
There are two types of forms for complex_relation data storage, the first is stored based on relational database, and another kind is diagram data
Library storage.Complex relationship relationship sheet form is converted to based on relational database storage to be stored in entity relationship scheme；Based on figure
Database purchase indicates and stores complex_relation data according to the structure and attribute of figure.Relational database technology comparative maturity,
But being to look for complex relationship needs multilist JOIN to operate, inefficient；Although chart database technology, which searches complex relationship, compares appearance
Easily, but insertion speed is slow, expends memory, causes data directory convergence slow.
It in the prior art, will in patent document 1 " a kind of diagram data storage method and subgraph query method based on external memory "
Complex_relation data is abstracted into figure, is made of different type side and point, to pressing in figure when carrying out classification storage and to every class
B+-Tree index is established according to lexcographical order, and point index of the picture and terminal bitmap are set up according to the starting point on side, terminal label information
Index.This method can also be applicable in small-scale data volume, but can not be suitable for big index of the picture.Because with a B+-Tree
Support one big figure infeasible in reality, and the situation that bitmap index is less suitble to element very big.Patent document 2 is " a kind of
Graph data query method and device ", by figure vertex subregion and layering, the search efficiency of Lai Tigao diagram data.It is this
Method can not be in face of the quick variation of complex_relation data and the sparse characteristic of big figure, i.e. some nodes and the performance of frontier juncture system complexity
Very dense, many nodes and side are very sparse without especially more relationship performances.
Therefore, the shortcomings that high cost JOIN is operated when how to overcome relational database query complex relationship, and overcome
The excessive disadvantage of chart database committed memory, realizes the quick search of complex_relation data, becomes prior art urgent need to resolve
Technical problem.
Summary of the invention
It is an object of the invention to propose that a kind of graph structure that is based on is suitable for complex_relation data storage and search method, lead to
The complex relationship for crossing graph structure storing data, supports the quick indexing of large-scale graph data, and supports scalability well.
To achieve this purpose, the present invention adopts the following technical scheme:
A kind of complex_relation data storage and search method based on graph structure, it is characterised in that:
Including constructing complex_relation data storage model, the storage model is graph structure, is had in the model multiple
Node has side to be connected with each other between different nodes, wherein the node table shows entity object, the side presentation-entity
Between relationship, each node has itself multiple attributes, wherein the relationship and the attribute are all made of the mode of chained list
It is saved.
Optionally, the mode that the relationship and the attribute are all made of doubly linked list is stored.
Optionally, in store the 1st relationship for belonging to the entity object of the node and the 1st attribute.It should by inquiry
Relationship, so that it may all relationships for belonging to the entity object are found along the chained list of relationship.Likewise, passing through the 1st category of inquiry
Property, so that it may along the chained list of attribute, find all properties for belonging to the entity object.
Optionally, the side start node and terminal node composition, and start node includes the preposition relationship of the node
Pointer (Prev) and postposition relationship pointer (Next), terminal node include the preposition relationship pointer of the node (Prev) and postposition relationship
Pointer (Next), preposition pointer are directed toward the upper relationship with starting point, and postposition pointer is directed toward the latter relation of the same starting point.
Optionally, all relationships of start node form a doubly linked list, and all relationships of terminal node form one
Doubly linked list.
Optionally, establish and hybrid index structure formed by node structure storage and relational structure, according to relationship type for
Node is grouped, and entity object node is ranked up with certain sequence, forms the index structure based on attribute of a relation.
Optionally, when being retrieved, query node first inquires corresponding first relationship of the node；Then, root
According to the relationship inquired, search relationship storage organization finds the beginning and end of node, along starting point preposition pointer or after
The corresponding relation of node can be found by setting pointer.
Optionally, the graph structure of the complex_relation data storage model is stored in relational database, is connected by table
Operation is connect, the relationship of figure interior joint and node is found.
Therefore, the present invention has the advantage that combining the storage method of the node structure based on entity relationship, based on figure
The date storage method of the side structure of structure, and the building hybrid index structure of node structure and side structure is combined, support big rule
The quick indexing of mould diagram data, and support scalability well, using a pair of of relationship of relational database memory node and side, gram
The excessive disadvantage of chart database committed memory is taken, realizes the quick search of complex_relation data.
Detailed description of the invention
Fig. 1 is the storage model of the complex_relation data based on graph structure of specific embodiment according to the present invention；
Fig. 2 is the storage method of relationship and attribute in the node of specific embodiment according to the present invention；
Fig. 3 is the storage organization of the relationship of specific embodiment according to the present invention；
Fig. 4 is the data query example of the complex relationship based on graph structure of specific embodiment according to the present invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
The present invention establishes the storage method of the node structure based on entity relationship, and multiple nodes are connected by entity relationship
It is connected together, establishes the date storage method of the side structure based on graph structure, and combine the building of node structure and side structure mixed
Index structure is closed, to provide complex_relation data querying method.
Referring to figures 1-3, respectively illustrate the storage model of the complex_relation data based on graph structure, in node relationship with
The storage method of attribute and the storage organization of relationship.
Specifically, a kind of complex_relation data storage and search method based on graph structure, including building complex relationship number
According to storage model, the storage model is graph structure, has multiple nodes in the model, has Bian Yijin between different nodes
Row is connected with each other, wherein the node table shows that entity object, the relationship between the side presentation-entity, each node have itself
Attribute, wherein the mode that the relationship and the attribute are all made of chained list is saved.Therefore, it is by complicated Relationship Change
One graph structure.
Further, the mode that the relationship and the attribute are all made of doubly linked list is stored.That is two nodes
All relationships form a doubly linked list.
Specifically, referring to fig. 2, in store the 1st relationship for belonging to the entity object of the node and the 1st attribute.It is logical
It crosses and inquires the relationship, so that it may find all relationships for belonging to the entity object along the chained list of relationship.Likewise, passing through inquiry
1st attribute, so that it may along the chained list of attribute, find all properties for belonging to the entity object.
Referring to Fig. 3, the side start node and terminal node composition, and start node includes the preposition relationship of the node
Pointer (Prev) and postposition relationship pointer (Next), terminal node include the preposition relationship pointer of the node (Prev) and postposition relationship
Pointer (Next), preposition pointer are directed toward the upper relationship with starting point, and postposition pointer is directed toward the latter relation of the same starting point.
All relationships of start node form a doubly linked list, and all relationships of terminal node form a Two-way Chain
Table.
Further, storage of the invention and search method, which are also set up, forms mixing by node structure storage and relational structure
Index structure is grouped node according to relationship type, and entity object node is ranked up with certain sequence, this
Sample just forms the index structure based on attribute of a relation.
It establishes the index of entity object node respectively according to different relationships, includes with this relationship under each index
The entity node list of attribute, the list of entities are arranged according to certain sequence, in an alternative embodiment, suitable according to letter
Sequence sequence.
Querying method of the invention is: query node first, inquires corresponding first relationship of the node；Then,
According to the relationship inquired, search relationship storage organization finds the beginning and end of node, along starting point preposition pointer or
Postposition pointer can find the corresponding relation of node.
Illustratively, referring to fig. 4, all relationships that there is the big treasured of father node king of set membership to have for inquiry are shown
For attribute.
There is following index according to different relationships in the figure:
Set membership (father): node A, node E
Set membership (son): node B, node C
Peer Relationships: node D, node E
Conjugal relation (husband): node A, node E
Conjugal relation (wife): node D, node C
Query process is as follows:
(1) index structure based on attribute is inquired according to set membership, inquires some entity node, is here node A.
(2) storage organization of the entity node is read, as shown in Fig. 2, inquiring first relationship R1 of the entity node
With first attribute P1 (herein are as follows: name attribute).
(3) storage organization for reading relationship R1, as shown in figure 3, the preposition pointer Prev and postposition along the starting point of R1 refer to
Needle Next traverses relationship doubly linked list, reads all relationships for belonging to start node.In this example, developing the preposition pointer of R1 is
The postposition pointer that the postposition relationship of postposition pointer points relationship R3, R3 of NULL, R1 are directed toward R7, R7 is directed toward NULL, obtains entity
All relationships of Object node A are R1, R3, R7.
(4) similar, the storage organization of reading attributes P1 develops the preposition pointer Prev and postposition pointer Next of P1, time
Go through the doubly linked list of attribute.
Read all properties for belonging to start node.So far, inquiry is completed.The preposition pointer Prev of P1 is directed toward NULL, P1
Postposition pointer be directed toward attribute gender, read gender attribute: male；The Next pointer of attribute gender is directed toward the attribute age, reads year
Age attribute: 32；The Next pointer at attribute age is directed toward NULL, and so far obtaining all properties name of entity object node A, (king is big
It is precious), gender (male), age (32).
Further, the present invention also has following improvement, can store the graph structure of complex_relation data storage model
In relational database, by table attended operation, the relationship of figure interior joint and node is found.
Therefore, the present invention has the advantage that combining the storage method of the node structure based on entity relationship, based on figure
The date storage method of the side structure of structure, and the building hybrid index structure of node structure and side structure is combined, support big rule
The quick indexing of mould diagram data, and support scalability well, using a pair of of relationship of relational database memory node and side, gram
The excessive disadvantage of chart database committed memory is taken, realizes the quick search of complex_relation data.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that
A specific embodiment of the invention is only limitted to this, for those of ordinary skill in the art to which the present invention belongs, is not taking off
Under the premise of from present inventive concept, several simple deduction or replace can also be made, all shall be regarded as belonging to the present invention by institute
Claims of submission determine protection scope.
Claims (8)
1. a kind of complex_relation data storage and search method based on graph structure, it is characterised in that:
Including constructing complex_relation data storage model, the storage model is graph structure, has multiple nodes in the model,
Have side to be connected with each other between different nodes, wherein the node table shows entity object, between the side presentation-entity
Relationship, each node has itself multiple attributes, wherein the mode that the relationship and the attribute are all made of chained list carries out
It saves.
2. storage according to claim 1 and search method, it is characterised in that:
The mode that the relationship and the attribute are all made of doubly linked list is stored.
3. storage according to claim 1 and search method, it is characterised in that:
In store the 1st relationship for belonging to the entity object of the node and the 1st attribute.By inquiring the relationship, so that it may
All relationships for belonging to the entity object are found along the chained list of relationship.Likewise, passing through the 1st attribute of inquiry, so that it may edge
The chained list of attribute, find all properties for belonging to the entity object.
4. storage according to claim 1 and search method, it is characterised in that:
The side start node and terminal node composition, and start node include the preposition relationship pointer of the node (Prev) and
Postposition relationship pointer (Next), terminal node include the preposition relationship pointer of the node (Prev) and postposition relationship pointer (Next),
Preposition pointer is directed toward the upper relationship with starting point, and postposition pointer is directed toward the latter relation of the same starting point.
5. storage according to claim 4 and search method, it is characterised in that:
All relationships of start node form a doubly linked list, and all relationships of terminal node form a doubly linked list.
6. storage according to claim 1 and search method, it is characterised in that:
It establishes and hybrid index structure is formed by node structure storage and relational structure, node is divided according to relationship type
Group, and entity object node is ranked up with certain sequence, form the index structure based on attribute of a relation.
7. storage according to claim 1 and search method, it is characterised in that:
When being retrieved, query node first inquires corresponding first relationship of the node；Then, according to inquiring
Relationship, search relationship storage organization find the beginning and end of node, along the preposition pointer or postposition pointer of starting point
Find the corresponding relation of node.
8. storage according to claim 1 and search method, it is characterised in that:
The graph structure of the complex_relation data storage model is stored in relational database, by table attended operation, is searched
To the relationship of figure interior joint and node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811643795.2A CN109726305A (en) | 2018-12-30 | 2018-12-30 | A kind of complex_relation data storage and search method based on graph structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811643795.2A CN109726305A (en) | 2018-12-30 | 2018-12-30 | A kind of complex_relation data storage and search method based on graph structure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109726305A true CN109726305A (en) | 2019-05-07 |
Family
ID=66299530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811643795.2A Pending CN109726305A (en) | 2018-12-30 | 2018-12-30 | A kind of complex_relation data storage and search method based on graph structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109726305A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125265A (en) * | 2019-12-13 | 2020-05-08 | 四川蜀天梦图数据科技有限公司 | Method and device for generating mapping data based on relational database data |
| CN111523000A (en) * | 2020-04-23 | 2020-08-11 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for importing data |
| CN112116951A (en) * | 2020-08-14 | 2020-12-22 | 中国科学院计算技术研究所 | Proteome data management method, medium and equipment based on graph database |
| CN112348934A (en) * | 2020-10-20 | 2021-02-09 | 珠海金山网络游戏科技有限公司 | Game map display method, device and medium based on large linked list |
| WO2021083239A1 (en) * | 2019-10-28 | 2021-05-06 | 北京大学 | Graph data query method and apparatus, and device and storage medium |
| CN112765418A (en) * | 2021-04-08 | 2021-05-07 | 中译语通科技股份有限公司 | Alias merging and storing method, system, terminal and medium based on graph structure |
| CN112836063A (en) * | 2021-01-27 | 2021-05-25 | 四川新网银行股份有限公司 | Method for realizing feature tracing |
| CN113448964A (en) * | 2021-06-29 | 2021-09-28 | 四川蜀天梦图数据科技有限公司 | Hybrid storage method and device based on graph-KV |
| CN113961754A (en) * | 2021-09-08 | 2022-01-21 | 南湖实验室 | Graph database system based on persistent memory |
| CN114996297A (en) * | 2022-04-14 | 2022-09-02 | 建信金融科技有限责任公司 | Data processing method, device, equipment, medium and product |
| CN115658329A (en) * | 2022-12-22 | 2023-01-31 | 杭州欧若数网科技有限公司 | Method, system and medium for optimizing memory of graph data structure |
| WO2023160137A1 (en) * | 2022-02-25 | 2023-08-31 | 苏州浪潮智能科技有限公司 | Graph data storage method and system, and computer device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6665863B1 (en) * | 2000-05-31 | 2003-12-16 | Microsoft Corporation | Data referencing within a database graph |
| CN105608232A (en) * | 2016-02-17 | 2016-05-25 | 扬州大学 | Bug knowledge modeling method based on graphic database |
| CN106383879A (en) * | 2016-09-13 | 2017-02-08 | 深圳市华傲数据技术有限公司 | Method and system for handling government administration data |
| CN108549731A (en) * | 2018-07-11 | 2018-09-18 | 中国电子科技集团公司第二十八研究所 | A kind of knowledge mapping construction method based on ontology model |
-
2018
- 2018-12-30 CN CN201811643795.2A patent/CN109726305A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6665863B1 (en) * | 2000-05-31 | 2003-12-16 | Microsoft Corporation | Data referencing within a database graph |
| CN105608232A (en) * | 2016-02-17 | 2016-05-25 | 扬州大学 | Bug knowledge modeling method based on graphic database |
| CN106383879A (en) * | 2016-09-13 | 2017-02-08 | 深圳市华傲数据技术有限公司 | Method and system for handling government administration data |
| CN108549731A (en) * | 2018-07-11 | 2018-09-18 | 中国电子科技集团公司第二十八研究所 | A kind of knowledge mapping construction method based on ontology model |
Non-Patent Citations (5)
| Title |
|---|
| BONELEE: "Neo4j图数据库简介和底层原理", 《HTTPS://WWW.CNBLOGS.COM/BONELEE/P/6211290.HTML》 * |
| JEWEL_JIA: "图数据存储初见", 《HTTPS://WWW.CNBLOGS.COM/JEWELL/P/5789270.HTML》 * |
| WEB8: "neo4j中索引的使用", 《HTTPS://WWW.CNBLOGS.COM/WEB100/P/NEO4J.HTML》 * |
| 王莉: "《情报研究前沿聚焦》", 31 December 2017 * |
| 范仁义: "Neo4j图数据库简介和底层原理", 《HTTPS://WWW.CNBLOGS.COM/RENYI-FAN/P/7833274.HTML》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021083239A1 (en) * | 2019-10-28 | 2021-05-06 | 北京大学 | Graph data query method and apparatus, and device and storage medium |
| CN111125265B (en) * | 2019-12-13 | 2020-10-02 | 四川蜀天梦图数据科技有限公司 | Method and device for generating mapping data based on relational database data |
| CN111125265A (en) * | 2019-12-13 | 2020-05-08 | 四川蜀天梦图数据科技有限公司 | Method and device for generating mapping data based on relational database data |
| CN111523000A (en) * | 2020-04-23 | 2020-08-11 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for importing data |
| CN112116951A (en) * | 2020-08-14 | 2020-12-22 | 中国科学院计算技术研究所 | Proteome data management method, medium and equipment based on graph database |
| CN112116951B (en) * | 2020-08-14 | 2023-04-07 | 中国科学院计算技术研究所 | Proteome data management method, medium and equipment based on graph database |
| CN112348934A (en) * | 2020-10-20 | 2021-02-09 | 珠海金山网络游戏科技有限公司 | Game map display method, device and medium based on large linked list |
| CN112836063A (en) * | 2021-01-27 | 2021-05-25 | 四川新网银行股份有限公司 | Method for realizing feature tracing |
| CN112765418B (en) * | 2021-04-08 | 2022-04-01 | 中译语通科技股份有限公司 | Alias merging and storing method, system, terminal and medium based on graph structure |
| CN112765418A (en) * | 2021-04-08 | 2021-05-07 | 中译语通科技股份有限公司 | Alias merging and storing method, system, terminal and medium based on graph structure |
| CN113448964A (en) * | 2021-06-29 | 2021-09-28 | 四川蜀天梦图数据科技有限公司 | Hybrid storage method and device based on graph-KV |
| CN113448964B (en) * | 2021-06-29 | 2022-10-21 | 四川蜀天梦图数据科技有限公司 | Hybrid storage method and device based on graph-KV |
| CN113961754A (en) * | 2021-09-08 | 2022-01-21 | 南湖实验室 | Graph database system based on persistent memory |
| WO2023160137A1 (en) * | 2022-02-25 | 2023-08-31 | 苏州浪潮智能科技有限公司 | Graph data storage method and system, and computer device |
| CN114996297A (en) * | 2022-04-14 | 2022-09-02 | 建信金融科技有限责任公司 | Data processing method, device, equipment, medium and product |
| CN114996297B (en) * | 2022-04-14 | 2023-09-26 | 建信金融科技有限责任公司 | Data processing method, device, equipment and medium |
| CN115658329A (en) * | 2022-12-22 | 2023-01-31 | 杭州欧若数网科技有限公司 | Method, system and medium for optimizing memory of graph data structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109726305A (en) | A kind of complex_relation data storage and search method based on graph structure | |
| CN106227800B (en) | Storage method and management system for highly-associated big data | |
| CN103345521B (en) | A kind of method and apparatus processing key assignments in Hash table database | |
| CN111190904B (en) | Method and device for hybrid storage of graph-relational database | |
| CN106933833B (en) | Method for quickly querying position information based on spatial index technology | |
| CN103561133B (en) | A kind of IP address attribution information index method and method for quickly querying | |
| CN102662974B (en) | A network graph index method based on adjacent node trees | |
| CN111309868B (en) | Knowledge graph construction and retrieval method and device | |
| CN108829880B (en) | Method for configuration management of optical network terminal equipment | |
| CN103064908B (en) | A kind of method by the quick duplicate removal list of internal memory | |
| EP3767486B1 (en) | Multi-record index structure for key-value stores | |
| CN104731945A (en) | Full-text searching method and device based on HBase | |
| CN110888930A (en) | Financial knowledge inquiry service interface design and implementation method based on knowledge map | |
| CN104615734B (en) | A kind of community management service big data processing system and its processing method | |
| WO2015058500A1 (en) | Data storage method and device | |
| CN102571752A (en) | Service-associative-index-map-based quality of service (QoS) perception Top-k service combination system | |
| CN101963993B (en) | Method for fast searching database sheet table record | |
| CN102779186A (en) | Whole process modeling method of unstructured data management | |
| CN114860727A (en) | Zipper watch updating method and device | |
| CN103365960A (en) | Off-line searching method of structured data of electric power multistage dispatching management | |
| CN116150436B (en) | Data display method and system based on node tree | |
| CN112540987A (en) | Big data management system of distribution and utilization electricity based on data mart | |
| CN106933844B (en) | Construction method of reachability query index facing large-scale RDF data | |
| CN109271350B (en) | Database and information point table automatic comparison and synchronization method based on telecontrol communication | |
| CN110263108A (en) | A kind of keyword Skyline fuzzy query method and system based on road network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190507 |