Our performance-based exams (Data Engineer and Data Scientist) require two levels of skills:
- Theoretical skills (skills to know what the right approach is)
- Practical skills (skills to implement the solution)
Think of it as a multiplce-choice test wrapped inside a hands-on test. We don't tell you "ingest this with Sqoop, partition the data like this, and write it to this format with this compression" -- we give you a scenario, inside of which is wrapped a number of theoretical decisions you must recognize -- but these are quite routine scenarios. There's nothing exotic or tricky, just real.
We've noticed some people really struggle with translating "book learning" into practice and just as many translating "practice into implementation."
As such, we really recommend the training and are not offering practice sessions divorced from a full experience and context with an instructor. As stated before, the best preparation is Developer Training for Spark & Hadoop I (http://university.cloudera.com/instructor-led-training/developer-training-for-spark-and-hadoop-i) of which there is the first public beta August 18 (50% off). That gives you the VM and lots of hands-on tutorials, and an instructor to help guide you.
We are considering doing a series of free monthly tech talks where one of our exam developers walks through a scenario and how to approach it, but this takes time and we're still focused on the beta. We want to see where people struggle and where they don't before spending lots of time developing guides.
If you don't want to take the time or spend the money to learn these skills in an environment designed to teach you these skills, there are many options if you search the internet (EdX, Coursera, etc.). So many great free tutorials and blogs and classes available.
We've collected many of them
After that, all you really need is the apache docs which are also made available to you during the exam. Everything in the exam is answered in the docs.
