Spark on Zeppelin

1 . Zeppelin Zeppelin Spark We are hiring P7 jeffzhang.zjf@alibaba-inc.com 

2 .Spark on Zeppelin Jeff Zhang 

3 .Who is Jeff Zhang 2008 2011 2013 2014 2018 

4 .Pain points of Big Data ● Hard to Test ● Hard to move development to production ● Hard to integrate with downstream application ● Hard to integrate with other tools 

5 .What is Apache Zeppelin Web-based notebook that enables data-driven, interactive data analytics and collaborative documents with SQL, Scala, Python and more. 

6 .Usage Scenarios ● Ad-hoc analytics ● ETL ● Build dashboard ● Machine Learning 

7 .Zeppelin Architecture Spark Interpreter Zeppelin Server WebSocket RPC Interpreter Client Launcher Jdbc Interpreter Rest Notebook Interpreter Manager Manager Shell Interpreter 

8 .Multiple language support SparkContext / SparkSession 

9 .Multiple spark version support Spark 1 Spark 2 Zeppelin Server Spark 3 

10 .Multiple mode support ● Local ● Standalone ● Yarn ● K8s 

11 .Inline Visualization 

12 .Dynamic forms TextBox Select Checkbox 

13 .Other features ● Inline Configuration ● Hive Integration ● Multiple user support Impersonation ● ZeppelinContext 

14 .Demo ● Configure & Start Zeppelin ● Configure Spark Interpreter ● Run Spark Tutorial ● Rest API 

15 .Pain points of Big Data ● Hard to Test ● Hard to move development to production ● Hard to integrate with downstream application ● Hard to integrate with other tools 

16 . Zeppelin Zeppelin Spark We are hiring P7 jeffzhang.zjf@alibaba-inc.com