Apache Kylin : Implementing OLAP on the Hadoop platform

A Comprehensive Course for Learning How to Build and Query Big Data OLAP Cubes Using Apache Kylin.

Apache Kylin is an Apache top-level project that bring OLAP to Big data. This simply means that we can now write complex aggregation queries with different levels of aggregation and expect to get a second or micro-seconds response to our query. 

Online analytical processing (OLAP) has been a common word in traditional business intelligence for years but has not been easy with hadoop platform that has become a data lake solution for many.  These data lake often have hundreds of millions and even billions of records that organizations want to slice and dice for insights. However, the high latency of query execution in SQL on Hadoop technologies like Apache Hive or Apache Drill often meant that data architect opted to transfer their data back to traditional systems that allow for real time response to query.

Kylin solves all of this. 

With Apache Kylin, anyone with the skills can now build OLAP, ROLAP or MOLAP structures using a web UI, deploy it and expect to query these structure with second of response time in mind. Also, one can connect their applications or favorite visualization tools to Kylin to integrate data either for system processing or for visualization. 

In this course, we are going to review 

• What Kylin is
• How it works
• How to build OLAP cubes in batch and streaming model
• How to deploy the cubes
• How to query cubes
• How to connect external tools and applications to Kylin

What is the target audience?

Big Data Engineers/Developers
Data Architects
Data Analysts.
Anyone who wishes to be able to write simple to complex aggregation queries of large dataset and wants a low latency response time.

What are the requirements?

You need access to a Big Data Sandbox like Cloudera quickstart VM, Hortonworks HDP sandbox or a cloud-based Hadoop environment with a least 10GB of Ram.
You should have some familiarity SQL and be able to use ODBC or JDBC based tools.
Some familiarity with Linux will be helpful

What do I need to know to get the best out of this course?

Because Kylin uses other hadoop projects to achieve its design a fair understanding of projects like Apache Hive, Apache Kafka, Apache HBase, MapReduce is great for this course. However, one can still use Kylin without any knowledge of these technologies. 

It is also worth knowing that no prior knowledge of any big data technology is required to query Kylin or use data integration in running report or data visualizations.