In the rapidly evolving world of search, analytics, and observability, OpenSearch has emerged over the last few years as a prominent open-source alternative to Elasticsearch. Its story is deeply intertwined with licensing controversies, cloud economics, and the open-source ethos. Here’s a detailed look at how OpenSearch came to be, who built it, why it matters — and what it looks like today.
1. Origins & Genesis
1.1 The License Shift by Elastic
The seeds of OpenSearch were sown when Elastic NV, the company behind Elasticsearch and Kibana, changed the licensing of their products. Until version 7.10, Elasticsearch and Kibana were licensed under the permissive Apache 2.0 license, which allowed broad use, modification, and redistribution. Elastic+2datanami.com+2
However, in early 2021, with the release of Elastic Stack 7.11, Elastic moved to a dual-license model: SSPL (Server Side Public License) and the Elastic License v2. Elastic+2Elastic+2
This change was highly controversial: SSPL is not an OSI-approved open-source license, and critics argued that Elastic was limiting how cloud providers could offer Elasticsearch as a managed service. theregister.com+2datanami.com+2
Elastic defended the change, saying it was necessary to protect their business model: they wanted to prevent cloud providers from repackaging Elasticsearch as a service without contributing “back” their modifications. Elastic
1.2 AWS Responds — From Open Distro to OpenSearch
Amazon Web Services (AWS) had already been uneasy with the trajectory. AWS had launched Open Distro for Elasticsearch earlier (in 2019) — a fully open-source distribution of Elasticsearch including various plugins for security, alerting, SQL, etc. theregister.com
But after Elastic’s relicensing, AWS decided to go further: in January 2021, it announced OpenSearch, a clean fork of Elasticsearch and Kibana. OpenSearch+1
Just a few months later, in July 2021, OpenSearch 1.0 was officially released under the Apache License, Version 2.0.
That decision meant that OpenSearch would remain truly open source, with no restrictive licensing preventing wide use or forking.
1.3 Community & Foundation
OpenSearch is now governed by the OpenSearch Software Foundation, which ensures that development is not purely driven by a single corporate entity. Wikipedia+1
Since its inception, the project has grown significantly: hundreds of contributors, dozens of repositories, and a roadmap that spans search, analytics, observability, security, and more. docs.opensearch.org
Over time, AWS was joined by other major players: companies like Red Hat, SAP, Logz.io, and Capital One showed interest or support. Wikipedia+1
In 2024, the project further strengthened its governance by joining the Linux Foundation, which added legitimacy and neutrality. Reddit
2. Why Was OpenSearch Created?
The motivations behind OpenSearch were a mix of technical, business, and community-driven factors:
- Preserving Open Source
OpenSearch was created to maintain a truly open-source search and analytics engine under Apache 2.0 — countering Elastic’s move to more restrictive licenses. - Avoiding Cloud Vendor Lock-in
AWS and others were concerned by Elastic’s license change because it seemed tailored to restrict cloud providers from offering Elasticsearch as a managed service. By forking, AWS could continue offering a fully open version without those limitations. theregister.com - Community Control & Innovation
Through the OpenSearch Foundation, the project aimed to democratize the development of search technology, inviting community contributions and making governance more transparent. - Compatibility & Stability
The fork was initially based on Elasticsearch 7.10.2, a familiar and stable codebase. Wikipedia This allowed users of older ES versions to migrate more easily without losing open-source guarantees.
3. What OpenSearch Offers Today
OpenSearch has matured quickly since its fork, offering a rich suite of capabilities. Some of the highlights of what OpenSearch (and its associated tools) provides today:
- OpenSearch Engine: Based on Lucene, supports full-text search, complex queries, aggregations, and analytics. Wikipedia
- OpenSearch Dashboards: A UI layer (forked from Kibana) to visualize data, build dashboards, and explore logs. Wikipedia
- Plugins and Extensions: Security (RBAC, encryption), alerting, SQL support, machine learning, Observability analytics, and more. docs.opensearch.org+1
- Data Ingestion Tools: Components like Data Prepper help gather, transform, and feed data into OpenSearch. OpenSearch
- Cross-platform Support: OpenSearch runs on Linux, Windows, Docker, and supports clustering, high availability, and scalability. docs.opensearch.org
- Community-driven Roadmap: The project roadmap is publicly available; development is driven by both AWS and the broader community. docs.opensearch.org
4. What About Elasticsearch? Is There a “Free” Version Still?
The situation around Elasticsearch’s “free” or open version is more nuanced now than it used to be.
4.1 Licensing History & Changes
- As mentioned, in 2021 Elastic switched from Apache 2.0 to dual licensing under SSPL and Elastic License v2. Elastic+1
- Under this model, while the code remains “source-available,” it is not considered fully open-source by some, because SSPL is not OSI-approved. theregister.com
- In 2024, Elastic introduced a third license option: AGPLv3, which is OSI-approved. ir.elastic.co
4.2 What the “Free” Tier Means Now
- Elastic continues to offer a free “Basic” tier for Elasticsearch. Elastic
- According to their FAQ, users of the default distribution (running on-premises) are largely unaffected by licensing changes: many of the “free” features remain free under the Elastic License v2 or SSPL. Elastic
- That said, the nature of “free” is different now: while the source code is visible, the licensing imposes restrictions, especially around providing Elasticsearch as a service. theregister.com+1
- The addition of AGPLv3 in 2024 is a gesture toward open-source legitimacy, but the dual/triple licensing model means that not all parts are under the same license. ir.elastic.co+1
5. Why the Fork Matters — Implications & Impact
- Competition & Innovation: OpenSearch provides a viable open-source alternative to Elasticsearch, which fosters more competition, innovation, and choice.
- Cloud Strategy: By offering a fully open search engine, AWS can continue to provide managed services without the legal and licensing constraints imposed by Elastic’s model.
- Community Trust: For many in the open-source community, the Apache 2.0 license of OpenSearch is more trustworthy than SSPL, because it aligns with traditional permissive open-source licenses.
- Sustainability: With its own foundation and growing contributor base, OpenSearch is positioned to be sustainable and vendor-neutral in the long run.
Conclusion
OpenSearch is not a successor (yeah, that hook was just to reel you in), is not just a technical fork; it’s a strategic and philosophical response to changes in the open-source licensing landscape. Born out of a desire to retain a truly open, permissively licensed search engine, it has grown rapidly under the stewardship of a foundation and a broad community. Today, it offers powerful search, analytics, security, and observability capabilities — all under the Apache 2.0 license.
On the other hand, Elasticsearch continues to be widely used and developed, but its licensing model has become more complex. While there is still a free Basic tier, the shift to SSPL/Elastic License and the recent addition of AGPL means its “free” status carries different conditions than before.
For organizations and developers who care deeply about open source, transparency, and long-term independence, OpenSearch represents an increasingly compelling option — and its rise is reshaping how we think about open search architecture.