i
Search icon

Hadoop Tutorial

Self Learning   /   Study Material   /   Resource Manager
Previous
Next

Resource Manager

Resource Manager

  • It is the ultimate authority in the allocation of resources.

  • It transfers parts of requests to corresponding node managers when receiving the requests for processing where the actual processing takes place.

  • It is the cluster resource arbitrator and decides to allocate the resources available to competing applications.

  • Optimizes the use of clusters such as keeping all resources in use all the time against various constraints such as guarantees of availability, equality, and SLAs

  • Resource Manager has two major components: 

a) Scheduler

1. It is the responsibility of the scheduler to allocate resources to different running applications subject to space limitations, queues,  etc.

2. In Resource Manager, it is called a pure scheduler, which means it does not monitor or track the status of the application.

3. The scheduler does not guarantee to restart the failed tasks if there is an application failure or hardware failure.

4. Perform scheduling, depending on the resource requirements of the application.

5. It has a plug-in policy which is responsible for partitioning the resources of the cluster between the different applications. It has two such plug-ins: Capacity Scheduler and Fair Scheduler, actually used as Resource Manager Schedulers.

b) Application Manager

  • Application Manager is responsible for accepting job submissions.

  • Negotiates the Resource Manager's first container to execute the application-specific Application Master.

  • Manages the Application Masters running in a cluster and provides service in case of failure to restart the Application Master container.

© All rights Reserved | SelfLearning.io