Why distributed architecture ensures reliability in Kafka - Challenge Your Understanding

Challenge - 5 Problems

🧠 Conceptual

intermediate

Why does distributed architecture improve reliability?

In Kafka, distributed architecture is key to reliability. Which reason best explains why?

A Because distributed architecture reduces network traffic by centralizing data.

B Because all data is stored on a single broker to avoid synchronization issues.

C Because data is stored on multiple brokers, so if one fails, others still have the data.

D Because distributed architecture eliminates the need for data replication.

💡 Hint

Think about what happens if one part of the system stops working.

Distributed architecture stores copies of data on multiple brokers. This means if one broker fails, others can continue serving data, ensuring reliability.

❓ Predict Output

intermediate

What is the output of this Kafka consumer group behavior?

Given a Kafka topic with 3 partitions and 2 consumers in a group, what happens to message consumption if one consumer fails?

Kafka

Consumer group with 2 consumers consuming from 3 partitions.
One consumer stops working unexpectedly.
Which statement is true?

A The remaining consumer takes over the partitions of the failed consumer and continues processing.

B The topic stops sending messages until the failed consumer restarts.

C Messages from the failed consumer's partitions are lost permanently.

D Both consumers stop consuming messages until manual intervention.

💡 Hint

Think about how Kafka balances load among consumers.

Kafka automatically reassigns partitions from the failed consumer to the remaining consumer, ensuring continuous processing.

🔧 Debug

advanced

Identify the cause of data loss in a Kafka cluster

Consider a Kafka cluster with replication factor 3. A topic has min.insync.replicas set to 2. If a producer sends data with acks=all but data is lost after a broker failure, what is the likely cause?

Kafka

Producer config: acks=all
Topic config: replication.factor=3, min.insync.replicas=2
Broker failure occurs
Data loss observed

A The producer sent data when fewer than min.insync.replicas were available, causing data loss.

B The replication factor was too high, causing delays and data loss.

C The producer used acks=0, so no acknowledgments were received.

D The topic had min.insync.replicas set to 3, which caused the failure.

💡 Hint

Check if the producer waited for enough replicas to confirm the write.

If the producer sends data when fewer than the minimum required replicas are in sync, Kafka may accept the write but data can be lost if a broker fails.

📝 Syntax

advanced

Which Kafka producer config ensures highest reliability?

Choose the correct Kafka producer configuration snippet that guarantees no data loss in case of broker failure.

Kafka

KafkaProducer<String, String> producer = new KafkaProducer<>(props);

A "acks" = "0", "enable.idempotence" = false, "retries" = 3

B "acks" = "all", "enable.idempotence" = false, "retries" = 1

C "acks" = "1", "enable.idempotence" = false, "retries" = 0

D "acks" = "all", "enable.idempotence" = true, "retries" = Integer.MAX_VALUE

💡 Hint

Think about settings that prevent duplicate messages and ensure all replicas confirm writes.

Setting acks to all, enabling idempotence, and allowing infinite retries ensures the producer waits for all replicas and avoids duplicates, maximizing reliability.

🚀 Application

expert

Design a fault-tolerant Kafka system for critical data

You must design a Kafka system that ensures no data loss and continuous availability even if multiple brokers fail. Which combination of features is essential?

A Replication factor = 1, min.insync.replicas = 1, acks=1, enable.idempotence=false

B Replication factor >= 3, min.insync.replicas >= 2, acks=all, enable.idempotence=true

C Replication factor = 2, min.insync.replicas = 1, acks=0, enable.idempotence=true

D Replication factor >= 3, min.insync.replicas = 1, acks=1, enable.idempotence=false

💡 Hint

Consider how replication and acknowledgments work together to prevent data loss.

High replication factor with minimum in-sync replicas and full acknowledgments combined with idempotence ensures data is safely stored and no duplicates occur, even if brokers fail.