Apache Zeppelin + IRIS Quick Start
Apache Zeppelin it's a Multi-purpose notebook that allow you:
- Data Ingestion
- Data Discovery
- Data Analytics
- Data Visualization and Collaboration.
Apache Zeppelin interpreter concept allows any language/data-processing-backend to be plugged into Zeppelin. Currently Apache Zeppelin supports many interpreters such as Apache Spark, Apache Flink, Python, R, JDBC, Markdown and Shell.
It provides you a safe environment to get insigth of your data. You can take avantage of Zeppelin's higligths using both , JDBC and Spark IRIS native connectors.
This is a Quick Start Guide just to put you on the track.
1.- Donwload and save Zeppellin from the official web site --> (Here)
2.- Unpackage the distro in the installation folder (your election)
3.- You will need install a JDK (if you don't have it)
4.- When JDK is Installed you can run Zeppelin #bash> sudo ./<zeppelin_path>/bin/zeppelin-daemon.sh start (This is a Linux / Unix based install ... but there is a bash file for the Windows based systems.)
5.- Once it's running you can use it through the browser http://[server_ip]:8080
6.- If everything was OK you should see this screen
7.- Now it's necessary to setup the interpreters
8.- Let's config Spark Connector (for using the full potencial of IRIS + Zeppelin i.e. with Scala or R)
8.1 In the "Artifacts" section include routes to jdbc and IRIS spark connector
9.- Now let's configure the JDBC Interpreter , in defaul_url you must set jdbc:IRIS://<your_iris_server>:<port>/<namespace>, also your default user and default password, finally default.driver is com.intersystems.jdbc.IRISDriver
9.1 We continue configuring the JDBC IRIS dependencies
10.- It's Done! now we can use Zeppelin + IRIS
Let's use JDBC interpreter as a quick example
1.- Back in to the Zeppelin's landing page. and create a new notebook
2.- in the new Notebook create a new Paragraph, in the case of using the jdbc interpreter, the sentence must start as % jdbc
You will get your data as a table by default, but quickly you can toggle to a graph just changing and setting the view
You can combine different queries , views and layouts to easily get really powerfull dashboards
I really hope that information was useful and feel free to share it or improve it.
Register or login to poll