Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
2 contributors

Users who have contributed to this file

@Amber1990Zhang @whitewum
306 lines (234 sloc) 8.25 KB

Quick Start

Try Nebula Graph from Docker

The easiest way to get Nebula Graph up and running is using Docker. Before you start, make sure that you have:

  • Installed the latest version of Docker

  • Pulled the latest images of Nebula from Nebula Docker Hub. If not, pull the images using the following command:

> docker pull vesoft/nebula-graph:nightly

Try Nebula Graph Deployment From Docker Compose

The easiest way to try nebula cluster is using docker compose. For details please refer nebula docker deployment.

Startup Service

When nebula image is ready, run a container:

> docker run -it vesoft/nebula-graph:nightly /bin/bash

After login, you're in the root directory and you should switch to the nebula directory

> cd /usr/local/nebula/

Start meta service, storage service and graph service:

> scripts/nebula.service start all

Check services' status:

> scripts/nebula.service status all

Connect to Service

> bin/nebula -u=user -p=password
  • -u is to set the user name, user is the default Nebula user account
  • -p is to set password, password is the default password for account user

Further more, if you run the console (bin/nebula) in a remote host from graphd, please specify the server's ip and port:

> bin/nebula -u=user -p=password -addr={graphd ip} --port={graphd port}

If you have any questions or concerns about the deployment procedures, please do not hesitate to open an issue on GitHub.

Build Your Own Graph

This section describes how to build a graph and run queries. The example is built on the graph below:

image

There are three kinds of tags (course, building and student) and two edge types (choose and follow).

Create a Graph Space

First, we need to create a space and use it before other operations.

To list all existing spaces:

nebula> SHOW SPACES;

To create a new space named test:

nebula> CREATE SPACE test(partition_num=1024, replica_factor=1);
-- Use this space
nebula> USE test;
  • replica_factor specifies the number of replicas in the cluster.
  • partition_num specifies the number of partitions in one replica.

Define Graph Schema

The CREATE TAG statement defines a tag, with a type name and an attribute list.

nebula> CREATE TAG course (name string, credits int);
nebula> CREATE TAG building (name string);
nebula> CREATE TAG student (name string, age int, gender string);

The CREATE EDGE statement defines an edge type.

nebula> CREATE EDGE follow(likeness double);
nebula> CREATE EDGE choose(grade int);

To list the tags and edge types that we just created:

-- Show tag list
nebula> SHOW TAGS;
============
| Name     |
============
| building |
------------
| course   |
------------
| student  |
------------
-- Show edge type list
nebula> SHOW EDGES;
==========
| Name   |
==========
| choose |
----------
| follow |
----------

To show the attributes of a tag or an edge type:

-- Show attributes of a tag
nebula> DESCRIBE TAG student;
===================
| Field  | Type   |
===================
| name   | string |
-------------------
| age    | int    |
-------------------
| gender | string |
-------------------
-- Show attributes of an edge type
nebula> DESCRIBE EDGE follow;
=====================
| Field    | Type   |
=====================
| likeness | double |
---------------------

Insert Data

Insert the vertices and edges based on the graph above.

-- Insert vertices
nebula> INSERT VERTEX student(name, age, gender) VALUES 200:("Monica", 16, "female");
nebula> INSERT VERTEX student(name, age, gender) VALUES 201:("Mike", 18, "male");
nebula> INSERT VERTEX student(name, age, gender) VALUES 202:("Jane", 17, "female");
nebula> INSERT VERTEX course(name, credits),building(name) VALUES 101:("Math", 3, "No5");
nebula> INSERT VERTEX course(name, credits),building(name) VALUES 102:("English", 6, "No11");

-- Insert edges
nebula> INSERT EDGE choose(grade) VALUES 200 -> 101:(5);
nebula> INSERT EDGE choose(grade) VALUES 200 -> 102:(3);
nebula> INSERT EDGE choose(grade) VALUES 201 -> 102:(3);
nebula> INSERT EDGE choose(grade) VALUES 202 -> 102:(3);
nebula> INSERT EDGE follow(likeness) VALUES 200 -> 201:(92.5);
nebula> INSERT EDGE follow(likeness) VALUES 201 -> 200:(85.6);
nebula> INSERT EDGE follow(likeness) VALUES 201 -> 202:(93.2);

Update Data

-- Update vertex
nebula> UPDATE VERTEX 101 SET course.credits = $^.course.credits + 1, building.name = "No8" YIELD $^.course.name AS Name, $^.course.credits AS Credits, $^.building.name
-- Update edge
nebula> UPDATE EDGE 200 -> 101 OF choose SET grade = choose.grade +1

Fetch Data

Fetch the data updated.

nebula> FETCH PROP ON course 101;
================================
| course.name | course.credits |
================================
| Math        | 4              |
--------------------------------
nebula> FETCH PROP ON choose 200 -> 101
================
| choose.grade |
================
| 6            |
----------------

Delete Data

Delete vertex.

nebula> DELETE VERTEX $vid

Sample Queries

Q1. Find the vertexes that 201 follows:

nebula> GO FROM 201 OVER follow;
=============
| follow._dst |
=============
| 200       |
-------------
| 202       |
-------------

Q2. Find the vertexes that 201 follows, whose ages are greater than 17. Return their name, age and gender, and alias the columns as Friend, Age and Gender, respectively.

nebula> GO FROM 201 OVER follow WHERE $$.student.age >= 17 YIELD $$.student.name AS Friend, \
  $$.student.age AS Age, $$.student.gender AS Gender;
=========================
| Friend | Age | Gender |
=========================
|   Jane |  17 | female |
-------------------------

You can use backslash (\) as line break;

  • YIELD specifies what values or results you might want to return from query.
  • $^ represents the source vertex.
  • $$ indicates the target vertex.

Q3. Find the chosen courses and corresponding grades of students followed by 201.

-- By pipe
nebula> GO FROM 201 OVER follow yield follow._dst as id \
  | GO FROM $-.id OVER choose YIELD $^.student.name AS Student,\
  $$.course.name AS Course, choose.grade AS Grade;
=============================
| Student |  Course | Grade |
=============================
|  Monica |    Math |     6 |
-----------------------------
|  Monica | English |     3 |
-----------------------------
|    Jane | English |     3 |
-----------------------------

-- By temporary variable
nebula> $var=GO FROM 201 OVER follow yield follow._dst as id; \
  GO FROM $var.id OVER choose YIELD $^.student.name AS Student,\
  $$.course.name AS Course, choose.grade AS Grade;
=============================
| Student |  Course | Grade |
=============================
|  Monica |    Math |     6 |
-----------------------------
|  Monica | English |     3 |
-----------------------------
|    Jane | English |     3 |
-----------------------------
  • | denotes a pipe. The output of the formal query acts as input to the next one follow a pipeline.
  • $- refers to the input stream.

The second approach adopts a user-defined variable $var. The scope of this variable is within the compound statement.

Tips

In practice, it often takes several retries to find a proper graph schema during testing. You can put all the DDL statements in a file as follows.

DROP SPACE test; -- Delete useless space
CREATE SPACE test(partition_num=1024, replica_factor=1);
USE test;
CREATE TAG course(name string, credits int);
CREATE TAG building(name string);
CREATE TAG student(name string, age int, gender string);
CREATE EDGE like(likeness double);
CREATE EDGE select(grade int);

Then batch write to console

cat  schema.ngql | ./bin/nebula -u user -p password

Likewise, you can put hundreds or thousands DML statements in a data.ngql to insert. If you have millions of records to insert, please use the csv importer tool (or sst ingest tool).

You can’t perform that action at this time.