Mapreduce job flow on YARN involves below components.
A Client node, which submits the Mapreduce job.
The YARN Resource Manager, which allocates the cluster resources to jobs.
The YARN Node Managers, which launch and monitor the tasks of jobs.
The MapReduce Application Master, which coordinates the tasks running in the
MapReduce job.
The HDFS file system is utilized for sharing job files between the above entities.
Application Master performs the role of negotiating resources from the Resource Manager and working with the Node Manager(s) toexecute and monitor the tasks.
Application Master requests containers for all map tasks and reduce tasks.Once Containers are assigned to tasks, Application Master starts containers by notifying its Node Manager. Application Master collects progress information from all tasks and aggregate values are propagated to Client Node or user.
Application master is specific to a single application which is a single job in classic mapreduce or a cycle of jobs. Once the job execution is completed, application master will no longer exist.