Building Data Streaming Applications with Apache Kafka

Using Kafka in Stream processing

Kafka is the persistence queue for data where data is stored in order of time stamp. The following properties of Kafka allow it to occupy its place in most Streaming architecture:

• Persistence queue and loose coupling: Kafka stores data in topics, which store data in order. The data producer does not have to wait for a response from the application that processes data. It simply puts data into the Kafka queue, and the processing application consumes data from the Kafka queue to process it. 
• Fault tolerance: Data stored in Kafka will not be lost even if one or two brokers fail. This is because Kafka replicates topic partition in multiple brokers, so if a broker fails, data will be served from the other broker ...