github.com/hazelcast/hazelcast ↗
Hazelcast is a unified real-time data platform combining stream processing with a fast data store, allowing customers to act instantly on data-in-motion for real-time insights.
Open this visualization on its own page →
Contributors
329
Lines of Code
273,834
From
2008-12-15
To
2021-02-20
About hazelcast/hazelcast
Hazelcast is a distributed, in-memory data platform designed for real-time processing and analysis. It combines stream processing capabilities with a fast data store, enabling organizations to act on data in motion before it reaches traditional databases or data lakes. The platform unifies multiple workloads including stateful data processing, SQL querying over streaming and batch data, distributed messaging, and low-latency caching, making it suitable for scenarios requiring instant insights and immediate action on incoming data.
The core of Hazelcast is its Jet engine, a built-in data processing system capable of handling massive data volumes with exceptional performance—a single node can aggregate 10 million events per second with sub-10 millisecond latency, while clusters can process billions of events per second. Key features include comprehensive connectors for sources like Kafka and S3, distributed key-value storage with queryable maps, pub-sub and queue-based messaging, support for machine learning model deployment via Python integration, and multi-region data replication through WAN capabilities. The platform provides at-least-once and exactly-once processing guarantees and offers client libraries across Java, Python, Node.js, .NET, C++, and Go.
Hazelcast is built in Java and targets enterprise organizations seeking to modernize applications and build real-time systems. The project is actively maintained, requires JDK 17 or higher for building from source, and employs an extensive test suite to ensure reliability. It is licensed under Apache License 2.0 or the Hazelcast Community License, with the codebase dual-licensed depending on components.