当前位置：网站首页>Neo4j: basic introduction (I) installation and use

Neo4j: basic introduction (I) installation and use

2022-06-27 12:35:00 【stay_ foolish12】

Link to the original text ：https://blog.csdn.net/sinat_36226553/article/details/108541370

# Graph database

link ： What is primordial (Native) Graph database

Generally considered to have “ No index adjacency ” Feature graph databases are called native graph databases

link ： Common graph databases

Graph storage can be divided into attribute graphs 、 Triple storage and hybrid system [ source ]

Property Graphs

Triple Stores and Semantic Databases

Hybrid and Exotic Systems

# Neo4j Introduce

Neo4j Is a high-performance NOSQL Graphic database , It stores structured data on the network instead of tables .Neo4j It can also be seen as a high-performance graph engine , The engine has all the features of a mature database .

There are two basic data types in a graph ：Nodes（ node ） and Relationships（ Relationship ）.Nodes and Relationships contain key/value Properties of form .Nodes adopt Relationships The relationships defined are connected , Form a relational network structure . Its form is as follows ：

be based on neo4j This storage feature of , Often use it to store owl、rdf file .

Neo4j It's open source. , A lot of programming languages , Such as Node.js、Python、PHP、Ruby、.NET And so on Neo4j Support for . Such as Node.js It can be done by command “ npm install neo4j-driver ” stay Node.js Install in Neo4j Driver;python Can pass " pip install py2neo " Install dependency packages .

# install

Win10 install Neo4j

1、 Prerequisite needs

install JAVA SE Of JDK（cmd Input java -version Check that it is installed

Be careful ：jdk and neo4j Version matching is required [ link ] （ Because of my jdk yes 1.8, Choose install... Here neo4j-3.5.3,neo4j 4.x The above version requires java11

2、 Version difference

Community Edition （Community Edition）： Basic 、 High performance graphic database with complete functions .

Enterprise Edition （Enterprise Edition）： Compared with the community version, the enterprise version has added many typical enterprise functions , Like clusters 、 Advanced monitoring 、 Advanced caching 、 Online backup, etc

notes ：Neo4j Most users start to develop applications from the community version , Then use the enterprise edition in the production environment .Neo4j The enterprise edition is charged , Only enterprise edition can use distributed installation , And it can provide load balancing, high availability configuration and other functions . The Community Edition can only be used on a stand-alone computer , Up to billions of nodes can be used 、 Relationships and attributes .

《 Knowledge map ： Method 、 Practice and Application 》：Neo4j The disadvantage of is that the community version is a stand-alone system , although Neo4j The enterprise edition supports high availability （High availability） colony , But its biggest difference from the distributed graph storage system is that the complete copy of the graph database is stored on each node （ Replica cluster similar to relational database mirroring ）, Instead of dividing graph data into subgraphs for distributed storage , It is not a real distributed database system . If the graph data exceeds a certain scale , System performance will be affected by disk 、 Memory and other restrictions

3、 The installation process

Official website address ：https://neo4j.com/download-center （ Up to now, the latest version is 4.1.1

Domestic mirror ：http://doc.we-yun.com:1008/neo4j

Unzip to custom Directory , As I am W:\Neo4j\neo4j-community-3.5.3

Neo4j The application has the following main directory structure ：

bin Catalog ： Used to store Neo4j Executable program of ;
conf Catalog ： Used to control the Neo4j Boot profile ;
data Catalog ： Used to store core database files ;
plugins Catalog ： Used to store Neo4j Plug in for ;

Configure environment variables NEO4J_HOME, And in Path Add %NEO4J_HOME%\bin

open cmd Command line , Enter the command ：neo4j console （ Sometimes the command is not recognized , Open in administrator mode instead cmd

notes ： stop it neo4j The server cmd type CTRL+C that will do

In the browser. http://localhost:7474 View in Neo4j database （ default host yes bolt://localhost:7687, The default user is neo4j, The default password is ：neo4j, First successful login to Neo4j After the server , Password reset required

If you forget the password , The solution is to stop the service , Get into /data/dbms Catalog , Delete auth file

The other way ： Use exe Files installed Neo4j

Linux install Neo4j

Use Linux System , The minimum requirement is to use EXT4 file system （ Or similar ）. Recommended EXT4、ZFS Equal file system .CentOS 6.5 It uses EXT4 file system ,CentOS 7 It uses XFS file system .XFS It's a ratio. EXT4 More advanced file systems , therefore CentOS 6.5 The above versions meet the requirements . namely Linux The operating system can use Ubuntu、Debian or CentOS 6.5 And above .

CentOS 6.9 course ：https://blog.csdn.net/sinat_36226553/article/details/108583059

Chinese version Neo4j

Cloud clustering （ Beijing ） Ltd. independently researched and developed Neo4j Simplified Chinese version . This version provides Neo4j Simplified Chinese interface , It also provides a set of extended functions that are especially suitable for Chinese users .

Neo4j Simplified Chinese version Download address ：http://we-yun.com/index.php/blog/release-56.html （ At present, we support Windows、Linux

import wizard ： A support will Excel、MySQL and Oracle Such data can be easily imported into Neo4j Tools for

import wizard Download address ：http://we-yun.com:8000/ToNeo4j/

# Neo4j To configure

The configuration document is stored in conf Under the table of contents ,Neo4j Through the configuration file neo4j.conf Control the work of the server . By default , No arbitrary configuration is required , You can start the server .

Neo4j Figure detailed explanation of database configuration file

1. Location of the core data file

for example , Where the core data files are stored , The default is in data/graph.db Directory , To change the default storage directory , Configuration options can be updated ：

# The name of the database to mount
#dbms.active_database=graph.db

2. Security verification , Enabled by default

# Whether requests to Neo4j are authenticated.
# To disable authentication, uncomment this line
#dbms.security.auth_enabled=false

Cancel the comment , The console will no longer output the warning shown in the following figure

3. To configure JAVA The size of heap memory

If there's not enough memory , Will report a mistake ：Neo4j bulk import “neo4j-admin import” OutOfMemoryError: Java heap space and OutOfMemoryError: GC overhead limit exceeded


   
    
      
       
        
       
       
       
         # Java Heap Size: by default the Java heap size is dynamically 
        
      
      
       
        
       
       
       
         # calculated based on available system resources. 
        
      
      
       
        
       
       
       
         # Uncomment these lines to set specific initial and maximum heap size. 
        
      
      
       
        
       
       
       
         #dbms.memory.heap.initial_size=512m 
        
      
      
       
        
       
       
       
         #dbms.memory.heap.max_size=512m

4. Configure access

Neo4j The default configuration of only supports local access , To enable it to support remote access , Relevant configurations can be modified .

Neo4j Three different connection modes are supported ：Bolt、HTTP、HTTPS.Neo4j Of Web The client uses HTTP How to connect , Generally, it can be used in program development HTTP and Bolt How to connect .HTTPS Security certificate support is required , If you want to make the database open for access in the Internet , You can use this connection .

5. Load balancing

When using graph database clustering , We should consider the load balancing of the cluster , This maximizes throughput and reduces latency .Neo4j It has no load balancing function , It depends on the load balancing capability of the network infrastructure .

# Cypher

image Oracle Database has query language SQL,Neo4j have Cypher As a query language .

Cypher Is a declarative pattern matching query language , Make diagram database management easy to understand 、 Easy to operate , Even less skilled users can use Cypher Easy operation diagram database .

neo4j Of Cypher Command box （ To wrap a line, press Shift + Enter、Cypher use // Annotate

grammar

If we want to use it Cypher To express Romeo love Juliet , that Cypher The grammar is as follows ：

(romeo:Person{name:"Romeo"}) - [:LOVES] -> (juliet:Person{name:"Juliet"})

namely

(NODE1) - [:RELATION] -> (NODE2)

Grammatical conventions

Nodes are humped in lowercase ; The label is in capital letters ; The relation is capitalized in serpentine form （ similar IS_A）; Attribute names are in lowercase humps ; All keywords are capitalized .

About case

Case sensitive ： Node labels 、 Type of relationship 、 Property name / key

Case insensitive ：Cypher keyword

Create data （CREATE）

stay neo4j Create the node and its relationships . We created a node romeo, It has Person Label and a character type name attribute , The property value is Romeo. Then I created a new one called LOVES Is connected to another node juliet

CREATE (romeo:Person{name:"Romeo"}) - [:LOVES] - > (juliet:Person{name:"Juliet"})

Reading data （READ）


   
    
      
       
        
       
       
        
        MATCH (n:Person) 
        - [:LOVES] 
        - () 
        
      
      
       
        
       
       
        
        WHERE toLower(n.name) 
        ="juliet" 
        
      
      
       
        
       
       
        
        RETURN n 
        ORDER 
        BY n.age 
        ASC 
        
      
      
       
        
       
       
        
        SKIP 
        2 LIMIT 
        5; 
        
      
      
       
        
       
       
        
        
      
      
       
        
       
       
       
         # 
        asc From small to large , No, age The node of the attribute will be placed at the end of the list  
        
      
      
       
        
       
       
       
         # 
        SKIP and LIMIT Used to page the results

Query a node according to a certain attribute

Query all attributes of the node

Query all keys of the node

Update data （UPDATE）


   
    
      
       
        
       
       
        
        MATCH (n:Person) 
        
      
      
       
        
       
       
        
        SET n.age 
        =n.age 
        + 
        1 
        
      
      
       
        
       
       
        
        RETURN n;

Delete data （DELETE）


   
    
      
       
        
       
       
       
         MATCH (r:Person{name: 
        'romeo'}) 
        
      
      
       
        
       
       
       
         DELETE r;

MERGE

MERGE More like MATCH and CREATE Mixing . If MERGE The pattern described in the statement already exists in the diagram , The identifier described by the statement will be bound to the existing data , This is like comparison. MATCH; If the schema it describes does not exist in the diagram , that MERGE It will be created first , This is like comparison. CREATE.

understand CQL For more commands, click [ link ]

Dr. yufanghua's series of tutorials

Neo4j Cypher Figure database query language quick reference - 【1】 Basic grammar : PDF

Neo4j Cypher Figure database query language quick reference - 【2】 Create and delete nodes 、 Relationships and attributes ：PDF

Neo4j Cypher Figure database query language quick reference - 【3】 Indexes and restrictions ：PDF

# Figure database and relational database

Native graph storage and native graph processing are not necessarily better or worse than other methods —— This is just a typical engineering trade-off . In addition to adopting specific storage and processing methods , The graph database also uses a specific data model . Several different graph data models are commonly used , Including attribute maps 、 Hypergraphs and triples .

The benefits of native graph storage are , Its stack is specifically designed and built for performance and scalability . But the relative , Non native graph storage is usually built on a very mature non graph back end （ Such as MySQL） above , The O & M team is familiar with their features . Native map processing （ Index free adjacency ） Although the performance advantage of traversal query is great , But the cost is that some queries that do not use traversal will be more difficult , And it takes up a lot of memory .

although E-R A graph is a graph , But we can immediately see the weakness of relational models in modeling complex domains . Although it can name contacts like a graph database （ Relational databases don't have this feature ）, But between two entities ,E-R The graph only allows us to create one 、 Undirected named connections . In this respect , The connections between real-world entities are colorful 、 A wide variety , The relational model is not suitable for representing .