Responsive SDK


extensions for Kafka Streams.

The Responsive SDK is a library extension for Kafka Streams that swaps out internal components to enable remote storage, asynchronous processing and additional features. The improved architecture eliminates unpredictable lag due to restoration and rebalances while maintaining 100% code compatibility.

Eliminate Restores
Rebalancing and Restoration are the leading causes of lag in Kafka Streams applications. Never worry about them again thanks to remote state.
Break Partition Limits
Open source Kafka Streams supports parallelism up to the number of Kafka partitions. The async processor supports parallelism up to the number of keys.
Expand the API
The Kafka Streams API is powerful, but misses critical features like Time-to-Live.
Java SDK
Eliminate lag by changing one line of code.
ResponsiveKafkaStreams extends KafkaStreams, so you can be confident none of your business logic needs to change.
Disaggregated Storage
Responsive supports remote state stores, meaning no more RocksDB or other local state. With no state restorations and instant rebalances, the #1 cause of Kafka Streams application lag is eliminated. Learn More.
Async Processing
Async processing allows your apps to scale beyond Kafka partition limits by delivering key-level parallelism for Kafka Streams. Unlock high throughput for operations like remote calls and CPU intensive tasks. Read the blog post.
BYO Storage
Responsive SDK allows you to bring your own MongoDB or Cassandra compatible database, ensuring data never leaves your security perminter. JDBC and DynamoDB support coming soon. Check your options.
Reduce Kafka Costs
Since Responsive doesn’t run with standby replicas, there is no longer any egress on the changelog topics (we still use them as a transactional write-ahead-log). 
Time-To-Live
Responsive supports row-level TTL, which means you can expire table entries without any punctuations or out of band processing. See How.
Simplified Deployment
Stateless Kafka Streams means a simplified deployment model. Run without StatefulSets on Kubernetes and don’t worry about standby or warmup replicas. 

Ditch RocksDB to level up Kafka Streams.

The separation of compute and storage is the hallmark of modern systems, and yet most stream processors embed RocksDB in their compute nodes. This coupling results in severe operational pain.

Want to see rebalances disappear?

Get your free 60 day trial for the Responsive Java SDK.