Producer Default Partitions and Key Hashing

The video tutorial explains the process of key hashing in Kafka, focusing on the serialization of keys into bytes and their transformation using the Murmur2 algorithm. It discusses the default partitioner class, advising against overriding it unless necessary. The tutorial also covers the hashing formula used in Kafka, emphasizing that the same key will always map to the same partition unless the number of partitions changes. This change can disrupt the hashing formula, leading to different partition assignments for the same key. The content is aimed at students curious about advanced Kafka settings.

What algorithm does Kafka use for hashing keys?

• MD5

• CRC32

• Murmur2

• SHA-256

Why is it generally not recommended to override the default partitioner in Kafka?

• It requires additional hardware

• It is not supported by Kafka

• It may lead to inefficient data distribution

• It is too complex to implement

What is the purpose of a custom partitioner class in Kafka?

• To reduce latency

• To provide more control over data distribution

• To increase the number of partitions

• To enhance security

What happens if you change the number of partitions in a Kafka topic?

• The topic becomes read-only

• The same key may go to a different partition

• The data is automatically rebalanced

• The hashing formula remains unchanged

What is the result of applying the target partition formula in Kafka?

• The data is compressed

• The record key is encrypted

• A new topic is created

• The target partition for a record key is determined