What are the different modes in which we can configure/install Hadoop?

We can install the Hadoop in following three modes:

(1) Standalone / Local Mode: We can install Hadoop on a single node in standalone (or local) mode. In this mode, Hadoop runs on a single node in single Java Process. In this mode, there is no need to start any daemon. This mode could be used for development purpose.

(2) Pseudo Distributed Mode: Here we simulate a Hadoop cluster on the single node. Hadoop can be installed on a single node but various daemons (master as well as the slave daemons) of Hadoop will run on the same machine as separate Java processes. This mode is usually used for development, testing, research, learning, etc.

(3) Distributed Mode: Here Hadoop is deployed on the real multi-node cluster. Master daemons (NameNode, ResourceManager) are deployed on the master node, while slave daemons (DataNode, NodeManager) are deployed on slave nodes. This mode is used in the production, here we get the actual power of Hadoop

Hadoop can be configured in 3 different modes and those are as follows:

1. Local(Standalone) Mode
This is a default mode of Hadoop configuration to run on a single, non-distributed mode. This is used for debugging.Its usage is very limited and is used for experimentation.All daemons are as part of single JAVA process

2.Pseudo Distributed Mode
-In this mode also Hadoop runs on a single node but in pseudo distributed mode. Each hadoop daemon runs as a separate JAVA process.Usage is limited and being used for experimentation

3. Fully-Distributed Mode
In this, all daemons are executed as a separate JAVA process on different nodes forming a multi-node cluster. This mode provides read advantages of Hadoop like Fault Tolerant, distributed storage, high availability,data reliability, Scalability.This works in Master-Slave mode of operation.In real world this is being used in all environments as standard mode of operation