Practical technology training
Course outline for Apache Kafka
Connect the concepts, work through practical problems, and choose a training plan that fits your goals.
About Apache Kafka
Kafka stores partitioned event logs and supports independently paced producers and consumers. Build a small event-processing workflow and make ordering, delivery, retention, and recovery assumptions explicit instead of assuming every consumer sees an event exactly once.
Apache Kafka Course Objectives
- Model topics, partitions, keys, and consumer groups.
- Implement producer and consumer workflows.
- Test offset, retry, replay, and duplicate-handling behavior.
- Compare Kafka Connect and Kafka Streams responsibilities.
- Diagnose lag and plan bounded recovery and retention.
Pre-requisites
- Programming experience in Java or another language with a supported Kafka client.
- Familiarity with distributed systems, networking, and Linux.
- Basic JSON and application-data modeling skills.
Lab Setup
- Version-pinned local Kafka environment using KRaft, with supplied producer/consumer scaffolds.
- Supported language runtime, container tooling, and synthetic event datasets.
- Use disposable topics and lab identities; multi-broker failures can use supplied traces where laptop resources are limited.
Detailed Course Outline
Event Streaming Foundations
- Events, records, and retained logs
- Producers, brokers, and consumers
- Streaming versus request/response integration
- Selecting an event boundary
Kafka Architecture and Setup
- Topics, partitions, and replication
- KRaft controller responsibilities
- Local broker and client configuration
- Inspecting topics and metadata
Producing Records
- Keys, serialization, and partition selection
- Acknowledgments, batching, and retries
- Idempotent producer concepts and limits
- Handling invalid and failed sends
Consumer Groups and Offsets
- Group membership and partition assignment
- Polling and offset commits
- Rebalances and processing failures
- Separating consumed offsets from completed external effects
Ordering and Delivery Semantics
- Ordering within a partition
- At-most-once and at-least-once processing
- Transactions and their defined scope
- Application deduplication and idempotent consumers
Retention and Replay
- Time/size retention and log compaction
- Resetting offsets safely in a lab
- Backlog growth and catch-up capacity
- Choosing recovery windows from arrival rates
Integration and Stream Processing
- Kafka Connect sources and sinks
- Schema evolution and compatibility responsibilities
- Kafka Streams topology concepts
- Stateful processing and windowing examples
Security and Operations
- Authentication, TLS, and authorization roles
- Consumer lag and broker/resource metrics
- Partition and replication planning
- Failure diagnosis and capacity assumptions
Practical Exercise
- Produce keyed events and process them with a consumer group
- Inject a repeated record and a processing failure
- Replay a bounded range without duplicating accepted application results
- Report offsets, lag, retention, and the limits of the chosen delivery contract
How we train
Online training for Apache Kafka
- Instructor-led live cohorts
- Self-paced learning with expert coaches
- 24x7 cloud labs with end-to-end examples
All sessions are 100% hands-on. Labs and activities are derived from real-world work our engineers deliver.
Classroom training
Available for corporate teams in:
- Bengaluru
- Chennai
- Hyderabad
- Mumbai
- Delhi/Gurgaon/NCR
- Pune
Note: Classroom training is for corporate clients only.
Self-paced hands-on sessions are delivered via VirtualCoach.