Cloudera Hadoop Developer Certification Questions and Answer (Dumps and Practice Questions)

Question : Which statement is true

1. Output of the reducer could be zero
2. Output of the reducer is written to the HDFS
3. In practice, the reducer usually emits a single key-value pair for each input key
4. All of the above



Correct Answer 4 :


Explanation: Reducer can either have zero output or more final key-value pairs. And Whatever is the output it will be written to HDFS.

In Practice, In practice, the reducer usually emits a single key-value pair for each input key.

Refer HadoopExam.com Recorded Training Module : 1 and 3





Question : What is data localization ?

1. Before processing the data, bringing them to the local node.
2. Hadoop will start the Map task on the node where data block is kept via HDFS
3. 1 and 2 both are correct
4. None of the 1 and 2 is correct


Correct Answer 2 :


Explanation:

Map Reduce Data Locality: Whenever possible, Hadoop will attempt to ensure that a MapTask on a node is working on a block of data stored locally
On that node via HDFS.

There is no concept of data locality for the Reducers. All mappers in general have to communicate with all reducers.

Refer HadoopExam.com Recorded Training Module : 3





Question : All the mappers, have to communicate with all the reducers...

1. True
2. False



Correct Answer : 1


Explanation: There is no concept of data locality for the Reducers. All mappers in general have to communicate with all reducers.

Refer HadoopExam.com Recorded Training Module : 3

Related Questions

---

Question : Mapper and Reducer runs on the same machine then output of the Mapper will not be transferred via network to the reducer

1. True
2. False



Question : No reducers can start until all the mappers have finished ?

1. True
2. False




Question :

Hadoop will start transferring the data as soon as Mapper finishes it task and it will not wait till last Map Task finished

1. True
2. False


Question : If a Mapper runs slow relative to other than ?



1. No reducer can start until last Mapper finished
2. If mapper is running slow then another instance of Mapper will be started by Hadoop on another machine
3. Hadoop will kill the slow mapper if it keep running if the new one finished
4. The result of the first mapper finished will be used
5. All of the above


Question : What is the Combiner ?


1. Runs locally on a single Mappers output
2. Using Combiner can reduce the network traffic
3. Generally, Combiner and Reducer code is same
4. None of the 1,2 and 3
5. All 1,2 and 3 applicable to the Combiner



Question : Using the Combiner will increase the network overhead ?


1. True
2. False