Block storage and replication in Hadoop - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding HDFS Block Replication Factor

In Hadoop Distributed File System (HDFS), what happens if the replication factor of a file is set to 1?

A The file is stored as a single block on one DataNode without any copies.

B The file is split into multiple blocks, each replicated twice on different DataNodes.

C The file is stored on the NameNode only without any DataNode storage.

D The file is replicated on all DataNodes in the cluster.

💡 Hint

Think about what replication factor means for data copies.

A replication factor of 1 means there is only one copy of each block, stored on a single DataNode. No additional copies exist for fault tolerance.

❓ data_output

intermediate

Calculating Number of Blocks for a File

Given a file size of 450 MB and HDFS block size of 128 MB, how many blocks will HDFS create to store this file?

A 3 blocks

B 5 blocks

C 4 blocks

D 6 blocks

💡 Hint

Divide file size by block size and round up.

450 MB divided by 128 MB equals about 3.52, so HDFS creates 4 blocks to store the entire file.

❓ Predict Output

advanced

HDFS Block Replication Command Output

What is the output of the following HDFS command? hdfs fsck /user/data/file.txt -files -blocks -racks

Hadoop

Sample output snippet:

Status: HEALTHY
 Total size: 2560 B
 Total blocks: 2 (avg. block size 1280 B)
 Minimally replicated blocks: 2
 Over-replicated blocks: 0
 Under-replicated blocks: 0

Block replica on datanode1:50010
Block replica on datanode2:50010

A The file has 2 blocks but some blocks are under-replicated.

B The file is missing blocks and is unhealthy.

C The file has 1 block with over-replication.

D The file has 2 blocks, each fully replicated with no under or over replication.

💡 Hint

Look at the replication status lines carefully.

The output shows 2 blocks with minimal replication and no under or over replication, indicating healthy block storage.

❓ visualization

advanced

Visualizing Block Distribution Across DataNodes

You have a file split into 3 blocks with replication factor 3. Which visualization best represents the block distribution across 3 DataNodes?

A Each block is stored on exactly 3 different DataNodes, with blocks distributed evenly.

B Each block is stored on exactly one DataNode only.

C Each DataNode stores all 3 blocks (full replication on each node).

D Blocks are stored randomly with some blocks missing replicas.

💡 Hint

Replication factor 3 means 3 copies of each block on different nodes.

With replication factor 3, each block has 3 copies on different DataNodes, ensuring fault tolerance and balanced storage.

🔧 Debug

expert

Diagnosing Under-Replication in HDFS

You notice that the HDFS report shows some blocks as under-replicated. Which of the following is the most likely cause?

A The replication factor is set to 1 for the file.

B One or more DataNodes storing replicas are down or unreachable.

C The file size is smaller than the block size.

D The NameNode is running out of memory.

💡 Hint

Think about what causes missing replicas in a distributed system.

Under-replication usually happens when some DataNodes holding replicas are offline or unreachable, reducing the number of available copies.