Armada

license

Apache-2.0(OSI)

source

public(armadaproject/armada is the whole system)

core features withheld

no — no enterprise, ee, commercial or proprietary path in the tree and no paid tier named first-party

Armada is released under Apache-2.0, and its public repository builds and deploys the whole system, with no enterprise or commercial code path in the tree. It is a CNCF Sandbox project, but one company runs it: every current maintainer works at G-Research, which has used Armada in production since 2020, and the repository has no governance document. The project's assets are held by Armada a Series of LF Projects, LLC.

• https://api.github.com/repos/armadaproject/armada/git/trees/master?recursive=1 recorded 2026-09-15

  Full untruncated recursive tree of the default branch, 2,416 entries. One root LICENSE. No path matches enterprise, ee, commercial or proprietary. A tree lists paths; it is cited for that and for nothing about the vendor's offerings.

• https://armadaproject.io/ recorded 2026-10-08

  The project site's footer reads "Copyright © Armada a Series of LF Projects, LLC". The landing page names no pricing, support plan or commercial edition.

• https://github.com/armadaproject/armada recorded 2026-09-15

  Repository page for armadaproject/armada: Apache-2.0 license, public and unarchived, described as 'A multi-cluster batch queuing system for high-throughput workloads on Kubernetes'. Establishes the license and that the source is public.

• https://raw.githubusercontent.com/armadaproject/armada/master/LICENSE recorded 2026-10-08

  The repository's root LICENSE is the "Apache License", "Version 2.0, January 2004".

• https://raw.githubusercontent.com/armadaproject/armada/master/MAINTAINERS.md recorded 2026-10-08

  The maintainer list: all fifteen current maintainers, and all nine past ones, list their affiliation as G-Research.

• https://raw.githubusercontent.com/armadaproject/armada/master/README.md recorded 2026-10-08

  The README runs the system from this repository: "Armada components run as host processes built from source", and "you can install Armada manually by using the Helm charts defined in the `deployment` directory". It names no paid tier or separate edition, and says "Armada is a CNCF Sandbox project and is used in production at G-Research".

• https://www.cncf.io/projects/armada/ recorded 2026-09-16

  Armada's project page on cncf.io, establishing CNCF hosting rather than vendor ownership.

Armada ships as container images and a Helm chart rather than through a package registry, leaving GitHub stars as the only adoption signal. A star is not a use, so the signal shows attention rather than real usage.

• https://github.com/armadaproject/armada recorded 2026-09-15

  Repository page for armadaproject/armada showing 633 stars.

Armada decides which cluster a job goes to and in what order, then leaves pod placement to that cluster's own scheduler, the same shape of decision as Kubeflow Pipelines. Reaching across many clusters is breadth, not depth, so it does not add the finer-grained placement a deeper scheduler performs.

• https://github.com/armadaproject/armada recorded 2026-09-15

  Repository page and README for armadaproject/armada describing multi-cluster batch queuing and per-queue fair share.