Changes Hadoop brought in ways of dealing with data
Hadoop has brought something new to the way in which data is dealt with, stored and processed. These are the three changes brought by Hadoop which are disruptive or absolutely against the classical approaches to handling and manage data.
- The code moves to the data rather than data moving to the code. Till the introduction of Hadoop, it was data that moved to the code but since the data has gone Huge it is much more efficient to move the code to the data rather than data to the code.
- Use of commodity hardware rather than custom and highly expensive hardware. To have a big, high computation server it may cost something around 50K and additionally investment has to be made on the software licenses as well. Hadoop being the open source it is cheaper; additionally even forming a cluster is far cheaper than buying one super expensive server. Studies have shown the Hadoop cluster can give 10 times the throughput at 1/10th of the cost in certain situations.
- The third and last game changer feature is read on schema; the schema can be decided on the time of read rather than at the time of storage. This is a big change as compared to fundamental, classical database storage we have seen till that point of time.
So these are the three big blows Hadoop has brought to the fundamental concepts of dealing with data before Hadoop came into the picture. And since these are more efficient and low cost as compared to the earlier practices, Hadoop has gained such great prominence in dealing with Big Data.