CCB-400 Online Practice Questions and Answers

You have an "Employees" table in HBase. The Row Keys are the employees' IDs. You would like to retrieve all employees who have an employee ID between 'user_100' and 'user_110'. The shell command you would use to complete this is:

A. scan'Employees', {STARTROW =>'user_100', STOPROW =>'user_111'}

B. get'Employees', {STARTROW =>'user_100', STOPROW =>'user_110'}

C. scan'Employees', {STARTROW =>'user_100', SLIMIT => 10}

D. scan'Employees', {STARTROW =>'user_100', STOPROW =>'user_110'}

Under default settings, which feature of HBase ensures that data won't be lost in the event of a RegionServer failure?

A. AllHBase activity is written to the WAL, which is stored in HDFS

B. All operations are logged on theHMaster.

C. HBase is ACID compliant, which guarantees that itis Durable.

D. Data is stored on the local filesystem of the RegionServer.

You need to create a "WebLogs" table in HBase. The table will consist of a single Column Family called "Errors" and two column qualifiers, "IP" and "URL". The shell command you should use to create the table is:

A. create'WebLogs',{NAME =>'Errors:IP', NAME =>'Errors:URL'}

B. create'WebLogs','Errors'{NAME =>'IP', NAME =>'URL'}

C. create'WebLogs','Errors:IP','Errors:URL'

D. create'WebLogs','Errors'

You have data already stored in HDFS and are considering using HBase. Which additional feature does HBase provide to HDFS?

A. Random writes

B. Batchprocessing

C. Fault tolerance

D. Scalability

You need to free up disk space on your HBase cluster. You delete all versions of your data that is older than one week. You notice your delete has had minimal impact on your storage availability. This is because:

A. You have large store file indexes

B. HBase has not flushed the MemStore

C. HBasehas not run a minor compaction

D. HBase has notrun a major compaction

You have a table with 5 TB of data, 10 RegionServers, and a region size of 256MB. You want to continue with puts to widely disbursed row ids in your table. Which of the following will improve write performance?

A. Increase your buffer cachein theRegionServers

B. Increase the number of RegionServers to 15

C. Decrease your number of RegionServersto 5

D. Decreaseyour regionsize to 128MB

You want to do mostly full table scans on your data. In order to improve performance you increase your block size. Why does this improve your scan performance?

A. It does not. Increasing block size does not improve scan performance.

B. It does not. Increasing block size means that fewer blocks fit into your block cache. This requires HBase to read each block from disk rather than cache for each scan, thereby decreasing scan performance.

C. Increasing block size requires HBase to readfrom disk fewer times, thereby increasing scan performance.

D. Increasing block size means fewer block indexes that need to be read from disk, therebyincreasing scan performance.

You have a table where keys range from "A" to "Z", and you want to scan from "D" to "H." Which of the following is true?

A. A MultiGet must be issued for rows D, E, F, G, H.

B. The scan class supports ranges via the stop and start rows.

C. All scans are full table scans, the client must implement filtering.

D. In order to range scan, raw scan mode must be enabled.

You have a key-value pair size of l00 bytes. You increase your HFile block size from its default 64k. What results from this change?

A. scan throughput increases and random-access latency decreases

B. scan throughput decreases and random-access latencyincreases

C. scan throughput decreases and random-access latencydecreases

D. scan throughput increases and random-access latency increases

You have tin linage table live in production. The table users as the rowkey. You want to change the existing rowkeys to . Which of the following should you do?

A. Modify the client application to write to both the old table and a new table while migrating the old data separately

B. Use theALTERtable command to modify the rowkeys

C. Use theASSIGNcommand to modify the rowkeys

D. Add a new columnto store the userid