AI Potluck
Contribute
Back to Gap Map Infrastructure / ML orchestration

Armada

Armada
open source / Overall score: 1.8

Multi-cluster batch queuing system for high-throughput workloads on Kubernetes, built at G-Research for quantitative research and machine-learning jobs. Armada holds a queue in front of many Kubernetes clusters, applies per-queue fair share across them, and decides which cluster each job is sent to, so a fleet of clusters behaves as one pool.

Ships as container images and Helm charts with no package channel, so adoption bands on stars. Verified 2026-09-15 via the repository and its recursive tree.

Openness

5 high confidence
5.0
license
Apache-2.0(OSI)
source
public(armadaproject/armada is the whole system)
core-gated
ungated(a 2,416-entry recursive tree carries no path matching enterprise, ee, commercial or proprietary)

Apache-2.0 with no commercial edition. The 2,416-entry recursive tree carries one governing LICENSE and no enterprise, ee or commercial path.

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

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

Adoption

1 low confidence
1.0

633 GitHub stars. The project ships as container images and a Helm chart and publishes no package-registry artifact, so no usage-volume channel exists and the band rests on stars, which measure attention rather than use. The stars route caps at level 3 for that reason. It is the smallest record in this category by this measure, and the note says so rather than leaving the reader to infer it.

Capability

3 medium confidence
3.0

Armada decides which cluster a job goes to and in what order, which is band 3 in this category: it chooses the pool, and the chosen cluster's own scheduler chooses the machine. Multi-cluster reach is breadth rather than depth of the placement decision, so it does not lift the band.

Verified 2026-09-15