Apache Kafka: An Introduction to Stream Processing


Apache Kafka is a distributed stream processing platform capable of handling trillions of events in a day. Originally designed by LinkedIn and later open-sourced, Kafka is now maintained by the Apache Software Foundation. Over the years, Kafka has become an essential component of enterprise architecture, particularly for businesses aiming to build real-time data pipelines and streaming applications.

Apache Kafka: An Introduction to Stream Processing

What is Kafka?

Kafka is a distributed event streaming platform that allows you to:

  • Publish and subscribe to streams of records.
  • Store streams of records in a fault-tolerant manner.
  • Process streams of records as they occur.

In simpler terms, Kafka can be seen as a blend of a messaging system and a distributed storage system with real-time processing capabilities.

Core Concepts

  • Producers: Entities that publish data (events) to Kafka topics.
  • Brokers: Kafka servers that store data and serve client requests.
  • Consumers: Entities that consume (or read) data from Kafka topics.
  • Topics: Categories or named feeds to which records can be published.
  • Partitions: Topics can be divided into partitions, which are the basic data storage units in Kafka. Records in each partition are uniquely identified by a sequence number called the offset.
  • ZooKeeper: An external service that Kafka uses to manage distributed brokers.

Why Kafka?

Scalability: Kafka can handle high throughput, thanks to its distributed nature. You can expand it by adding more nodes to the Kafka cluster.

Fault-Tolerance: Even if a few nodes fail, Kafka continues to run, ensuring data isn’t lost.

Low Latency: Kafka provides real-time handling of data, making it suitable for use cases where immediacy is critical.

Durability: Data in Kafka is persistent and survives node failures.

Common Use Cases

  • Real-time analytics: Analyzing user behavior or system performance in real-time.
  • Monitoring: Observing services and generating alerts based on events or patterns.
  • Log aggregation: Collecting logs from various services and making them available in a central place.
  • Stream processing: Processing and transforming data as it arrives.
  • Event sourcing: Capturing changes to an application’s state as a series of events.

Kafka’s Ecosystem

Apache Kafka is not just about sending and receiving messages. Over time, its ecosystem has grown and now includes:

  • Kafka Connect: A tool for importing data into and exporting data out of Kafka.
  • Kafka Streams: A library for building real-time streaming applications.
  • KSQL: A SQL interface for querying data in Kafka in real-time.
  • Confluent Platform: A more extensive platform built around Kafka, offering additional functionalities and improved manageability.

Apache Kafka has become an indispensable tool in the modern data-driven world, bridging the gap between data sources and applications that need real-time operations. Whether you’re building a microservices architecture, a real-time analytics dashboard, or a complex event-processing system, Kafka offers robust, scalable, and fault-tolerant capabilities to meet the needs of a wide variety of applications.

Integrating Kafka with a Spring Boot Doctor’s Appointment Service

The combination of Apache Kafka and Spring Boot can be a powerful tool in building scalable and responsive applications. In this article, we’ll delve into integrating Kafka with a Spring Boot service focused on managing doctor’s appointments. We’ll highlight how Kafka can enhance this service, making it more resilient and capable of handling high loads.

Overview of the Use Case

Imagine a doctor’s appointment service where patients can book, modify, or cancel appointments. By using Kafka, we can ensure that the appointment notifications (reminders, changes, or cancellations) are communicated to both doctors and patients in real time.

Setting Up Kafka with Spring Boot

1. Maven Dependencies:

Include the Kafka and Spring Kafka dependencies in your pom.xml:


2. Configuration:

In the application.properties or application.yml file, specify the Kafka properties:


3. Producer Configuration:

public class KafkaProducerConfig {

    private String bootstrapServers;

    public ProducerFactory<String, Appointment> producerFactory() {
        Map<String, Object> configProps = new HashMap<>();
        configProps.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, bootstrapServers);
        configProps.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
        configProps.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, JsonSerializer.class);
        return new DefaultKafkaProducerFactory<>(configProps);

    public KafkaTemplate<String, Appointment> kafkaTemplate() {
        return new KafkaTemplate<>(producerFactory());

4. Consumer Configuration:

public class KafkaConsumerConfig {

    private String bootstrapServers;

    private String groupId;

    public ConsumerFactory<String, Appointment> consumerFactory() {
        Map<String, Object> configProps = new HashMap<>();
        configProps.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, bootstrapServers);
        configProps.put(ConsumerConfig.GROUP_ID_CONFIG, groupId);
        configProps.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class);
        configProps.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, JsonDeserializer.class);
        return new DefaultKafkaConsumerFactory<>(configProps, new StringDeserializer(),
                new JsonDeserializer<>(Appointment.class));

    public ConcurrentKafkaListenerContainerFactory<String, Appointment> kafkaListenerContainerFactory() {
        ConcurrentKafkaListenerContainerFactory<String, Appointment> factory = new ConcurrentKafkaListenerContainerFactory<>();
        return factory;

5. Publishing an Event:

Whenever a new appointment is booked, modified, or canceled, an event is sent to Kafka:

public class AppointmentService {

    private KafkaTemplate<String, Appointment> kafkaTemplate;

    public void bookAppointment(Appointment appointment) {
        // ... service logic ...
        kafkaTemplate.send("appointment-topic", appointment);

6. Consuming an Event:

Listen to the topic and take the necessary action (e.g., send a notification):

public class AppointmentNotificationService {

    @KafkaListener(topics = "appointment-topic", groupId = "doctors-appointment-group")
    public void consume(Appointment appointment) {
        // Send notification to doctor and patient

Integrating Kafka with Spring Boot services like the doctor’s appointment example ensures that event-driven actions are processed smoothly. By using this architecture, services can handle large influxes of users while still providing timely feedback and notifications. This setup not only enhances scalability and fault tolerance but also improves the user experience by ensuring timely notifications and responses.

Apache Kafka Interview Questions: A Comprehensive Guide

Apache Kafka is a widely-used distributed event streaming platform that has become a cornerstone for many real-time data processing pipelines. If you’re preparing for an interview centered around Kafka, this article will guide you through some of the most commonly asked questions.

Apache Kafka: An Introduction to Stream Processing

1. What is Apache Kafka?

Apache Kafka is a distributed event streaming platform designed for building real-time data pipelines and streaming applications. It’s horizontally scalable, fault-tolerant, and provides high throughput.

2. What are the core components of Kafka?

  • Producer: Sends messages to Kafka topics.
  • Consumer: Reads messages from a topic.
  • Broker: Kafka servers that store data and handle client requests.
  • Topic: A channel where producers send messages and from which consumers read.
  • Partition: Topics are divided into partitions for scalability and parallel processing.

3. Explain the role of the ZooKeeper in Kafka.

ZooKeeper is used to manage and coordinate Kafka brokers. It helps in:

  • Keeping track of topic, partition details, and replicas.
  • Monitoring broker failures and triggering leader election for partitions.
  • Storing metadata essential for brokers.

4. What are consumer groups?

A consumer group includes multiple consumers that together read data from one or more Kafka topics. The group ensures that each message is read by only one consumer instance.

5. How is data stored in Kafka?

Data in Kafka is stored in topics, which are split into partitions. Each partition is an ordered, immutable sequence of records and is continuously appended to. These records are stored in a set of log files.

6. What is a Kafka topic partition?

A partition is a subset of your data. By breaking topics into partitions, Kafka allows parallelism, as different consumers can read different partitions at the same time.

7. How does Kafka provide durability and fault-tolerance?

Kafka replicates each message across multiple brokers. If one broker fails, messages are available on another broker. This is determined by the replication factor set for the topic.

8. What is a Kafka offset?

An offset is a unique ID given to each record in a Kafka partition. It’s used to uniquely identify each message within the partition and to track the consumption of messages.

9. How can you ensure exactly-once semantics in Kafka?

Kafka 0.11 introduced idempotent producers to ensure that records are written exactly once to the broker even if a producer retries sending messages.

10. How does Kafka handle unprocessed messages?

Messages in Kafka are retained for a configurable amount of time, regardless of whether they’ve been processed. Consumers track their offset position in each partition, enabling them to pick up where they left off.

11. Describe the difference between a leader and a follower in Kafka.

For each Kafka partition, one broker acts as a leader, and others act as followers. The leader handles all reads and writes for the partition, while followers replicate the data.

12. What are the different acknowledgment modes in Kafka?

The acknowledgment modes (acks) determine how the producer gets a confirmation that a message was received:

  • acks=0: No acknowledgment.
  • acks=1: Acknowledgment from the leader.
  • acks=all: Acknowledgment from all replicas.

13. How does log compaction work in Kafka?

Log compaction retains only the last message (by key) within a partition, reducing the size of stored data. It ensures that Kafka retains at least the last known value for each message key within the log of a topic partition.

14. Why are partitions important in Kafka?

Partitions enable horizontal scaling in Kafka, as they can be distributed over multiple brokers. They also allow parallelism in data processing.

15. How do you secure Kafka clusters?

Kafka provides multiple security features:

  • Authentication using TLS or SASL.
  • Authorization using Access Control Lists.
  • Data encryption with TLS.


Apache Kafka plays a crucial role in real-time data processing systems. If you’re interviewing for a position involving Kafka, it’s vital to understand its core concepts, architecture, and potential challenges. This guide has covered some foundational questions, but always be ready for deeper dives and scenario-based questions to test your understanding of Kafka in real-world applications.

Dive into this insightful post on CodingReflex to unlock the power of Quarkus, Java’s revolutionary framework for building ultra-speed applications.

  • For real-time updates and insights, follow our tech enthusiast and expert, Maulik, on Twitter.
  • Explore a universe of knowledge, innovation, and growth on our homepage, your one-stop resource for everything tech-related.

For more information on related topics, check out the following articles: Best Practices for Java Architects on GitHub