Skip to main content

Advances, Systems and Applications

Table 3 Hadoop cluster memory configurations

From: Source camera identification: a distributed computing approach using Hadoop

Configuration name

Default value (MB)

Details

yarn.scheduler.minimum-allocation-mb

32

Minimum resource allocation for every container.

yarn.scheduler.maximum-allocation-mb

11,520

Maximum resource allocation for every container

yarn.nodemanager.resource.memory-mb

11,520

Total amount of resource that can be allocated for the containers

mapreduce.map.memory.mb

1440

The amount of memory for each map task.

mapreduce.map.java.opts

-Xmx1152m

Maximum JVM heap memory for map task.

mapreduce.reduce.memory.mb

2880

The amount of memory for each reduce task.

mapreduce.reduce.java.opts

-Xmx2304m

Maximum JVM heap memory for reduce task.

yarn.app.mapreduce.am.resource.mb

2880

The memory allocation for MapReduce Application Master.