By | July 22, 2015

Hadoop local mode, Distributed mode, Pseudo Distributed

[spacer height=”20px”]

There is three different modes are available in hadoop that is

  • Local Mode Hadoop setup
  • Pseudo Distributed Mode Setup
  • Distributed Mode Setup

Let’s describe this terms in more detail

What is Local or Standalone Mode in Hadoop ?

  • This is a default mode of hadoop, for references Hadoop uses local filesystem
  • If you run any job in local mode, it will start running locally means it will use only single jvm process.
  • you can directly point to bin directory of hadoop and run example all the default local or standalone configuration is setup

Note – Alternative Ex. “Hadoop jar myjar.jar ….”

If you mention only jar then jobs will run locally, to run hadoop jobs on cluster mode you should specify “-jar” you need to add hyphen sign

What is Distributed Mode in Hadoop ?



  • If you are having namenode, Jobtracker running on different servers we can call it as a Distributed mode cluster where your replication factor is greater than 2 or 3
  • In a distributed mode you can take advantage of parallel processing and data replication.
  • For example you can consider hadoop production cluster.


What is Pseudo Distributed Mode in Hadoop ?

  • Pseudo distributed mode mainly used by developers to test the code on their machine before deploying it on production cluster.
  • in pseudo distributed mode jobtracker, namenode, tasktracker and datanode services are running on single machine
  • In this case the limitation is you can’t set the replication factor greater than one

[spacer height=”20px”]


Leave a Reply

Your email address will not be published. Required fields are marked *