mapreduce - difference between hadoop mr1 and yarn and mr2? -

can pls tell differece between mr1 , yarn , mr2

my understanding mr1 having below components namenode, secondary name node, datanode, job tracker, task tracker

yarn node manager resource manager

is yarn consists of mr1 or mr2 ( or both mr2 , yarn same?)

sorry if asked basic level question

mrv1 uses jobtracker create , assign tasks task trackers, can become resource bottleneck when cluster scales out far enough (usually around 4,000 clusters).

mrv2 (aka yarn, "yet resource negotiator") has resource manager each cluster, , each data node runs node manager. in mapreduce mrv2, functions of jobtracker have been split between 3 services. resourcemanager persistent yarn service receives , runs applications (a mapreduce job application) on cluster. contains scheduler, which, previously, pluggable. mapreduce-specific capabilities of jobtracker have been moved mapreduce application master, 1 of started manage each mapreduce job , terminated when job completes. jobtracker function of serving information completed jobs has been moved jobhistory server. tasktracker has been replaced nodemanager, yarn service manages resources , deployment on host. responsible launching containers, each of can house map or reduce task.