● Beta — KubeEra is in active development. Profiles are live; AI-perception measurement is rolling out.
KubeEra
H

Hazelcast IMDG

App Definition & Delivery
Open source · NOASSERTION

Hazelcast IMDG (In-Memory Data Grid) is a distributed in-memory computing platform — think distributed maps, caches, queues, and compute grid capabilities, all embeddable in a JVM application or run as a standalone cluster.

updated 2026-07-01

What it is

Hazelcast IMDG (In-Memory Data Grid) is a distributed in-memory computing platform — think distributed maps, caches, queues, and compute grid capabilities, all embeddable in a JVM application or run as a standalone cluster. It gives you data structures (Map, Set, Queue, MultiMap, RingBuffer) that are automatically partitioned and replicated across cluster members, plus a distributed execution service for running compute tasks near the data. It's commonly used as a caching layer, a session store, or a backbone for event-driven and stream-processing workloads. Note: Hazelcast has since rebranded/split its product line (Hazelcast Platform now bundles IMDG with Jet stream processing) — if you're evaluating this today, confirm which product line you're actually looking at.

Who builds it and why

Built primarily by Hazelcast Inc., a commercial company that open-sources the core grid and sells enterprise features (security, WAN replication, management center, support) on top. 486 contributors on GitHub is a healthy number for a project this age, suggesting real community engagement beyond the core team, though the majority of commits in projects like this typically come from paid maintainers. The business model is standard commercial-open-source: give away the engine, charge for the operational and enterprise layer.

Production readiness signal

  • 6,581 GitHub stars — respectable but modest for a project that's been in production use for over a decade; this is a workhorse tool, not a hype project.
  • 486 contributors — solid, indicates active multi-team development, not a single-vendor side project.
  • Latest release v5.7.0 — active versioning, feature and patch releases are ongoing.
  • Last commit 2026-07-01 — actively maintained (verify this date against current time when evaluating; if it's genuinely recent, that's a strong signal).
  • License: NOASSERTION — this is a red flag operationally, not necessarily legally. It usually means the repo's LICENSE file isn't machine-parseable by scanning tools, not that there's no license. Confirm the actual license (historically Apache 2.0 for core IMDG) directly from the repo before you build a compliance decision on GitHub metadata.
  • CNCF maturity and founding date: not publicly available in the data provided — Hazelcast IMDG is not a CNCF project, so CNCF maturity levels don't apply here.

Bottom line: this is a mature, battle-tested piece of infrastructure with a long production track record in finance, telco, and e-commerce, even though the metadata here is thin.

Who should use this

  • Java/JVM shops that need a distributed cache or data grid embedded directly in their application tier without standing up a separate Redis cluster.
  • Teams needing distributed compute close to data (executor services, entry processors) rather than just a K/V cache.
  • Organizations already invested in Hazelcast's ecosystem (Jet, management center) looking for a unified in-memory + stream processing stack.

Who should NOT use this

  • Teams whose primary need is a simple, fast K/V cache — Redis or Memcached are lighter-weight and have a much larger ops community.
  • Non-JVM shops. Hazelcast's client libraries exist for other languages, but the deepest integration and best performance is JVM-native; if your stack is Go/Python/Node-heavy, the operational overhead of running a JVM-based grid isn't worth it.
  • Teams wanting a fully managed, batteries-included cloud service without ops overhead — you'll be running and tuning JVM clusters yourself unless you pay for Hazelcast Cloud.
  • Anyone needing strict CNCF-governed, vendor-neutral tooling for procurement reasons — this is vendor-controlled, not a CNCF project.

Alternatives

  • Redis / Redis Cluster — simpler operational model, broader language support, the default choice for most caching needs.
  • Apache Ignite — closer architectural peer to Hazelcast (data grid + compute grid), Apache-governed rather than single-vendor.
  • Infinispan (Red Hat) — JVM-native distributed cache/data grid, tightly integrated with the JBoss/Quarkus ecosystem.

Pricing

Core Hazelcast IMDG is open source. Enterprise features (security suite, WAN replication, rolling upgrades, management center at scale, support SLAs) require a paid Hazelcast Enterprise or Platform Enterprise license. Exact current pricing: not publicly available — contact Hazelcast sales directly.

Frequently asked

What is Hazelcast IMDG?+
Hazelcast IMDG (In-Memory Data Grid) is a distributed in-memory computing platform — think distributed maps, caches, queues, and compute grid capabilities, all embeddable in a JVM application or run as a standalone cluster.
Who builds Hazelcast IMDG?+
Built primarily by Hazelcast Inc., a commercial company that open-sources the core grid and sells enterprise features (security, WAN replication, management center, support) on top.
Is Hazelcast IMDG production ready?+
6,581 GitHub stars — respectable but modest for a project that's been in production use for over a decade; this is a workhorse tool, not a hype project. 486 contributors — solid, indicates active multi-team development, not a single-vendor side project. Latest release v5.7.
Who should use Hazelcast IMDG?+
Java/JVM shops that need a distributed cache or data grid embedded directly in their application tier without standing up a separate Redis cluster. Teams needing distributed compute close to data (executor services, entry processors) rather than just a K/V cache.
Who should not use Hazelcast IMDG?+
Teams whose primary need is a simple, fast K/V cache — Redis or Memcached are lighter-weight and have a much larger ops community. Non-JVM shops.
What are the alternatives to Hazelcast IMDG?+
Redis / Redis Cluster — simpler operational model, broader language support, the default choice for most caching needs. Apache Ignite — closer architectural peer to Hazelcast (data grid + compute grid), Apache-governed rather than single-vendor.
How much does Hazelcast IMDG cost?+
Core Hazelcast IMDG is open source. Enterprise features (security suite, WAN replication, rolling upgrades, management center at scale, support SLAs) require a paid Hazelcast Enterprise or Platform Enterprise license. Exact current pricing: not publicly available — contact Hazelcast sales directly.