I've tested the materials mostly on Ubuntu, with python 2.7 and Oracle Java 1.8, and a prebuilt version of spark 1.3 (we'll be doing local mode, so most would work, I've been using the prebuilt for hadoop 2-4; available at http://d3kbcqa49mib13.cloudfront.net/spark-1.3.0-bin-hadoop2.4.tgz.
David Branner has set up a vagrant file, available in github at https://github.com/DataBranner/spark-pycon15_study Thanks David !!! If you like VMs and have a laptop powerful enough, this is amazing.
I'm preparing a script for install on ubuntu; you can download and run it it is at https://raw.githubusercontent.com/okaram/spark-pycon15/master/scripts/install_ubuntu.sh or just type
curl https://raw.githubusercontent.com/okaram/spark-pycon15/master/scripts/install_ubuntu.sh | bash
if you like to live dangerously :)