OrientDB with Apache TinkerPop 3

(since v 3.0)

OrientDB adheres to the Apache TinkerPop standard and implements TinkerPop Stack interfaces.

In versions pior to 3.0, OrientDB uses the TinkerPop 2.x implementation as the default for Java Graph API.

Starting from version 3.0, OrientDB ships its own APIs for handling Graphs (Multi-Model API). Those APIs are used to implement the TinkerPop 3 interfaces.

The OrientDB TinkerPop development happens here


Since TinkePop stack has been removed as a dependency from the OrientDB Community Edition, starting with version 3.0 it will be available for download as an Apache TinkerPop 3 enabled edition of OrientDB based on the Community Edition.

It contains all the features of the OrientDB Community Edition plus the integration with the Tinkerpop stack:

Source Code Installation

In addition to the binary packages, you can compile the OrientDB TinkerPop enabled distribution from the source code. This process requires that you install Git and Apache Maven on your system.

As a pre-requirement, you have to build the same branch for OrientDB core distribution:

$ git clone https://github.com/orientechnologies/orientdb
$ cd orientdb
$ git checkout develop
$ mvn clean install -DskipTests
$ cd ..

Then you can proceed and build orientdb-gremlin

$ git clone https://github.com/orientechnologies/orientdb-gremlin
$ cd orientdb-gremlin
$ git checkout develop
$ mvn clean install

It is possible to skip tests:

$ mvn clean install -DskipTests

This project follows the branching system of OrientDB.

The build process installs all jars in the local maven repository and creates archives under the distribution module inside the target directory. At the time of writing, building from branch develop gave:

$ls -l distribution/target/
total 266640
drwxr-xr-x   2 staff  staff    68B Mar 21 13:07 archive-tmp/
drwxr-xr-x   3 staff  staff   102B Mar 21 13:07 dependency-maven-plugin-markers/
drwxr-xr-x  12 staff  staff   408B Mar 21 13:07 orientdb-community-3.0.0-SNAPSHOT/
drwxr-xr-x   3 staff  staff   102B Mar 21 13:07 orientdb-gremlin-community-3.0.0-SNAPSHOT.dir/
-rw-r--r--   1 staff  staff    65M Mar 21 13:08 orientdb-gremlin-community-3.0.0-SNAPSHOT.tar.gz
-rw-r--r--   1 staff  staff    65M Mar 21 13:08 orientdb-gremlin-community-3.0.0-SNAPSHOT.zip

The directory orientdb-gremlin-community-3.0.0-SNAPSHOT.dir contains the OrientDB Gremlin distribution uncompressed.

Gremlin Console

The Gremlin Console is an interactive terminal or REPL that can be used to traverse graphs and interact with the data that they contain. For more information about it see the documentation here

To start the Gremlin console run the gremlin.sh (or gremlin.bat on Windows OS) script located in the bin directory of the OrientDB Gremlin Distribution

$ ./gremlin.sh 

         (o o)
plugin activated: tinkerpop.orientdb

Alternatively if you downloaded the Gremlin Console from the Apache Tinkerpop Site, just install the OrientDB Gremlin Plugin with

gremlin> :install com.orientechnologies orientdb-gremlin {version}

and then activate it

gremlin> :plugin use tinkerpop.orientdb

Open the graph database

Before playing with Gremlin you need a valid OrientGraph instance that points to an OrientDB database. To know all the database types look at Storage types.

When you're working with a local or an in-memory database, if the database does not exist it's created for you automatically. Using the remote connection you need to create the database on the target server before using it. This is due to security restrictions.

Once created the OrientGraph instance with a proper URL is necessary to assign it to a variable. Gremlin is written in Groovy, so it supports all the Groovy syntax, and both can be mixed to create very powerful scripts!

Example with a memory database (see below for more information about it):

gremlin> graph = OrientGraph.open();

Some useful links:

Working with in-memory database

In this mode the database is volatile and all the changes will be not persistent. Use this in a clustered configuration (the database life is assured by the cluster itself) or just for test.

gremlin> graph = OrientGraph.open()

Working with local database

This is the most often used mode. The console opens and locks the database for exclusive use. This doesn't require starting an OrientDB server.

gremlin> graph = OrientGraph.open("embedded:/tmp/gremlin/demo");

Working with a remote database

To open a database on a remote server be sure the server is up and running first. To start the server just launch server.sh (or server.bat on Windows OS) script. For more information look at OrientDB Server

gremlin> graph = OrientGraph.open("remote:localhost/demodb");

Use security

OrientDB supports security by creating multiple users and roles associated with certain privileges. To know more look at Security. To open the graph database with a different user than the default, pass the user and password as additional parameters:

gremlin> graph = OrientGraph.open("remote:localhost/demodb","reader","reader");

With Configuration

gremlin> config = new BaseConfiguration()
gremlin> config.setProperty("orient-url","remote:localhost/demodb")
gremlin> graph = OrientGraph.open(config)

Available configurations are :

  • orient-url: Connection URL.
  • orient-user: Database User.
  • orient-pass: Database Password.
  • orient-transactional: Transactional. Configuration. If true the OrientGraph instance will support transactions (Disabled by default).

Mutating the Graph

Create a new Vertex

To create a new vertex, use the addVertex() method. The vertex will be created and a unique id will be displayed as the return value.


Create an edge

To create a new edge between two vertices, use the v1.addEdge(label, v2) method. The edge will be created with the label specified.

In the example below two vertices are created and assigned to a variable (Gremlin is based on Groovy), then an edge is created between them.

gremlin>  v1 = graph.addVertex();

gremlin>  v2 = graph.addVertex();

gremlin> e = v1.addEdge('friend',v2)

Close the database

To close a graph use the close() method:

gremlin> graph.close()

This is not strictly necessary because OrientDB always closes the database when the Gremlin Console quits.


OrientGraph supports Transactions. By default Tx are disabled. Use configuration to enable it.

gremlin> config = new BaseConfiguration()
gremlin> config.setProperty("orient-url","remote:localhost/demodb")
gremlin> config.setProperty("orient-transactional",true)
gremlin> graph = OrientGraph.open(config)

When using transactions OrientDB assigns a temporary identifier to each vertex and edge that is created. use graph.tx().commit() to save them.

Transaction Example

gremlin> config = new BaseConfiguration()
gremlin> config.setProperty("orient-url","remote:localhost/demodb")
gremlin> config.setProperty("orient-transactional",true)
gremlin> graph = OrientGraph.open(config)
gremlin> graph.tx().isOpen()
gremlin> v1 = graph.addVertex()
gremlin> v2 = graph.addVertex()
gremlin> e = v1.addEdge("friends",v2)
gremlin> graph.tx().commit()


The power of Gremlin is in traversal. Once you have a graph loaded in your database you can traverse it in many different ways. For more info about traversal with Gremlin see here

The entry point for traversal is a TraversalSource that can be easily obtained by an opened OrientGraph instance with:

gremlin> graph = OrientGraph.open()
gremlin> g = graph.traversal()
==>graphtraversalsource[orientgraph[memory:orientdb-0.41138060448051794], standard]

Retrieve a vertex

To retrieve a vertex by its ID, use the V(id) method passing the RecordId as an argument (with or without the prefix '#'). This example retrieves the first vertex created in the above example.

gremlin> g.V('#33:0')

Get all the vertices

To retrieve all the vertices in the opened graph use .V() (V in upper-case):

gremlin> g.V()

Retrieve an edge

Retrieving an edge is very similar to retrieving a vertex. Use the E(id) method passing the RecordId as an argument (with or without the prefix '#'). This example retrieves the first edge created in the previous example.

gremlin> g.E('145:0')

Get all the edges

To retrieve all the edges in the opened graph use .E() (E in upper-case):

gremlin> g.E()

Basic Traversal

To display all the outgoing edges of a vertex use .outE(<label>) . Label is optional and if passed the outgoing edges will be filtered by that label


gremlin> g.V('#41:1').outE()

To display all the incoming edges of a vertex use .inE(<label>). Example:

gremlin> g.V('#41:1').inE()

For more information look at the Gremlin Traversal.

Filter results

This example returns all the outgoing edges of all the vertices with label equal to 'friend'.

gremlin> g.V('#41:1').inE('HasProfile')

Create complex paths

Gremlin allows you to concatenate expressions to create more complex traversals in a single line:


Of course this could be much more complex. Below is an example with the graph taken from the official documentation:

gremlin> g.V('44:0').out('HasFriend').out('HasFriend').values('Name').dedup()

Gremlin Server

There are two ways to use OrientDB inside the Gremlin Server

  • Use the OrientDB-TP3 distribution that embedd the Gremlin Server
  • Install the OrientDB-Gremlin Driver into a GremlinServer


Dowload the latest version of OrientDB-TP3 here. and start OrientDB to automatically start the embedded Gremlin Server. The configuration of the Gremlin Server is in $ORIENTDB_HOME/config.

When using the embedded version by default the Authentication manager authenticate the users against OrientDB server user with permission gremlin.server.

Install OrientDB-Gremlin

Download the latest Gremlin Server distribution here

and then install the OrientDB Gremlin driver with

bin/gremlin-server.sh -i com.orientechnologies orientdb-gremlin ${version}

OrientDB Gremlin Server configuration

YAML configuration example (save as gremlin-server.yaml)

host: localhost
port: 8182
scriptEvaluationTimeout: 30000
channelizer: org.apache.tinkerpop.gremlin.server.channel.WebSocketChannelizer
graphs: {
  graph : conf/orientdb-empty.properties
scriptEngines: {
  gremlin-groovy: {
    plugins: { org.apache.tinkerpop.gremlin.server.jsr223.GremlinServerGremlinPlugin: {},
               org.apache.tinkerpop.gremlin.orientdb.jsr223.OrientDBGremlinPlugin: {},
               org.apache.tinkerpop.gremlin.jsr223.ImportGremlinPlugin: {classImports: [java.lang.Math], methodImports: [java.lang.Math#*]},
               org.apache.tinkerpop.gremlin.jsr223.ScriptFileGremlinPlugin: {files: [ scripts/empty-sample.groovy]}}}}
  - { className: org.apache.tinkerpop.gremlin.driver.ser.GryoMessageSerializerV3d0, config: { ioRegistries: [org.apache.tinkerpop.gremlin.orientdb.io.OrientIoRegistry] }}             # application/vnd.gremlin-v3.0+gryo
  - { className: org.apache.tinkerpop.gremlin.driver.ser.GryoMessageSerializerV3d0, config: { serializeResultToString: true }}                                                                       # application/vnd.gremlin-v3.0+gryo-stringd
  - { className: org.apache.tinkerpop.gremlin.driver.ser.GraphSONMessageSerializerV3d0, config: { ioRegistries: [org.apache.tinkerpop.gremlin.orientdb.io.OrientIoRegistry] }}         # application/json
  - { className: org.apache.tinkerpop.gremlin.server.op.session.SessionOpProcessor, config: { sessionTimeout: 28800000 }}
  - { className: org.apache.tinkerpop.gremlin.server.op.traversal.TraversalOpProcessor, config: { cacheExpirationTime: 600000, cacheMaxSize: 1000 }}
metrics: {
  consoleReporter: {enabled: true, interval: 180000},
  csvReporter: {enabled: true, interval: 180000, fileName: /tmp/gremlin-server-metrics.csv},
  jmxReporter: {enabled: true},
  slf4jReporter: {enabled: true, interval: 180000}}
strictTransactionManagement: false
maxInitialLineLength: 4096
maxHeaderSize: 8192
maxChunkSize: 8192
maxContentLength: 65536
maxAccumulationBufferComponents: 1024
resultIterationBatchSize: 64
writeBufferLowWaterMark: 32768
writeBufferHighWaterMark: 65536
ssl: {
  enabled: false}

Graph Configuration properties example (e.g. save as orientdb-empty.properties. It has to be the one referenced in the gremlin-server.yaml file)


OrientDB TinkerPop Graph API

In order to use the OrientDB TinkerPop Graph API implementation, you need to create an instance of the OrientGraph class. If the database already exists, the Graph API opens it.

NOTE: When creating a database through the Graph API, you can only create PLocal and Memory databases. Remote databases must already exist.

NOTE: In v. 2.2 and following releases, when using PLocal or Memory,please set MaxDirectMemorySize (JVM setting) to a high value, like 512g -XX:MaxDirectMemorySize=512g

When building multi-threaded application, use one instance of OrientGraph per thread. Bear in mind that all graph components, such as vertices and edges, are not thread safe. So, sharing them between threads may result in unpredictable results.

Remember to always close the graph instance when you are done with it, using the .close() method.

Working with in-memory database

OrientGraph graph = OrientGraph.open();
try {
} finally {

Working with local database

OrientGraph graph = OrientGraph.open("embedded:/tmp/gremlin/demo");
try {
} finally {

Working with remote database

OrientGraph graph = OrientGraph.open("remote:localhost/demodb");
try {
} finally {

Using the Graph Factory

The OrientGraphFactory act as a factory for OrientGraph instances Using a Graph Factory in your application is relatively straightforward. First, create and configure a factory instance using the OrientGraphFactory class. Then, use the factory whenever you need to create an OrientGraph instance and shut it down to return it to the pool. When you're done with the factory, close it to release all instances and free up system resources.

OrientGraphFactory factory = new OrientGraphFactory("remote:localhost/demodb","admin","admin");

Then you can:

  • retrieve a transactional instance, use the getTx() method on your factory object:
OrientGraph txGraph = factory.getTx();
  • retrieve a non-transactional instance, use the getNoTx() method on your factory object:
OrientGraph noTxGraph = factory.getNoTx();

Shutting Down Graph Instances

When you're done with a Graph Database instance, you can return it to the pool by calling the close() method on the instance. This method does not close the instance. The instance remains open and available for the next requester:


Releasing the Graph Factory

When you're ready to release all instances, call the close() method on the factory. In the case of pool usage, this also frees up the system resources claimed by the pool:


results matching ""

    No results matching ""