Hazelcast Unified Real-Time Data Platform

New 5.5

Optimize geographically dispersed clusters

Client multi-member routing further improves the performance and stability of client applications connecting to geographically dispersed clusters (stretched clusters) by increasing throughput, reducing operation times and improving recovery times.  This feature uses partition groups to define regional subsets of members  and allows clients to prefer the in-region subset when connecting.  For existing customers who are familiar with “smart client” mode , multi-member routing is similar but the initial connections are limited to one partition group rather than the whole cluster. 

Learn More

New 5.5

Run Jet Jobs on specific compute resources

Jet Job Placement Control allows users to isolate and focus the compute component of Hazelcast Platform independently from the storage component. This allows compute jobs to be executed on specific nodes (e.g. lite members), makes the deployment of Hazelcast Platform more versatile and improves its resiliency for compute-heavy processing. Customers can distribute their workloads selectively to meet their needs -  for example, scaling out vector embedding, stream processing or any other compute intensive workloads. 

Learn More

New 5.5

Sustained and controlled innovation within a cadence that helps drive success

Hazelcast Platform has introduced long-term support (LTS) releases to simplify the overall upgrade experience. Upgrade directly from a previously supported release to an LTS release or directly between LTS releases using the rolling upgrade process with zero downtime.

In addition, between LTS releases, Hazelcast will provide short-term support (STS) releases. Choose to upgrade directly from one LTS to the next LTS, or upgrade from one LTS to each STS release and then from the last STS release to the next LTS release. 

Upgrade directly from 5.2.x or later to 5.5 without the need to upgrade to each release between the current version and the LTS release. It is recommended that you upgrade to 5.5 from the latest available patch for your current release.

Learn More

Continuously process data with easy-to-build distributed applications

Many modern data infrastructures need to keep data continuously updated to provide a real-time view of the business. But without the right application framework, it is difficult to write distributed, parallelized, scalable, resilient, and secure applications that keep your data up to date.

The Hazelcast Platform distributed compute engine enables large-scale data processing that efficiently takes advantage of your cluster’s resources. You write applications that process your data, and your Hazelcast cluster handles task orchestration so you can focus on the high value code, not coordination code.

Learn More

Ensuring strong consistency without compromising on performance, setting a new standard against other systems.

Sometimes you need the fastest performance possible. Sometimes data consistency is more important than blazing speed. With Hazelcast Platform, you get to choose. Our regular data structures are really fast and eventually consistent. Our newly improved Advanced CP Subsystem structures ensure data remains consistent across the CP groups while still providing high performance. Some of our largest, most demanding financial services customers count on Hazelcast Platform to ensure fast, strong consistency of their key mission-critical data. Learn more

Elevating Hazelcast Platform's performance to lead the industry, optimizing resource usage for unparalleled efficiency.

Thread per Core (or TPC) is a more recent approach to the problem of how to best use the multiple cores that are standard across virtualized compute. The traditional approach, staged event-driven architecture (SEDA), made use of threads dedicated to particular tasks in a data pipeline. When a task was done, a thread passed the event to another thread for executing the next stage. With today’s faster processors and more numerous threads, that approach can cause significant delay in the form of context-switching between the threads and there is a resulting inefficiency when trying to scale. TPC is a more modern approach that has a thread hold a event all the way through the data pipeline. This eliminates the performance hits of context switching and allows the system to scale linearly through adding more cores and threads. Hazelcast Platform now uses TPC in its core and will add more TPC-enabled data structures in coming releases. Learn more

Get faster access to data as one part of many other capabilities in a single platform

Need a cache to speed up your data accesses? We have you covered. Hazelcast Platform provides all the caching features you need to meet your speed, scale, resilience, and security requirements. It has a proven track record with the most demanding companies around the world.

In addition to caching, Hazelcast Platform provides other capabilities that you will need for your data-driven strategies. Avoid technical debt by exploring how distributed compute and stream processing capabilities in Hazelcast Platform can solve your big picture needs. Learn more

The core engine for processing your data in motion

Hazelcast Platform enables event stream processing and fast batch processing at any scale. It retrieves live data from databases, data lakes, applications, devices, and message brokers like Apache Kafka, Apache Pulsar, AWS Kinesis, or RabbitMQ. It then transforms raw, high-volume data streams into business events and actionable insights, making them easily consumable by applications, dashboards, and databases.

Reduce complexity and avoid disruptive garbage collection

High-performance, real-time systems cannot afford the disruption caused by Java garbage collectors (GC). Garbage collection is most disruptive when large amounts of memory is allocated in the Java virtual machine. To avoid this, DevOps teams typically deploy more nodes with less data per node, creating management complexity just to avoid GC disruption.

The Hazelcast High-Density Memory Store feature is a non-Java memory allocator that leverages off-heap memory to avoid GC disruptions. This helps to improve overall system performance while reducing node count to lower maintenance overhead. Learn more

The multi-purpose engine for your real-time deployments

Hazelcast connects a set of networked/clustered compute resources to let applications share data structures and run parallelized workloads in the cluster.

The primary advantage is speed, which has become critical in an environment with billions of mobile, IoT devices and other sources continuously streaming data. With all relevant information in RAM, there is no need to traverse a network to remote storage for transaction processing. The difference in speed is significant – minutes vs. sub-millisecond response times for complex transactions done millions of times per second.

Out-of-the-box connectivity to your existing data platforms

The platform empowers you to effortlessly integrate diverse applications and data systems, eliminating the need for additional code. Take your data architecture to the next level with our extensive range of pre-built connectors, including support for all connectors in the Kafka Connect ecosystem, enabling swift modernization and powerful integrations at any scale.

Learn More

Leverage SQL for both data in motion and data at rest

SQL support provides a familiar interface for running queries and offers the following benefits:

• Industry-standard querying: Query large volumes of data using an industry-standard approach, maintaining the same query specificity as the existing Hazelcast Predicate-based design.
• High-performance indexing: Take advantage of new, high-performance concurrent off-heap B+ tree indexes to optimize query performance.
• Advanced query optimization: Benefit from advanced query optimization techniques to enhance the efficiency of your queries.

Learn more

Manage and monitor your Hazelcast deployments

Hazelcast Management Center offers scripting and console modules to run scripts (JavaScript, Groovy, etc.) and commands across your Hazelcast cluster, however it is deployed. The visual tool helps analyze data flow and identify bottlenecks in real time. Developers gain real-time cluster insights.

During development, Management Center provides deep insights. In production, it can be used by IT operations or integrated with enterprise monitoring tools via REST and JMX. With a commercial license from Hazelcast, you can use Management Center to monitor a cluster of any size.

Learn more

Protect your data with role-based access controls and encryption

Hazelcast ensures industry-leading security with end-to-end TLS encryption, mutual authentication using X509 certificates, and roles-based authorization via the standard Java Authentication and Authorization Service (JAAS). It seamlessly integrates security into your application, streamlining the protection of sensitive data. By combining industry security standards and user-friendly APIs, it maintains optimal performance while providing peace of mind.

Learn more

Keep your systems running even in the event of a site-wide disaster

When site-wide disasters occur, business operations can come to a costly and damaging halt. With a proper disaster recovery strategy, you can ensure that your operations will continue with minimal downside.

Hazelcast Platform provides an efficient WAN Replication feature to incrementally copy data from the primary cluster to a standby cluster (or any of multiple topologies including active-active) to ensure an identical cluster with a very small recovery point objective (RPO). With its Automatic Disaster Recovery Failover feature, all clients/applications are quickly redirected to the backup cluster to minimize recovery time objective (RTO). Learn more

Reduce maintenance downtime along with upgrade risk

Blue-green deployment refers to an IT practice which reduces maintenance downtime and associated risk by deploying two separate clusters, “blue” and “green.” During normal operations, the blue cluster is in production and the green cluster is for staging.

When the green cluster is ready for production, clients and applications are redirected from the blue cluster to the green cluster. Upon error, clients/applications are reverted back to the original blue cluster. Hazelcast Platform provides the switchover capability to quickly alert applications which cluster to access for production use. Learn more