Producer throughput optimization in Kafka - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this Kafka producer throughput calculation?

Given the following Kafka producer configuration snippet, what is the effective throughput in messages per second if the batch.size is 16384 bytes, linger.ms is 5 ms, and the average message size is 512 bytes?

Kafka

batch_size = 16384
linger_ms = 5
message_size = 512
messages_per_batch = batch_size // message_size
effective_throughput = messages_per_batch / (linger_ms / 1000)
print(int(effective_throughput))

A 1024

B 6400

C 1638

D 8192

💡 Hint

Calculate how many messages fit in one batch, then divide by linger time in seconds.

The batch size is 16384 bytes, each message is 512 bytes, so 16384 // 512 = 32 messages per batch. The linger time is 5 ms, which is 0.005 seconds. Throughput = 32 / 0.005 = 6400 messages per second. The code computes this exactly and prints 6400.

❓ Predict Output

intermediate

What error does this Kafka producer config cause?

What error will this Kafka producer configuration cause when trying to send messages? Properties: acks=all batch.size=0 linger.ms=10

A No error, messages sent immediately

B TimeoutException due to linger.ms too high

C IllegalArgumentException: batch.size must be > 0

D SerializationException due to missing serializer

💡 Hint

Check the valid range for batch.size in Kafka producer configs.

batch.size=0 is valid in Kafka and disables batching. The producer sends each message as an individual request immediately. No exception is thrown during producer initialization or message sending.

🚀 Application

advanced

Optimize Kafka producer throughput with compression and batching

You want to optimize Kafka producer throughput for a high-volume topic. Which combination of settings will most likely increase throughput without significantly increasing latency?

A Set compression.type=none, batch.size=16384, linger.ms=0

B Set compression.type=gzip, batch.size=1024, linger.ms=50

C Set compression.type=snappy, batch.size=65536, linger.ms=20

D Set compression.type=lz4, batch.size=512, linger.ms=100

💡 Hint

Higher batch size and moderate linger time with fast compression helps throughput.

Using snappy compression is fast and reduces data size. Increasing batch.size to 65536 bytes allows more messages per batch. Setting linger.ms to 20 ms allows batching without too much delay. This combination improves throughput while keeping latency reasonable.

🔧 Debug

advanced

Identify the cause of low Kafka producer throughput

A Kafka producer is configured with batch.size=32768 and linger.ms=100, but throughput is unexpectedly low. Which of the following is the most likely cause?

A batch.size is too small causing frequent sends

B linger.ms is too high causing excessive delay

C acks=all causing slow acknowledgments

D max.in.flight.requests.per.connection=1 limiting parallelism

💡 Hint

Check settings that limit concurrency in sending requests.

Setting max.in.flight.requests.per.connection=1 limits the number of unacknowledged requests to one, reducing parallelism and throughput. This is often the cause of low throughput despite large batch size and linger time.

🧠 Conceptual

expert

Which Kafka producer setting most directly controls the trade-off between latency and throughput?

In Kafka producer configuration, which setting most directly controls the trade-off between latency and throughput by delaying sending to batch more messages?

A linger.ms

B batch.size

C compression.type

D acks

💡 Hint

This setting defines how long the producer waits before sending a batch.

linger.ms defines the maximum time to wait for additional messages before sending a batch. Increasing linger.ms increases batching and throughput but adds latency. It directly controls the latency-throughput trade-off.