Hadoop – Key Consideration for Optimal Performance – setting up Mapper & Reducer (1)

Configuration File Namemapred-site.xml
Property Name(s)              :  mapreduce.tasktracker.map.tasks.maximum
                                         (deprecated mapred.tasktracker.map.tasks.maximum)
                                         mapreduce.tasktracker.reduce.tasks.maximum
                                                (deprecated mapred.tasktracker.reduce.tasks.maximum)
Description                      :  The maximum number of map/reducer tasks that will be run simultaneously by a task tracker.
Default Value                  :  If no value is set the default is 2. The value specifies that the number of map task slots is based                on the total amount of memory reserved for MapReduce by the sysadmin.
Consideration                 :  These are the maximum number of map/reduce tasks permitted to execute simultaneously per node.If the map/reduce tasks are not CPU intensive,  this value may be equal to the number of cores on each node as long as there is sufficient memory in the system to support the total number of map and reduce tasks.
Recommended Value    : Recommended value: #CPU cores-1
Example                            :
<property>
     <name>mapreduce.tasktracker.map.tasks.maximum</name>
     <value>1</value>
</property>

<property>
<name>mapreduce.tasktracker.reduce.tasks.maximum</name>
     <value>1</value>
</property>

References:
http://hadoop.apache.org/docs/r2.6.0/hadoop-project-dist/hadoop-common/ClusterSetup.html
http://www.datastax.com/dev/blog/tuning-dse-hadoop-map-reduce