Confluent Certified Developer for Apache Kafka - Fact Sheet¶

Exam Overview¶

Exam Name: Confluent Certified Developer for Apache Kafka Duration: 90 minutes Questions: 60 multiple-choice questions Passing Score: 70% Cost: $150 USD Valid For: 2 years Delivery: Online proctored

📖 Official Certification Page - Registration and official details 📖 Confluent Developer - Free tutorials and courses 📖 Apache Kafka Documentation - Core Kafka reference

Target Audience¶

This certification is designed for: - Software engineers building applications with Apache Kafka - Data engineers implementing streaming data pipelines - Backend developers working with event-driven architectures - DevOps engineers managing Kafka-based applications - Professionals with 6+ months of Kafka development experience

Exam Domains¶

Domain 1: Application Design (15%)¶

Core Topics: - Event-driven architecture patterns - Topic design and partitioning strategies - Key selection for ordering guarantees - Exactly-once semantics (EOS) configuration - Idempotent producer design

Key Facts: - Kafka guarantees ordering within a partition, not across partitions - Default partition count cannot be decreased after topic creation - Idempotent producers require enable.idempotence=true - Transactional producers enable atomic writes across multiple partitions - Retention can be time-based (retention.ms) or size-based (retention.bytes)

📖 Kafka Design - Architecture and design principles 📖 Topic Configuration - Topic-level configuration reference

Domain 2: Development (30%)¶

Producer Configuration - Critical Settings: - acks=all - Wait for all in-sync replicas to acknowledge - acks=1 - Wait for leader acknowledgment only - acks=0 - No acknowledgment (fire and forget) - retries - Number of retry attempts (default: Integer.MAX_VALUE with idempotence) - batch.size - Batch size in bytes (default: 16384) - linger.ms - Delay before sending batch (default: 0) - buffer.memory - Total memory for buffering (default: 32MB) - max.in.flight.requests.per.connection - Max unacknowledged requests (set to 5 or less with idempotence) - compression.type - Compression codec (none, gzip, snappy, lz4, zstd)

Consumer Configuration - Critical Settings: - group.id - Consumer group identifier - auto.offset.reset - Behavior when no offset exists (earliest, latest, none) - enable.auto.commit - Automatic offset commit (default: true) - auto.commit.interval.ms - Auto-commit frequency (default: 5000) - max.poll.records - Max records per poll (default: 500) - max.poll.interval.ms - Max time between polls before rebalance (default: 300000) - session.timeout.ms - Consumer heartbeat timeout (default: 45000) - fetch.min.bytes - Minimum fetch size (default: 1) - fetch.max.wait.ms - Max wait for fetch.min.bytes (default: 500)

📖 Producer API - Producer API reference 📖 Consumer API - Consumer API reference 📖 Producer Configuration - All producer settings 📖 Consumer Configuration - All consumer settings

Domain 3: Kafka Streams (15%)¶

Key Abstractions: - KStream - Record stream (insert semantics, each record is independent) - KTable - Changelog stream (upsert semantics, latest value per key) - GlobalKTable - Fully replicated table on each instance (for enrichment joins)

Supported Operations: - Stateless: filter, map, flatMap, mapValues, branch, merge, selectKey - Stateful: aggregate, count, reduce, join, window - Windowing: Tumbling, Hopping, Sliding, Session windows

Key Facts: - KStream-KStream joins require a windowed join - KTable-KTable joins produce a new KTable - KStream-KTable joins are non-windowed (latest table value) - State stores use RocksDB by default - Exactly-once processing requires processing.guarantee=exactly_once_v2

📖 Kafka Streams Concepts - Core concepts guide 📖 Kafka Streams Developer Guide - Complete developer reference

Domain 4: ksqlDB (10%)¶

Key Concepts: - Streams - Immutable, append-only collections (backed by Kafka topics) - Tables - Mutable collections with latest value per key - Push Queries - Continuous queries that emit results as data changes (EMIT CHANGES) - Pull Queries - Point-in-time lookups against materialized views - Persistent Queries - Continuously running queries that write to topics

Important Syntax: - CREATE STREAM - Define a stream from a topic - CREATE TABLE - Define a table from a topic - CREATE STREAM AS SELECT - Create derived stream (persistent query) - INSERT INTO - Insert data into a stream - SHOW QUERIES - List running persistent queries

📖 ksqlDB Documentation - Complete ksqlDB reference 📖 ksqlDB Quickstart - Getting started guide

Domain 5: Kafka Connect (15%)¶

Architecture: - Workers - Processes that run connectors and tasks - Connectors - Define the source/sink and configuration - Tasks - Units of work that move data - Converters - Handle serialization (JSON, Avro, Protobuf)

Key Configuration: - connector.class - Connector implementation class - tasks.max - Maximum number of tasks - key.converter / value.converter - Serialization format - transforms - Single Message Transforms chain - errors.tolerance - Error handling (none, all) - errors.deadletterqueue.topic.name - DLQ topic for failed records

📖 Kafka Connect - Connect framework documentation 📖 Connect REST API - REST API reference

Domain 6: Schema Registry (15%)¶

Compatibility Modes: - BACKWARD (default) - New schema can read old data - FORWARD - Old schema can read new data - FULL - Both backward and forward compatible - BACKWARD_TRANSITIVE - Backward compatible with all previous versions - FORWARD_TRANSITIVE - Forward compatible with all previous versions - FULL_TRANSITIVE - Full compatibility with all previous versions - NONE - No compatibility checking

Key Facts: - Default subject naming: <topic>-key and <topic>-value - Schema IDs are globally unique integers - Supports Avro, Protobuf, and JSON Schema - Schemas are stored in the _schemas internal topic - Adding a field with a default is backward compatible - Removing a field with a default is forward compatible

📖 Schema Registry - Schema Registry documentation 📖 Schema Evolution - Compatibility and evolution guide

Critical Numbers to Remember¶