Describe the Hadoop Ecosystem, HDFS, and Data Ingestion Tools DATA-ENGINEERING-OPTIMIZE Practice Questions (Page 1)

Questions 1–5

1. 1expert · medium

   An HDFS cluster is configured with a replication factor of 3. A DataNode fails permanently. The NameNode detects the under-replicated blocks and begins replication. However, the cluster has no spare capacity to store the additional replicas. What is the most likely outcome?

   • AThe NameNode replicates the blocks to the same DataNode that failed.
   • BThe NameNode reduces the replication factor to 2 to match available capacity.
   • CThe NameNode deletes the under-replicated blocks to free space.
   • DThe blocks remain under-replicated until space is freed or new DataNodes are added.
   Reveal Solution Discussion Select an answer first
2. 2application · easy

   A data engineering team is building a new big data platform. They need a component that provides a SQL-like interface to query data stored in HDFS without writing Java code. Which Hadoop ecosystem component should they use?

   • AApache Hive
   • BApache Oozie
   • CApache Zookeeper
   • DApache Flume
   Reveal Solution Discussion Select an answer first
4. 3expert · medium

   A company needs to ingest data from multiple sources: web server logs, a relational database, and a message queue. They want a unified ingestion pipeline that can handle both batch and streaming data. Which combination of tools is most appropriate?

   • AUse Flume for all sources, with a JDBC source for the database and a Kafka source for the message queue.
   • BUse Flume for logs, Sqoop for the database, and Kafka for the message queue, then write all data to HDFS.
   • CUse only Sqoop for all sources, with a generic connector for logs and message queues.
   • DUse only Kafka for all sources, with a JDBC source connector for the database and a file source for logs.
   Reveal Solution Discussion Select an answer first
5. 4foundation · easy

   Which Hadoop ecosystem component provides a SQL-like interface for querying data stored in HDFS?

   • AHive
   • BFlume
   • CZooKeeper
   • DOozie
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   What is a key difference between Hadoop MapReduce and Spark for ETL workloads?

   • ASpark keeps data in memory for faster iterative processing, while MapReduce writes intermediate results to disk.
   • BMapReduce supports real-time stream processing, while Spark only supports batch processing.
   • CMapReduce is better for iterative algorithms because it caches data in memory.
   • DSpark requires a relational database to store data, while MapReduce uses HDFS.
   Reveal Solution Discussion Select an answer first