Submitting MapReduce job using jobConf object vs job object

Question

I am new to Hadoop & I am a bit confused about how to submit a job to Hadoop. Earlier I was submitting the job using the following command:

JobClient.runJob(jobconf);

But while going through an online tutorial, I found out that they are submitting the job using the below command:

Configuration conf = getConf();
Job job = new Job(conf, "secondary sort");

Can anyone explain me what is the difference between jobconf & job object? Which is the right way of submitting the job?

Answer 1

Both JobConf & Job objects are used for configuring & submitting the job. You can also control the execution using the objects. The only difference between them is JobConf object belong to the org.apache.hadoop.mapred package, which is the deprecated API, while Job object is from org.apache.hadoop.mapreduce package, which is the new API.

I would recommend using the new API.

You can refer the following link for the official documentation.

JobConf: https://hadoop.apache.org/docs/r2.7.3/api/org/apache/hadoop/mapred/JobConf.html

Job: https://hadoop.apache.org/docs/r2.7.3/api/org/apache/hadoop/mapreduce/Job.html