CODEWITHSUNDEEP
hadoop
Suri
Suri

Reputation: 229

Deploy JAR for MapReduce job in Hadoop Cluster

Sorry about the dum question. I've a 5 node hadoop cluster in CentOS linux.

Name node / Job Tracker / Secondary Name Node / Data node - 2

I wrote a map reduce and compiled the JAR. To which node I need to deploy the JAR?

Thanks in advance.

Upvotes: 2

Views: 929

Answers (1)

Sandeep Singh
Sandeep Singh

Reputation: 8010

In a typical hadoop cluster , there should be client node.Client nodes are used to run client applications and cluster administration tools on a hadoop cluster. The client machines performing the following tasks:

  1. Load data in the HDFS cluster
  2. Submit MapReduce jobs (describing how to process the data)
  3. Retrieve or view the results of the job after its completion
  4. Submit Pig or Hive queries

Hadoop administrator generally configure client machine during the installation and configuration of hadoop cluster. Please identify if you have client node configured in your cluster. If yes, then you can deploy and schedule/submit your job from there. You can also use NameNode to deploy and submit the jar but using namenode for deploying jar is not a efficient way.

If you are running oozie,then you can put your jar on HDFS and you can schedule hadoop jobs through oozie.

Upvotes: 1

Related Questions