There are many methods used to deploy Hadoop, few are as follows :
In this method, there are no daemons running, everythings runs as in a single JVM. This method is suitable for running MapReduce program during development and it is easy test, debug.
The Hadoop daemon process runs on a local machine simulating a cluster on a small scale.
Fully distributed :
Here, Hadoop runs on a cluster of machines which provides production environment.