Mapr (HP) Hadoop Developer Certification Questions and Answers (Dumps and Practice Questions)

Question : Without the metadata on the NameNode file can be recovered ?




1. True
2. False

Correct Answer : Get Lastest Questions and Answer :

Without the metadata on the NameNode, there is no way to access the files in the HDFS cluster




Question : Select the correct option ?

1. NameNode is the bottleneck for reading the file in HDFS
2. NameNode is used to determine the all the blocks of a file
3. Access Mostly Uused Products by 50000+ Subscribers
4. All of the above

Correct Answer : Get Lastest Questions and Answer :

Explanation : When a client application wants to read a file
- It communicates with the NameNode to determine which blocks make up the file, and which blocks make up the file
- It the communicates directly with the datanodes to read the data
- The NameNode is not a bottleneck.




Question : Which is the correct option for accessing the file which is stored in HDFS


1. Application can read and write files in HDFS using JAVA API
2. There is a command line option to access the files
3. Access Mostly Uused Products by 50000+ Subscribers
4. 1,2 and 3 are correct
5. 1 and 2 are correct

Correct Answer : Get Lastest Questions and Answer :

Application can access the files using Java API and typically the files are created in the local filesystem and
then moved to the HDFS, there is also one command hadoop fs which is used to access the files in HDFS

Related Questions

---

Question : Which is a true statement regarding MapR output compressions and codec?



1. Default compression codec in MapR is LZ4

2. ZLIB codec provides the highest level of compression but also requires the most CPU cycles.

3. Access Mostly Uused Products by 50000+ Subscribers

4. The data is compressed which is spilled to local disk in sort phase on the Mappers

5. Configure LD_LIBRARY_PATH to enable native codecs



Question : You have written a MapReduce job, which process months data in Mapper for year . However, all the required transformation is done in
Mapper only.

And you just want output to be stored in specified directory. How many reduce tasks you should configure



1. You cannot decide number of reducer. It is decided by Framework only

2. 0

3. Access Mostly Uused Products by 50000+ Subscribers

4. 1



Question : Which of the following property will help, defining number of reducer tasks on job level



1. mapred.reduce.tasks

2. mapred.max.reduce.tasks

3. Access Mostly Uused Products by 50000+ Subscribers

4. mapred.min.reduce.tasks



Question : Which of the following property will help to limit number of reduce tasks across all the Jobs



1. mapred.reduce.tasks

2. mapred.max.reduce.tasks

3. Access Mostly Uused Products by 50000+ Subscribers

4. mapred.min.reduce.tasks



Question : What is the ideal range for mapred.reduce.tasks parameters?



1. between 0.95 to 1.75

2. between 1 to 2

3. Access Mostly Uused Products by 50000+ Subscribers

4. You can choose any range, and it will derived based on relative value.



Question : You have a MapReduce jobs, which create unique data sets and finally insert each record in JDBC database table. Reducer is responsible writing
data in Database. There are chances that your cluster is very heavily loaded and few map tasks and reduce tasks can fail in between and re-launched in
different node. So, which statement is correct for above scenario?




1. to avoid slowness, we should enable speculative execution

2. we should not enable speculative execution

3. Access Mostly Uused Products by 50000+ Subscribers

4. We should reduce number of reducer