There are several editions of OrientDB Server, each of them meets some criteria, and you have to choose which one you need. Keep in mind, however, that the data files are compatible through the server versions, so you can switch from one version to another anytime.
The three available versions are:
OrientDB Standard Edition
OrientDB GraphEd Edition
OrientDB Enterprise Edition
There was another version which was focused on the key/value pair's management. It was called OrientKV Server but now it has deprecated and is no longer supported. In fact, the other versions of OrientDB have the capability to manage the key/value indexes as well, so if you need to manage large associative arrays or very big hash tables, you can use them to do that.
The Standard Edition is the "main" edition of OrientDB; the other ones are extensions of the Standard Edition and add some specific capabilities to it.
The Standard Edition is shipped with a rich set of out of the box features; all of them are immediately available after the server installation.
Graph database support
Document database support
Object database support
Dictionary indexes support
A SQL-like query language
Transaction support
Hooks (a sort of trigger, but at the system level rather than class level)
Custom SQL function (to write your own functions in Java to expand the SQL parser)
Stored procedures
Plugin support to add features to the server's core
REST interface
Clustering
Command-line administration console
Web-based administration console (OrientDB Studio)
The Graphed Edition adds the TinkerPop Blueprints interface and the Gremlin query language to the Standard Edition.
Note that the Graphed Edition does not add the graph capabilities to OrientDB.
Relations through edges and other graph features are first-class capabilities in OrientDB, and are present in the Standard Edition too.
The Graphed Edition provides a layer on top of the OrientDB databases, so you can manage them through the Blueprints interface and provide the support to use the Gremlin language, if you want to use this de-facto query language for the graph databases instead of the OrientDB SQL-like language.
The OrientDB Key Value server has not been supported since April 2011, and its source code was removed from the public repository in January 2013.
Its functionalities are available in the other versions.
In fact, each new OrientDB database has a special structure called Dictionary. The scope of the dictionary is to provide key/value management features to implement, for example, lookup tables, caches, and logs. You can also define as many dictionaries as you like, as per you needs.
To install OrientDB you have to download its latest stable release, and you must have a Java virtual machine (Java SE 6 or above) installed on your target system.
Since the JVM is the only requirement to run an OrientDB server, this means it can be installed on any system supported by the Java platform, even the Raspberry Pi!
You can find the binary packages on the OrientDB official site: www.orientdb.org.
Currently, the 1.3.0 version is available and you can find two packages under the download section:
Download the orientdb-1.3.0.tar.gz file from the OrientDB site, and extract its content to a directory on your system.
Now you should have a directory tree similar to the structure shown as follows:
/orientdb-1.3.0 /benchmarks /bin /config /databases /lib /log /www history.txt license.txt readme.txt
The following is an explanation for the preceding directory tree structure:
benchmarks: scripts to perform several benchmark testsbin: scripts to run and to stop the server and the command-line consoleconfig: XML configuration filesdatabase: default path for the database fileslib: the JAR fileslog: default path for the server logfileswww: the OrientDB Studio web applicationhistory.txt: the change logfilelicense.txt: the Apache 2 licensereadme.txt: instruction to build OrientDB from the source code
To run the server, just go into the bin directory, and launch the server.bat (on Windows OS) or server.sh (on Unix/Linux systems).
If you are using a Unix-based system, you may have to set the execution permission to the script files:
chmod +x ./bin/*.sh
If you plan to run OrientDB outside the bin directory, you have to set the bin path in the PATH environment variable, furthermore you have to set the ORIENTDB_HOME environment variable to the path's directory in which you extracted the tar.gz package.
The ORIENTDB_HOME variable is used by the scripts to guess the OrientDB position.
Now you can launch the server.
In the bin directory there are several scripts, for both Windows and Linux:
aserver.*: just ignore them, they are experiments for the next releasesconsole.*: run the command-line consoledserver.*: run the server in distributed modeorientdb.sh: script to run/stop OrientDB as a daemon on Unix-like systemsserver.*: run the servershutdown.*: shutdown the server in a clean way (that is, do not kill it)
You should have an output similar to this:
2013-04-19 09:20:21:600 INFO OrientDB Server v1.3.0 (build @[email protected]) is starting up... [OServer] 2013-04-19 09:20:22:936 INFO -> Loaded memory database 'temp' [OServer] 2013-04-19 09:20:23:148 INFO Listening binary connections on 0.0.0.0:2424 [OServerNetworkListener] 2013-04-19 09:20:23:151 INFO Listening http connections on 0.0.0.0:2480 [OServerNetworkListener] 2013-04-19 09:20:23:181 INFO Installing GREMLIN language v.2.2.0-SNAPSHOT [OGraphServerHandler] 2013-04-19 09:20:23:195 INFO OrientDB Server v1.3.0 is active. [OServer]
Now you can run the console to connect to the server and try some simple commands.
In another terminal window, go to the bin directory and launch the console script.
Type ? and see all the available commands:
The first command you have to supply is the connect command, to connect to a server:
connect remote:127.0.0.1/demo admin admin
The server replies:
Connecting to database [remote:127.0.0.1/demo] with user 'admin' ...OK orientdb>
You can try the simple command:
info
To exit from the console, type:
exit
Now to shutdown the server, go to the bin directory and launch the shutdown script.
The procedure is similar to that used for the Standard Edition.
Go to the download page www.orientdb.org and grab the orientdb-graphed-1.3.0.tar.gz file.
Once you have downloaded the file, unzip it in a convenient directory.
The tree structure is same as that of the Standard Edition. There are some differences in some of them:
bin: there are thegremlin.*scripts to run the Gremlin scripts from command line, batches, and shellsdatabases: there is theTinkerpopdemo database instead of the OrientDB onelib: in addition to the OrientDB libraries, there are also the libraries of the Tinkerpop Stack, and other necessary JARs
Again to run the server, just go into the bin directory, and launch the server.bat (on Windows OS) or server.sh (on Unix-like systems).
Remember to set the right execution permission to the script files:
chmod +x ./bin/*.sh
and to set the ORIENTDB_HOME environment variable.
Tip
Check if another instance of OrientDB is running. If so, terminate it to avoid the port binding conflicts.
Once the server is started, take a look at the last two output rows. Here you can see if the Gremlin language support is loaded:
2013-04-19 09:20:23:181 INFO Installing GREMLIN language v.2.2.0-SNAPSHOT [OGraphServerHandler] 2013-04-19 09:20:23:195 INFO OrientDB Server v1.3.0 is active. [OServer]
Now you can run the console tool from another terminal window. Also, in this case, you should see a notice indicating that the Gremlin language support is available.
In other terminal, go to the bin directory and launch the console command:
OrientDB console v.1.3.0 (build 16) www.orientechnologies.com
Type help to display all the commands supported.
Installing extensions for GREMLIN language v.2.2.0-SNAPSHOT orientdb>
OrientDB Graphed Edition is shipped with the Tinkerpop demo database.
Try to connect to the db:
connect remote:127.0.0.1/tinkerpop admin admin
And then try to execute the info command:
info
To exit from the console type:
exit
The source code is available through the GitHub platform at the URL http://github.com/nuvolabase/orientdb.
Since the build process will put the generated file in a directory called release, which will be a sibling of the source directory, I suggest you create a directory just for OrientDB. Inside that directory create a new directory to place the downloaded source code.
You can download the latest source code by selecting the master branch and downloading the master.zip file using the ZIP button.
Otherwise, you can clone the GIT repository on your machine using git. Type:
git clone git://github.com/nuvolabase/orientdb.git
To build OrientDB from source code you must have a JDK SE 6 or above, and the Ant Tool version 1.6.5 or above.
You can download Ant from http://ant.apache.org/.
Please note that the ant executable directory must be included in the PATH environment variable.
Once you have downloaded the source code and installed the necessary tool, you can proceed to build OrientDB.
Go into the source code directory and type:
ant clean install
You can also launch the build.bat (or build.sh in Unix-like system), however these scripts call ant.
To build the Graphed Edition too, type:
ant clean installg
At the end of the build process you should have a screen shown as follows:
The generated files are in the release\orientdb-x.y.z-SNAPSHOT and release\orientdb-graphed-x.y.z-SNAPSHOT directories, where x.y.z is the current version under development.
To run the test suite against a freshly-built OrientDB snapshot, you can type the following command:
ant test
After the test suite has finished (you should have 0 errors), in the database directory of the Standard Edition, you should have some databases created to run the tests, including the demo database.
There are another set of tests that need of Maven to be executed.
So, if you want to execute these tests you must download Maven from http://maven.apache.org/download.cgi
and install it following the instruction provided in the readme.txt file within the installation package.
Once you have Maven on your system, you can type:
mvn clean test
OrientDB can run as a background process. The setup process depends on the server platform.
OrientDB is shipped with a script that can be used to run OrientDB like a daemon.
It supports the start/stop/status parameters and can be configured to execute OrientDB with a specified user's credentials.
The script is orientdb.sh and is located in the /bin directory of the installation path.
You must open it and change the first lines:
ORIENTDB_DIR="YOUR_ORIENTDB_INSTALLATION_PATH" ORIENTDB_USER="USER_YOU_WANT_ORIENTDB_RUN_WITH"
Set the installation path and the user as stated, save the script, and deploy like other scripts for the other daemon.
Different Linux distribution uses different ways to manage the start/stop process at the system bootstrap/shutdown.
Since OrientDB is a Java application, it does not offer any native way to run as a Windows service.
However, there are some tools that can be wrapped with any executable so that it can be installed and managed as a Windows service.
The correct procedure to do this is illustrated at the wiki page https://github.com/nuvolabase/orientdb/wiki/Wrapping-As-A-Windows-Service. The steps are as follows:
You must download the Apache Commons Daemon tools for your system. The latest version is available at: http://www.apache.org/dist/commons/daemon/binaries/windows/.
Download and unzip it.
You will find a
prunsrv.exefile in the root directory. This is the file for x86/32 bit systems. In the directoryamd64there is a version of theprunsrv.exefile for x86/64 architectures, while the directoryia64contains a version of theprunsrv.exefile for the Itanium machines.Let's say you have installed OrientDB in the
%ORIENTDB_HOME%directory, you have to create a new directory called%ORIENTDB_HOME%/service.Copy to this new directory the two
.exefiles shipped with the Apache Commons Daemon tools:prunmgr.exeandprunsrv.exe, according to the architecture of your machine.Rename
prunsrv.exetoOrientDB.exe, andprunmgr.exetoOrientDBw.exe.Copy to the same directory the file
installService.bat.In order to execute this script you must locate the
jvm.dllfile installed on your system. Generally this file is in the%JAVA_HOME%\jre\bin\serverdirectory.Assuming that Java is installed in the
C:\Program Files\Java\jdk1.6.0_37path, and that OrientDB is installed in theC:\OrientDB\releases\orientdb-1.4.0-SNAPSHOTdirectory, you must type the following command:installService.bat "C:\Program Files\Java\jdk1.6.0_37\jre\bin\server\jvm.dll" C:\OrientDB\releases\orientdb-1.4.0-SNAPSHOTNow you can open the Windows Services Management Console and see OrientDB listed as a service.
In this chapter, we had our first encounter with OrientDB. We have seen the available versions, have learned to build it from the latest available source code, and have run a server instance.
We have also seen how to deploy it as a Windows service.
In the next chapter, we will go into more detail about the OrientDB architecture. We will explore the first basic concepts, such as how the data are organized and stored on filesystem, the security framework, and how to perform administrative tasks using the embedded tools.