Category: Big Data Technology

A company is using multiple services to clean the data in its Hive tables. Which service
will it use in the Hadoop ecosystem to export the Hive metadata to these services?

administrator
A company is using multiple services to clean the data in its Hive tables. Which service
will it use in the Hadoop ecosystem to export the Hive metadata to these services?
Atlas
Well Done. Your Answer is Correct Keep it Up!
Kafka
Flume
Tez

A company has deployed an ETL pipeline and wishes to implement data life cycle
tracking, data flow management, data audit and disaster recovery management modules to
protect their data. Which service will be most suitable for this purpose?

administrator
A company has deployed an ETL pipeline and wishes to implement data life cycle
tracking, data flow management, data audit and disaster recovery management modules to
protect their data. Which service will be most suitable for this purpose?
Falcon
Well Done. Your Answer is Correct Keep it Up!
Sqoop
Atlas
YARN

Which of the following is true about the Application Master and Node Manager of a job?

administrator
Which of the following is true about the Application Master and Node Manager of a
job?
Application Master negotiates resources of its job with the Resource Manager.
Well Done. Your Answer is Correct Keep it Up!
Application Master creates execution containers, monitors their resources and reports
them to the Resource Manager.
Application Master distributes resources among applications
Node manager is responsible for managing resources on all the nodes in a cluster.

Which of the following is true with regards to partitioning and bucketing?

administrator
Which of the following is true with regards to partitioning and bucketing?
Bucketed tables can help save time in operations such as mapside joins.
Well Done. Your Answer is Correct Keep it Up!
Partitioned tables provide a performance deficit by organizing a table’s data into
subfolders thereby increasing read times.
Tables can only either be partitioned or bucketed but not both.
Partitioning divides the table into equal buckets for more efficient sampling.

Which of these is true with regards to managed and external tables?

administrator
Which of these is true with regards to managed and external tables?
When an external table is deleted, the underlying directory is deleted as well.
Well Done. Your Answer is Correct Keep it Up!
The directory of a managed table is only created in /apps/hive/warehouse. A managed
table cannot point to a directory elsewhere.
When a managed table is deleted, the underlying directory is not deleted.
When an external table is deleted, the underlying directory is not deleted.

Consider a scenario where a NameNode restarts. Before restart, you had a file named
a.txt which you named to b.txt and then renamed to c.txt. FSImage will have a a.txt while
the edit log will have record of the two renames (a->b)(b->c). When the Namenode restarts
and the FSImage merges with the edit log to produce a new FSImage, what will be the new
FSImage and edit log contain?

administrator
Consider a scenario where a NameNode restarts. Before restart, you had a file named
a.txt which you named to b.txt and then renamed to c.txt. FSImage will have a a.txt while
the edit log will have record of the two renames (a->b)(b->c). When the Namenode restarts
and the FSImage merges with the edit log to produce a new FSImage, what will be the new
FSImage and edit log contain?
FSImage has c.txt with the edit log being empty
Well Done. Your Answer is Correct Keep it Up!
FSImage has b.txt with the edit log having a record of one rename (b->c)
FSImage has b.txt with the edit log having a record of the two renames (a->b)(b->c)
FSImage has a.txt with the edit log having a record of the two renames (a->b)(b->c)
Page 53 of 60
1 … 51 52 53 54 55 … 60