ITTA offers an Apache Kafka Fundamentals course covering the distributed event streaming platform: architecture (brokers, topics, partitions, replicas), producers and consumers, Kafka Connect, Kafka Streams, ksqlDB, schema registry, security (SASL, TLS, ACL), operations. Audience: back-end developers, data engineers, data architects, SREs, platform engineers. Sessions delivered in Geneva, Lausanne and interactive virtual classroom.
FILTER BY
- Domains
- Editors
- Location
- Format
- Level
- Certifying
- Confirmed training
Apache Kafka was built with the vision to become the central nervous system that makes real-time data available to all the applications that need to use it.
Apache Kafka was created in 2011 at LinkedIn by Jay Kreps, Neha Narkhede and Jun Rao, to solve the challenges of centralising data flows at scale. Open-sourced under the Apache Foundation, Kafka has become in a few years the most used distributed event streaming platform. The three creators then founded Confluent, which commercialises Confluent Platform and Confluent Cloud (managed Kafka multi-cloud).
In 2026, Kafka is a technical foundation present in most modern data and application architectures. It serves to decouple data producers and consumers via persistent topics, to propagate business events in real time between microservices, to feed analytical pipelines (streaming ETL, fraud detection, recommendation), to integrate heterogeneous systems via Kafka Connect, and to build stream processing applications via Kafka Streams or Apache Flink. On the evolution side, Kafka 3.x and 4.x have marked the progressive exit of ZooKeeper (KRaft mode now by default), simplifying deployment and operations.
Our Apache Kafka course at ITTA:
The course covers the entire platform in a project logic: Kafka architecture (brokers, topics, partitions, replicas, leader/follower, ISR), operating modes (KRaft vs ZooKeeper legacy), producers (configuration, acks, idempotence, transactions, batching), consumers (consumer groups, offset management, rebalancing, cooperative sticky assignor), message management (partition keys, ordering, retention, compaction), Kafka Connect for integration with external sources/sinks (databases via Debezium CDC, S3, Elasticsearch, HDFS), Kafka Streams for stream processing (KStream, KTable, windowing, joins, exactly-once), introduction to ksqlDB (SQL on Kafka), schema registry (Avro, Protobuf, JSON Schema), security (SASL, TLS, ACL), observability (JMX, Prometheus, Cruise Control), operations (sizing, monitoring, troubleshooting). The format is hands-on with a real Kafka cluster.
Kafka serves as backbone for asynchronous communication between microservices, replacing synchronous REST calls. Producers publish business events (order created, payment validated, stock updated), consumers react according to their responsibilities. The resulting event-driven architecture is more resilient and scalable.
Kafka Connect with Debezium captures changes in relational databases (PostgreSQL, MySQL, SQL Server, Oracle) and republishes them as events (CDC, change data capture). These streams feed in real time S3/ADLS data lakes, Snowflake/BigQuery warehouses, Elasticsearch search engines, or specialised data marts.
Kafka Streams or Apache Flink consume Kafka streams to produce real-time analytics: fraud detection, live aggregations (marketing KPI, sales), data enrichment (joins with reference KTables), personalisation. Sub-second latency enables use cases impossible in batch.
Kafka centralises application and infrastructure logs, then consumed by Elasticsearch, OpenSearch, Loki or Splunk. This architecture allows decoupling log production (services) from their consumption (observability tools).
Kafka fits in a broader landscape covered by our data catalogue. The database design and development sub-domain regroups modelling, relational and NoSQL database training. The data science sub-domain brings analytical and applied AI uses. The data and databases sub-domain on the IT pro side covers administration and BI competencies.
On the publisher side, the Apache Cassandra publisher brings the distributed NoSQL dimension, frequently coupled with Kafka for high-volume architectures. The Apache Hadoop publisher covers the historical big data ecosystem (HDFS, MapReduce, Hive, Spark), which often coexists with Kafka for streaming. The Open Source publisher regroups our open technologies training.
Our Kafka audience is broad: back-end developers needing to integrate Kafka in their microservices, data engineers building streaming ETL pipelines with Kafka Connect, data architects validating a Kafka choice for a new event-driven architecture, SREs needing to operate a Kafka cluster in production (sizing, monitoring, troubleshooting), platform engineers industrialising Kafka use for their internal teams, consultants intervening on data or application modernisation projects in French-speaking Switzerland and internationally.
The 2026 event streaming landscape offers several options. Apache Kafka remains the dominant option for substantial architectures, with a rich ecosystem (Connect, Streams, ksqlDB, Schema Registry). Apache Pulsar is a credible alternative for multi-tenant and geo-distributed contexts. Redpanda is a Kafka clone written in C++ positioning itself on operational simplicity and low-cost performance. Managed services (Confluent Cloud, AWS MSK, Azure Event Hubs with Kafka interface, Aiven Kafka) delegate operations. RabbitMQ and ActiveMQ remain relevant on smaller cases or transactional messages. Our course addresses these trade-offs at session start.
Several trends shape Kafka in 2026. KRaft mode (without ZooKeeper) has become the norm, simplifying deployment. Managed services (Confluent Cloud, AWS MSK Serverless) gain adoption for organisations wanting to avoid operations. Tiered storage (separates hot and cold storage) allows keeping very long retentions at controlled cost. Integration with Flink (for advanced stream processing) progresses to the detriment of Kafka Streams on demanding cases. Generative AI enters the data pipeline via RAG, often relying on Kafka streams for real-time ingestion. Schema governance (schema registry, evolution) becomes a major topic on organisations with many topics and teams.
Our Apache Kafka sessions are scheduled in Geneva, Lausanne and in interactive virtual classroom with a live trainer. The format is very practice-oriented on a real Kafka cluster. Material modalities are communicated in advance by our education team. For data or platform teams seeking grouped upskilling on their real Kafka cluster, we organise in-house sessions calibrated on your stack. This modality is well suited to banking, telecom, public sector, e-commerce and industrial contexts.
Do I need Java knowledge before this course?
Java culture helps for producer/consumer examples and Kafka Streams. The course also addresses Python clients (confluent-kafka), Node.js and Go for non-Java profiles. Prior programming experience is required.
Kafka Streams or Apache Flink?
Kafka Streams is relevant for simple to moderate cases, deployed as embedded library. Flink is more powerful for advanced cases (large state, distributed exactly-once, complex windowing). Our course introduces both and helps choose.
Confluent Cloud, AWS MSK or self-managed?
Self-managed offers complete control but requires real operational expertise. Confluent Cloud and AWS MSK Serverless delegate operations with additional cost. The choice depends on internal skills, budget and sovereignty constraints. Our course addresses these trade-offs.
Is Schema Registry addressed?
Yes, Schema Registry (Confluent, Apicurio) is introduced as the recommended path to govern schema evolution (Avro, Protobuf, JSON Schema) on organisations with many topics.
ITTA offers a coherent data catalogue from big data fundamentals (Hadoop, Cassandra) to event streaming (Kafka), relational databases, cloud data engineering (AWS Data Engineering, Azure Databricks, Microsoft Fabric) and applied AI. This continuity allows addressing a complete data stack. Our Kafka trainers are data engineers and architects active on Kafka projects in French-speaking Switzerland, providing concrete and current examples. Sessions available in Geneva, Lausanne and interactive virtual classroom, in-house and inter-company.
Monday to Friday
8:30 AM to 6:00 PM
Tel. 058 307 73 00
ITTA
Route des jeunes 35
1227 Carouge, Suisse
Monday to Friday, from 8:30 am to 06:00 pm.
