Hadoop, HDFS and Hive — Learn free

Self-check

Questions 1–10 of 10

1. What is the default block size in Hadoop?
Question 1A. 4 MBB. 64 MBC. 128 MBD. 256 MBE. 1 GB
2. How is HDFS best described in relation to Hadoop?
Question 2A. A Hadoop file system, comparable to FAT, NTFS or EXTB. A query language for Hadoop tablesC. The Java framework that schedules Hadoop jobsD. A wrapper that manages table structuresE. A compression codec for Hadoop blocks
3. What does Hive actually store when it presents data as tables?
Question 3A. A full copy of every HDFS file in its own databaseB. Only the compressed version of each fileC. The data blocks but not the file namesD. A cached subset of frequently queried rowsE. No data at all; it only manages table structures on top of HDFS files
4. What are the two ways in which Hive can work with data?
Question 4A. Store data in memory or on tapeB. Store files in HDFS directly, or create tables which create HDFS files at the back endC. Replicate files across nodes, or keep a single copyD. Query files with SQL, or query them with MapReduceE. Store data in HDFS, or store it in a relational database
5. If two 50 MB files are stored directly in HDFS, how many blocks are created?
Question 5A. None until a query is runB. One shared block, because together they are under 128 MBC. Four, because each file is split in halfD. Two, because each file stored in HDFS creates its own blockE. One block per node in the cluster
6. If the same two 50 MB files are loaded through Hive tables instead of directly into HDFS, how are they stored?
Question 6A. In two separate blocks, exactly as with direct HDFS storageB. Both in a single block, because Hive manages its own blocksC. In a Hive-owned database outside HDFSD. In one block per columnE. They are rejected because they are smaller than 128 MB
7. Why do thousands of 1 MB files arriving from social media cause a problem when stored directly in HDFS?
Question 7A. Each creates its own small block, and thousands of small blocks hurt MPP performanceB. HDFS cannot store files smaller than the block sizeC. Hive refuses to build tables over files that smallD. Small files cannot be replicatedE. They exceed the cluster's total block count
8. What block size is described as best practice for Hadoop?
Question 8A. 1 MBB. 50 MBC. 64 MBD. 100 MBE. 128 MB
9. Which kinds of data can Hadoop's open-source framework store?
Question 9A. Structured data onlyB. Unstructured data onlyC. Semi-structured and unstructured data, but not structuredD. Structured, semi-structured and unstructured dataE. Only data that has first been converted to CSV
10. Which statement about Hive is NOT correct?
Question 10A. Hive behaves like tables in a traditional databaseB. Hive is a wrapper on top of files stored in HDFSC. Hive keeps its own separate copy of the data apart from HDFSD. Hive can create HDFS files at the back end when tables are createdE. Hive manages its own blocks