Set Up and Preparation for sds-2.x


Here you are given instructions to setup and prepare for the workshop/course.

Here you have two main tasks:

  • TASK 1 is to set up a free databricks community edition account.
  • TASK 2 is to set up docker in your local system (laptop or VM).

1. Cloud-full Computing Environment Setup

TASK 1. Create an account at databricks

databricks, the commercial arm of the creators of Apache Spark, provides a free and managed Spark micro-cluster in the cloud (currently AWS) for learning purposes to anyone! They also intended to let these accounts remain open “indefinitely” (or at least in the forseeable future). This is our most convenient way to learn Spark.

Steps to complete TASK 1

  1. Go to Databricks Community Edition at the following URL:
  2. Signup for the databricks community edition account
    • use your private email address as your account name, so you can continue working on learning Spark
    • fill in your employer details, job description, etc.
    • say you are trying to learn Spark as the reason for requesting the community edition account
    • after submitting the form, follow instructions including checking your email address used for the account (check junk mailbox, in case it is not in your inbox)
    • Note that we have explicit permission from databricks to use the community edition for your learning (20190410, Dublin, Ireland.).

2. Cloud-free Computing Environment Setup

TASK 2. Set up Docker for your laptop

We have several Steps to complete TASK 2 as detailed below.

Step 0 for TASK 2: Learn basics of docker


Familiarise yourself with concepts in at least the first three parts of Getting Started with Docker:

Approximate reading times for the above four links is 10, 4, 14 and 9 minutes respectively (less than 1 hour).

Step 1 for TASK 2: Install docker

Follow the steps to install docker for your laptop (all three OSs are supported) from the following URL:


  1. Check if your laptop has virtual memory enabled.
  2. You may need to increase the permitted size of the disk image to around 112GB if possible for the full docker compositions that are possible

NOTES for Windows 7 OS

Step 2 for TASK 2: Install docker

You next need to run docker-compose from an open network environment as instructed in the following URLs in more details:

But briefly, you need to minimally do the following (see More Details section below).

After downloading as explained below, unzip it into say a directory called sds inside your home directory like so: /.../home/user/sds/, minimally run these commands in an open network envitonment. See file below to test Spark-shell, etc.

$ pwd
$ dockerCompose/
$ docker-compose -f docker-compose-hadoop.yml up -d
$ docker-compose down

Note that you will generally use stop and start as well as attach in docker-compose once everything is set-up properly so you don’t reformat the hdfs with injected data (this will happen if you docker-compose down). We will see these commands in the sequel.

More details at the raw github desktop-view level