The workshop series offers a brief introduction to concepts of parallel distributed computing and the Hadoop universe. Participants will learn to navigate among the various tools, and to write programs for large scale data analysis. Examples will be provided in Python; knowledge of the Java programming language is not required.

We will cover some of the core elements of Hadoop: the distributed file system (HDFS), MapReduce, Yarn, Hive, Kafka, and some other components.

1. Participants learn about the core concepts behind Hadoop, how to manage data files on HDFS, and use the MapReduce Streaming API for scalable, distributed data processing with Python.
2. The second session covers managing data in Hive and HBase, and best practices of using schema based versus key-value-based data stores.
3. Participants will use Kafka to ingest data from Internet sources, such as the Twitter Streaming API, and prepare the collected samples for processing.

Sessions will be accompanied with exercises on varying levels. All software components and the data sets are available on the Analytics-Research-Cluster http://arc.insight.gsu.edu.

• Participants should bring their own laptop.
• Windows users should install a SSH client (for terminal connection) like http://smartty.sysprogs.com/.
• Mac and Linux users already have this capability. In addition, a recent web-browser is required.

These steps are only needed if you want to run Hadoop on your own computer:

1. Install either VMWare Virtual Machine or Virtual Box on your laptop. (VMWare offers a free Workstation Player for Windows and Linux https://my.vmware.com/en/web/vmware/free#desktop_end_user_computing/vmware_workstation_player/12_0, OS X users may use VMWare Fusion https://www.vmware.com/products/fusion/fusion-evaluation.html. Oracle's VirtualBox https://www.virtualbox.org/wiki/Downloads is free for all platforms.)
2. Download the Hortonworks Sandbox for the respective virtual machine from http://hortonworks.com/products/hortonworks-sandbox/#install (it's over 9 GB, it may take a while).

1. MapReduce Design Patterns
2. Hadoop MapReduce v2 Cookbook, 2nd Edition