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

Elastic

Observability
Open source · NOASSERTION

Elastic is the company and platform behind Elasticsearch, Kibana, Logstash, and Beats — collectively the stack most engineers still call "the ELK stack.

updated 2026-07-01

What it is

Elastic is the company and platform behind Elasticsearch, Kibana, Logstash, and Beats — collectively the stack most engineers still call "the ELK stack." In the observability space, that means log aggregation, metrics, APM, and search-based analytics on a single data store. It's Java-based (Elasticsearch core), with a distributed, sharded architecture built on Lucene. If you've searched logs, built a dashboard in Kibana, or shipped traces into an APM backend at a company with more than a handful of engineers, you've probably touched this stack.

Who builds it and why

Elastic N.V. — the commercial company — drives the vast majority of development. This isn't a community-governed CNCF project (no CNCF maturity level applies here), it's a vendor-led open-core product. 2,503 contributors and 77,145 GitHub stars reflect real adoption, but don't mistake that for a neutral foundation project — Elastic controls the roadmap, the licensing terms, and the commercial upsell path. The motivation is straightforward: sell Elastic Cloud, Elastic Security, and enterprise support on top of a widely-deployed open core.

Production readiness signal

Active commit history through 2026-07-01 and a recent release (v9.4.3) show this is not an abandoned or stagnating project — it's continuously maintained at scale. Elasticsearch has run in production at massive scale for over a decade; operational maturity isn't in question.

What's worth flagging: license is listed as NOASSERTION, which tracks with reality — Elastic moved away from pure Apache 2.0 in 2021 to the Elastic License/SSPL dual license (partly to block AWS from repackaging it), then shifted core Elasticsearch and Kibana to AGPL in 2024 alongside SSPL and Elastic License options. If you're doing license compliance review, don't assume Apache 2.0 — verify per-component, per-version.

Who should use this

  • Teams that need full-text search as a first-class capability alongside logs/metrics — Elasticsearch's Lucene-based search is still best-in-class for this.
  • Organizations already running Kibana/ELK who need APM, security analytics (SIEM), or vector search bolted onto an existing investment.
  • Teams with the operational capacity (or budget for Elastic Cloud) to run and tune a distributed search cluster — this is not a "deploy and forget" system at scale.

Who should NOT use this

  • Small teams wanting simple log aggregation without operating a JVM-based distributed system — the operational overhead (shard management, heap sizing, cluster health) is real and unforgiving under load.
  • Cost-sensitive teams at high log volume — Elasticsearch's storage and compute footprint for logs is notably heavier than purpose-built alternatives like Loki, and Elastic Cloud pricing scales fast.
  • Anyone who needs a clean, unambiguous open-source license for legal/compliance reasons — the AGPL/SSPL/Elastic License split creates real friction for redistribution or embedding.
  • Teams prioritizing metrics-first observability (high-cardinality time series) — Elasticsearch's document model is not optimized for that the way Prometheus/Mimir/Thanos are.

Alternatives

  • Grafana Loki — log aggregation designed to be cheap at scale by indexing only metadata, not full text; weaker search capability but far lower operational cost.
  • OpenSearch — the AWS-led fork of Elasticsearch/Kibana pre-license-change, Apache 2.0 licensed, for teams that want the same technical model without Elastic's licensing terms.
  • Splunk — heavier commercial alternative with stronger enterprise security/SIEM tooling, at a materially higher price point.

Pricing

Not fully open source in the simple sense. Core components are source-available under a mix of AGPL, SSPL, and the Elastic License v2 depending on component and version — not a permissive OSI license across the board. Self-hosting is free under those terms; Elastic Cloud (managed) and paid tiers (security, advanced APM features, support) are commercial. Check licensing per-module before assuming what you can redistribute or fork.

Frequently asked

What is Elastic?+
Elastic is the company and platform behind Elasticsearch, Kibana, Logstash, and Beats — collectively the stack most engineers still call "the ELK stack." In the observability space, that means log aggregation, metrics, APM, and search-based analytics on a single data store.
Who builds Elastic?+
Elastic N.V. — the commercial company — drives the vast majority of development. This isn't a community-governed CNCF project (no CNCF maturity level applies here), it's a vendor-led open-core product.
Is Elastic production ready?+
Active commit history through 2026-07-01 and a recent release (v9.4.3) show this is not an abandoned or stagnating project — it's continuously maintained at scale. Elasticsearch has run in production at massive scale for over a decade; operational maturity isn't in question.
Who should use Elastic?+
Teams that need full-text search as a first-class capability alongside logs/metrics — Elasticsearch's Lucene-based search is still best-in-class for this. Organizations already running Kibana/ELK who need APM, security analytics (SIEM), or vector search bolted onto an existing investment.
Who should not use Elastic?+
Small teams wanting simple log aggregation without operating a JVM-based distributed system — the operational overhead (shard management, heap sizing, cluster health) is real and unforgiving under load.
What are the alternatives to Elastic?+
Grafana Loki — log aggregation designed to be cheap at scale by indexing only metadata, not full text; weaker search capability but far lower operational cost. OpenSearch — the AWS-led fork of Elasticsearch/Kibana pre-license-change, Apache 2.
How much does Elastic cost?+
Not fully open source in the simple sense. Core components are source-available under a mix of AGPL, SSPL, and the Elastic License v2 depending on component and version — not a permissive OSI license across the board.