Great question and one of the first question that comes to anyone who heard about Hadoop.
Yes Hadoop is built out of Java Classes and Objects, but do you need to Java before you get a job in Hadoop and Big data
Technologies – Yes and No! The real answer is it DEPENDS.
It totally depends on what you want to do in the big data space and in Hadoop and the role you are going to play. Do you want to be hard core map reduce developer or
You want to learn few of the many awesome tools that are built around the Hadoop Eco System which writes the map reduce jobs for you?
There is DEMAND for both, so start learning Hadoop today.
Three years before when Hadoop Was purely a batch oriented platform knowing Java was critical since Hadoop developers
have to write Java CODE to write MAP-REDUCE programs which is the very FUNDAMETAL thing Hadoop does.
But with so much effort from open source communities and big enterprises it has gone far from manually writing Java MAP REDUCE programs.
Think about it for last 3 or 4 decades all of the IT was dominated by SQL and companies like
Spent their time and money building SQL as the De Facto standard for getting data out of the database store. So totally abandoning the SQL and moving on to writing map reduce programs using Java, Python. Perl or C++ is like spending all your money you saved for years(since you didn’t like the way you saved your money) and stating to save again. Doesn’t make sense.
Here are some cool alternatives that have evolved in the last few years.
Hive – open source tool that you can use to run SQL on top of Hadoop HDFS table abstraction layer.
Pig – Using Lig Latin language to manipulate the data stored in HDFS.
These are not just the tools, but every major enterprises are building far better tools that the above tools and some of the examples are –
IBM – Bigsql(Performs better that Hive)
Teradata – SQL on Hadoop
Oracle – BIGDATASQL
SyncSORT – GUI tool like Informatica.
And many more.
So knowing basic Java would be more than sufficient in starting your career in big data and the same applies to Linux since Hadoop was primarily built for Linux.
So start your big data journey today!!!