Query Engines for Hive, MR, Spark, Tez and LLAP – Considerations!

Big Data Meetup in Canary Wharf, London!

Agenda:

18:00h - Networking & refreshments

18:30h - Kick Off & Intros

18:45h - Query Engines for Hive - MR, Spark, Tez and LLAP – Considerations!

Apache Hive has established itself to be the Data Warehouse of choice on Hadoop for storing data on HDFS and compatible file systems. Hive has been using the standard MapReduce as its execution engine for a while. Newer release of Hive allow Apache Spark or Apache Tez together with LLAP to be used as its query engines as well. With the ability of Hive to deploy these different performant execution engines, it is important to understand the benefits and limitations that each engine brings in coming to an informed decision in deploying them. Mich will discuss how to set up Spark as execution engine for Hive and will present some interesting results. In addition, the presentation has now been extended to cover the deployment of Tez together with LLAP as a hybrid execution engine for Hive with equally interesting findings.

About the Presenter

Mich Talebzadeh is an award winning technologist and architect who holds a PhD in Particle Physics from Imperial College of Science and Technology, University of London. He specializes in the strategic use of Big Data ecosystem, RDBMS, IMDB and CEP. Mich is the author of two books and the author of the forthcoming book “Complex Event Processing in Heterogeneous Environments” plus numerous articles on Big Data. Mich is also an active contributor to Big Data User Groups notably Apache Spark and Apache Hive.

19:45h - Networking and refreshments