Check out the docs at Apache Kafka web site: http://kafka.apache.org/documentation/streams/

Sub-pages:

• Using Kafka Streams:
  • Usage Patterns
  • Application Reset Tool

  • Join Semantics

  • Streams FAQs
  • Data (Re)Processing Scenarios
• Development:
  • Development Plan
  • Discussions
  • Data (Re)Processing Scenarios
• Internals:
  
  • Data Management
  • Architecture

Find more links about Kafka Streams at Kafka Ecosystem page.

Kafka Streams related KIPs:

Accepted

• future release Kafka 2.2:
  • KIP-149: Enabling key access in ValueTransformer, ValueMapper, and ValueJoiner (partially implemented in v1.1)
  • KIP-150 - Kafka-Streams Cogroup

• upcoming release Kafka 2.1:
  
  • KIP-206: Add support for UUID serialization and deserialization
  • KIP-312: Add Overloaded StreamsBuilder Build Method to Accept java.util.Properties

  • KIP-313: Add KStream.flatTransform and KStream.flatTransformValues

  • KIP-319: Replace segments with segmentInterval in WindowBytesStoreSupplier

  • KIP-321: Update TopologyDescription to better represent Source and Sink Nodes

  • KIP-324: Add method to get metrics() in AdminClient

  • KIP-328: Ability to suppress updates for KTables

  • KIP-330: Add retentionPeriod in SessionBytesStoreSupplier

  • KIP-331 Add default implementation to close() and configure() for Serializer, Deserializer and Serde

  • KIP-353: Improve Kafka Streams Timestamp Synchronization

  • KIP-356: Add withCachingDisabled() to StoreBuilder

  • KIP-365: Materialized, Serialized, Joined, Consumed and Produced with implicit Serde

  • KIP-366: Make FunctionConversions deprecated

  • KIP-358: Migrate Streams API to Duration instead of long ms times

  • KIP-372: Naming Joins and Grouping Repartition Topics

• Kafka 2.0:
  • KIP-244: Add Record Header support to Kafka Streams Processor API
  • KIP-245: Use Properties instead of StreamsConfig in KafkaStreams constructor
  • KIP-251: Allow timestamp manipulation in Processor API
  • KIP-261: Add Single Value Fetch in Window Stores
  • KIP-265: Make Windowed Serde to public APIs
  • KIP-267: Add Processor Unit Test Support to Kafka Streams Test Utils
  • KIP-268: Simplify Kafka Streams Rebalance Metadata Upgrade
  • KIP-270 - A Scala Wrapper Library for Kafka Streams
  • KIP-274: Kafka Streams Skipped Records Metrics
  • KIP-276 Add StreamsConfig prefix for different consumers
  • KIP-284: Set default retention ms for Streams repartition topics to Long.MAX_VALUE
  • KIP-292: Add transformValues() method to KTable
  • KIP-295 Add Streams Configuration Allowing for Optional Topology Optimization
  • KIP-303: Add Dynamic Routing in Streams Sink
• Kafka 1.1:
  
  • KIP-149: Enabling key access in ValueTransformer, ValueMapper, and ValueJoiner (partially implemented in v1.1)
  • KIP-171 - Extend Consumer Group Reset Offset for Stream Application
  • KIP-205: Add all() and range() API to ReadOnlyWindowStore
  • KIP-210 - Provide for custom error handling when Kafka Streams fails to produce
  • KIP-220: Add AdminClient into Kafka Streams' ClientSupplier
  • KIP-224: Add configuration parameter `retries` to Streams AP
  • KIP-233: Simplify StreamsBuilder#addGlobalStore
  • KIP-239 Add queryableStoreName() to GlobalKTable
  • KIP-247: Add public test utils for Kafka Streams
• Kafka 1.0:
  • KIP-120: Cleanup Kafka Streams builder API
  • KIP 130: Expose states of active tasks to KafkaStreams public API
  • KIP-138: Change punctuate semantics
  • KIP 157 - Add consumer config options to streams reset tool
  • KIP-160: Augment KStream.print(), KStream.writeAsText() to allow users pass in extra parameters in the printed string
  • KIP-161: streams deserialization exception handlers
  • KIP-167: Add interface for the state store restoration process
  • KIP-173: Add prefix to StreamsConfig to enable setting default internal topic configs
  • KIP-182: Reduce Streams DSL overloads and allow easier use of custom storage engines
  • KIP-198: Remove ZK dependency from Streams Reset Tool
  • KIP-202 Move merge() from StreamsBuilder to KStream
• Kafka 0.11.0:
  • KIP-114: KTable state stores and improved semantics
  • KIP-123: Allow per stream/table timestamp extractor
  • KIP-121: Add KStream peek method
  • KIP-129: Streams Exactly-Once Semantics
  • KIP-134: Delay initial consumer group rebalance
  • KIP 155 - Add range scan for windowed state stores
  • KIP 156 Add option "dry run" to Streams application reset tool
• Kafka 0.10.2:
  • KIP-77: Improve Kafka Streams Join Semantics
  • KIP-90: Remove zkClient dependency from Streams
  • KIP-93: Improve invalid timestamp handling in Kafka Streams
  • KIP-94: Session Windows
  • KIP-97: Improved Kafka Client RPC Compatibility Policy 
  • KIP-99: Add Global Tables to Kafka Streams
  • KIP-100: Relax Type constraints in Kafka Streams API
  • KIP-104: Granular Sensors for Streams

• Kafka 0.10.1:

  • KIP-62: Allow consumer to send heartbeats from a background thread
  • KIP-63: Unify store and downstream caching in streams
  • KIP-67: Queryable state for Kafka Streams
  • KIP-71: Enable log compaction and deletion to co-exist

• Kafka 0.10.0:

  • KIP-28: Add a processor client
  • KIP-32: Add timestamps to Kafka message

Under discussion

• KIP-159: Introducing Rich functions to Streams
• KIP-165: Extend Interactive Queries for return latest update timestamp per key
• KIP-213 Support non-key joining in KTable
• KIP-216: IQ should throw different exceptions for different errors
• KIP-221: Repartition Topic Hints in Streams
• KIP-228 Negative record timestamp support
• KIP 230: Name Windowing Joins
• KIP-258: Allow to Store Record Timestamps in RocksDB
• KIP-262: Metadata should include number of state stores for task
• KIP-300: Add Windowed KTable API in StreamsBuilder
• KIP-307: Allow to define custom processor names with KStreams DSL
• KIP-311: Async processing with dynamic scheduling in Kafka Streams

• KIP-314: KTable to GlobalKTable Bi-directional Join

• KIP-326: Schedulable KTable as Graph source

• KIP-335: Consider configurations for KafkaStreams

• KIP-362: Support dynamic gap session window

• KIP-364: Remove implicit Materialized, Consumed and Produced

• KIP-378: Enable Dependency Injection for Kafka Streams handlers

Inactive:

• KIP-95: Incremental Batch Processing for Kafka Streams

Discarded:

• KIP-116: Add State Store Checkpoint Interval Configuration
• KIP-132: Augment KStream.print to allow extra parameters in the printed string (duplicated by KIP-160)
• KIP-147: Add missing type parameters to StateStoreSupplier factories and KGroupedStream/Table methods (covered by KIP-182)