Showing posts with label hadoop. Show all posts
Showing posts with label hadoop. Show all posts

Wednesday, July 24, 2013

OSCON Getting Hadoop, Hive and HBase up and running

The Getting Hadoop, Hive and HBase up and running in fifteen minutes page on OSCON.

Mark Grover (Cloudera)
4:10pm Wednesday, 07/24/2013

Hadoop is a distributed batch processing system
Installation and configuring hadoop projects is hard
Integration testing not done across versions.

Apache Bigtop addresses this issue
- generates packages of various Dadoop ecosystem componenets for various distros
- provides deployment code for various projects
- convenience artifacts available e.g hadoop-conf-pseudo to fake multiple nodes on local workstation
- integration testing of latest project releases

Follow steps on wiki

github.com/markgrover/oscon-bigtop/blob/master/readme.md

Hive
- allows SQL syntax query into hadoop
- don't claim to be SQL complaint but are very close. do not support correlated sub-queries.

Code for demo:
github.com/markgrover/oscon-bigtop