Hazelcast Makes New
March 1, 2019
touted an advanced in-memory data grid solution, Hazelcast IMDG 3.12.
Built and proven to provide the performance at scale required by the
world’s largest organizations, IMDG 3.12 advances its technical
superiority with new features focused on data integrity, cloud
interoperability, performance-sensitive data storage and more.
“With a strong foundation of customers in the financial services,
e-commerce and telecommunication industries, many of the world’s largest
organizations rely on Hazelcast for its enterprise-grade performance,
scale, security and stability,” said Kelly Herrell, CEO of Hazelcast.
“The latest release is a significant advancement for in-memory computing
and expands the addressable use cases where customers can benefit from
our industry-leading performance to accelerate their business.”
Hazelcast IMDG 3.12 again drives the industry forward by addressing data
integrity issues during network failures, often described as CAP Theorem
concerns. Hazelcast now uniquely provides both a Consistency (CP)
Subsystem for sensitive concurrency structures which favors consistency
over availability and an Availability (AP) Subsystem for data storage
structures which favor availability over consistency. By offering both
AP and CP subsystems, Hazelcast offers customers the ability to
fine-tune the in-memory data grid to suit the application and provide a
consistent, stable end-user experience. The correctness of the CP
Subsystem has been confirmed with industry standard Jepsen tests.
“With this breakthrough, Hazelcast achieves the highest data correctness
standards in the industry,” said Greg Luck, chief technology officer for
Hazelcast. “Competitive offerings resoundingly fail in this area. By
solving these challenges, customers will be able to use Hazelcast’s
concurrent structures to build their applications, without relying on
additional clustering software, thus simplifying their operations.”
This new capability for both AP and CP subsystems is a technology
breakthrough and builds upon the advancements made in IMDG 3.10 that
solved for “split-brain” failures, a condition that threatens the
integrity of data in a distributed system.
In alignment with the growing trend of enterprises adopting multi-cloud
architectures, Hazelcast IMDG 3.12 also features the industry’s
strongest support for Kubernetes, which has become the de facto cloud
operating system with support from all major cloud providers and
increased adoption on-premises. Verified and validated on Kubernetes,
customers gain another simple method of deploying Hazelcast, regardless
of the cloud architecture. The location proximity between Hazelcast and
the application it supports provides improved performance and enables
new possibilities for microservices and other latency-sensitive
Multi-Model Database Support
Hazelcast IMDG 3.12 now supports first-class JSON data structures across
all language clients, a feature often found in document databases. Out
of the box, internal benchmarks of mixed workloads show that Hazelcast
achieves 4x greater throughput than popular NoSQL offerings. With this
support, Hazelcast now becomes a multi-model platform and can complement
slower NoSQL offerings for use cases that revolve around a need for low
latency access to short-lived information commonly encountered during
web sessions or transactions. Combined with its near limitless scaling
proficiency, Hazelcast IMDG with JSON support thrives where transaction
rates are volatile and quick scaling of services is required.
Upgrade and Failover
Hazelcast IMDG Enterprise also introduces Blue/Green Deployments, which
enables progressive rollouts and mitigates risk during software
upgrades. Additionally, Hazelcast IMDG 3.12 now provides Automatic
Disaster Recovery (DR) failover. In the event a company’s primary site
encounters an outage, Hazelcast will automatically failover to a
secondary DR site.
These two new features, in conjunction with Rolling Upgrades and Lights
Out cluster recovery support, offer unprecedented operational safety for
the world’s most demanding organizations.