Is the use of CGroups recommended for large Hadoop workloads?
Updated -
At this time, it recommended not to use CGroups to implement resource limitations for individual Hadoop tasks.
The YARN Yet Another Resource Negotiator (YARN) component allows limits to be imposed via CPU CGroups via the following YARN executor class:
org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
