October 2018
Beginner
220 pages
5h 33m
English
YARN (Yet Another Resource Negotiator) provides a cluster-wide dynamic computing platform for different Hadoop subsystem components such as Apache Spark and MapReduce. YARN applications can be written in any language, and can now utilize the capabilities of cluster and HDFS storage without any MapReduce programming. YARN can be set up in a single node or a cluster node. We will set up YARN in a cluster node.
First, we need to inform Hadoop that the cluster will be using YARN instead of the MapReduce framework for processing; this can be done by editing etc/hadoop/mapred-site.xml, and adding the following entry to it:
<configuration> <property> <name>mapreduce.framework.name</name> <value>yarn</value> </property> ...
Read now
Unlock full access