foreword
When a business event occurs, traditional data-at-rest systems record it but leave the use of the data to a much later time. In contrast, Apache Kafka is a data streaming platform designed to react to business events in real time. Over the past decade, Apache Kafka has become the standard for data streaming. Hundreds of thousands of organizations, including most of the largest enterprises in the world, are using Kafka to take action on what’s happening to their business. Those actions allow them to enhance customer experience, gain new business insights, improve efficiency, reduce risks, and so on, all within a few short seconds.
Applications built on top of Kafka are event driven. They typically take one or more data streams as the input and continuously transform these data into a new stream. The transformation often includes streaming operations such as filtering, projection, joining, and aggregation. Expressing those operations in low-level Java code is inefficient and error prone. Kafka Streams provides a handful of high-level abstractions for developers to express those common streaming operations concisely and is a very powerful tool for building event-driven applications in Java.