Hazelcast has officially released Hazelcast Platform 5.7.0, an update that introduces production-grade support for the Java 25 runtime environment. The release, launched on May 28, 2026, aims to strengthen the operational foundations of enterprise systems that rely on real-time processing.

🐞 Bug Fixes Native Image Hazelcast error: java.lang.NoSuchMethodError: com.hazelcast.map.IMap.getName() #5604 📔 Documentation Add documentation for @Counted #5613 🔨 Dependency Upgrades Bump or...

Compare Hazelcast to Redis with this in depth side-by side benchmarks and use cases.

Before the holiday season kicks in, we’re excited to host one more RotterdamJUG meetup in 2024! And again we've found a new host: **WeAreFrank!**, an open-source organizat

Software architecture

Before the holiday season kicks in, we’re excited to host one more RotterdamJUG meetup in 2024! And again we've found a new host: **WeAreFrank!**, an open-source organizat

# HazelCast: Distributed Computing Platform Distributed systems are complex. Managing data across multiple nodes, ensuring consistency, handling failures—these challenges grow exponentially as your application scales. That's where HazelCast enters the picture. It's an in-memory data grid and distributed computing platform that makes building scalable, fault-tolerant applications surprisingly approachable. ## What Is HazelCast? At its core, HazelCast is an open-source, distributed in-memory data grid (IMDG) written in Java. Think of it as a distributed cache on steroids: it stores data in memory across a cluster of machines, allowing you to access that data with sub-millisecond latency while maintaining high availability and automatic failover. Unlike traditional caching layers that sit passively between your app and database, HazelCast actively participates in your system. It offers distributed queues, maps, topics, locks, and semaphores—all synchronized across nodes automatically. No heartbeat requests, no stale copies: when one node updates a value, every other node sees the change immediately. ## Why HazelCast Matters for Java Developers **Speed:** In-memory operations are orders of magnitude faster than disk or network I/O. HazelCast gives you that speed without the complexity of managing a separate distributed system. **High Availability:** Data is automatically replicated across cluster members. If one node crashes, your data survives. If the network partitions, HazelCast can be configured to handle split-brain scenarios gracefully. **Embedded or Managed:** Run HazelCast as an embedded library inside your Spring Boot app, or connect to a standalone server cluster. Both approaches work seamlessly—your code doesn't change. **Built for the JVM:** Native Java support, Spring integration, and Kubernetes operators mean it feels native to your stack, not like a third-party bolted-on tool. ## A Practical Example // Start an embedded HazelCast instance HazelcastInstance instance = Hazelcast.newHazelcastInstance(); // Get a distributed map IMap<String, String> config = instance.getMap("config"); // Put data (replicated across the cluster automatically) config.put("feature_flag_new_ui", "enabled"); // Read from any node—same result String flag = config.get("feature_flag_new_ui"); // "enabled" // Listen for changes config.addEntryListener((event) -> { System.out.println("Config updated: " + event.getKey()); }, false); That's it. Your app now has a shared, distributed data store, and every node in your cluster sees updates in real-time. ## When to Use HazelCast **Session storage:** Distributed web sessions across multiple app instances without sticky sessions. **Configuration management:** Shared config that updates live across the cluster. **Rate limiting & quotas:** Distributed counters and atomic references ensure fairness. **Real-time analytics:** Stream events through distributed topics, aggregate results in memory. **Distributed locks:** Ensure critical sections run safely across a cluster. ## The Tradeoffs HazelCast isn't a silver bullet. All data lives in memory, so heap size is your limit. Network partitions and Byzantine failure modes demand careful configuration. And like any distributed system, debugging becomes harder—partial failures and timing issues can be tricky to reproduce. But for Java teams building microservices, especially those already using Spring Boot and Kubernetes, HazelCast brings order to the chaos of distributed state management. It abstracts away the plumbing and lets you focus on business logic. ## Next Steps Start with the embedded mode—add it to a Spring Boot app, experiment with distributed maps, and feel how natural it becomes. The official docs are excellent, and the community is active. Once you've felt the power of instant, replicated data across your cluster, you'll wonder how you ever built distributed systems without it. Have you used HazelCast in production? What problems did it solve for your team?

Deserialization of Untrusted Data vulnerability in Apache Camel Hazelcast component. The camel-hazelcast component creates and manages Hazelcast instances using a default configuration that applies no Java deserialization filter. When Camel builds the Hazelcast Config itself - that is, when no user