Yet Another Resource Negotiator (YARN) DATA-SCIENCE-OPTIMIZE Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A YARN cluster administrator notices that a particular NodeManager is reporting high memory usage and is frequently killing containers. The administrator suspects that a single application is requesting more memory than the container's allocated limit. What is the NodeManager's role in enforcing resource limits on containers?

   • AThe NodeManager monitors each container's resource usage and kills containers that exceed their allocated memory or CPU limits.
   • BThe NodeManager requests the ApplicationMaster to reduce the container's memory allocation.
   • CThe NodeManager only monitors CPU usage and relies on the operating system to handle memory limits.
   • DThe NodeManager reports the resource usage to the ResourceManager, which then decides whether to kill the container.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A developer is submitting a new application to a YARN cluster. The developer wants to know who is responsible for requesting the initial container for the ApplicationMaster. What is the correct answer?

   • AThe ResourceManager allocates the first container for the ApplicationMaster when the application is submitted.
   • BThe NodeManager on the first available node allocates the container for the ApplicationMaster.
   • CThe client requests the first container directly from a NodeManager.
   • DThe ApplicationMaster requests its own container from the ResourceManager.
   Reveal Solution Discussion Select an answer first
4. 13application · easy

   A new administrator is learning about the YARN architecture. They are confused about the difference between the ResourceManager (RM) and the NodeManager (NM). Which statement correctly describes the relationship between these two components?

   • AThe ResourceManager and NodeManager are interchangeable; either can be used as the cluster master.
   • BThe ResourceManager executes tasks, while the NodeManager only monitors resource usage.
   • CThe ResourceManager runs on every node, while the NodeManager runs only on the master node.
   • DThe ResourceManager is the global master that schedules resources, while the NodeManager is the per-node slave that manages containers and reports status to the RM.
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   Which YARN scheduler is the simplest and processes jobs in the order they are submitted?

   • AFIFO Scheduler
   • BRound Robin Scheduler
   • CFair Scheduler
   • DCapacity Scheduler
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A developer is tuning a MapReduce job on a YARN cluster. The job's map tasks are running out of memory and being killed by the NodeManager. The developer wants to increase the memory available to each map task. What is the correct way to do this?

   • AIncrease the total memory of the NodeManager by editing yarn.nodemanager.resource.memory-mb.
   • BIncrease the container memory allocation for map tasks in the job configuration (e.g., mapreduce.map.memory.mb).
   • CIncrease the number of map tasks to distribute the memory load.
   • DSet the mapreduce.map.java.opts to a lower value to reduce memory usage.
   Reveal Solution Discussion Select an answer first