Product comparisons

Decided Designed

Consolidated index of every product/architecture comparison the GPUaaS team has done. Many of these were originally embedded inside other docs as the comparisons turned into requirements.

Comparisons drive requirements. A capability we choose to build, a vendor we choose to defer, a competitor whose feature set we choose to track — every decision leaves evidence somewhere in doc/. This section pulls those scattered evidences into one navigable surface.

Taxonomy

flowchart TB
    classDef ext fill:#e3f2fd,stroke:#1565c0
    classDef bvb fill:#fff3e0,stroke:#e65100
    classDef par fill:#e8f5e9,stroke:#2e7d32
    classDef gap fill:#fff3cd,stroke:#332701
    classDef rev fill:#f8d7da,stroke:#42101e

    A[Comparisons]
    A --> B[A. External clouds / products]:::ext
    A --> C[B. Build-vs-buy &<br/>option evaluations]:::bvb
    A --> D[C. Internal evolution<br/>parity audits]:::par
    A --> E[D. Point-in-time<br/>gap analyses]:::gap
    A --> F[E. External review]:::rev

    B --> B1[AWS / GCP / Azure / Nebius<br/>tenancy hierarchy]
    B --> B2[GPU cloud landscape<br/>CoreWeave / Lambda / RunPod /<br/>DGX Cloud / hyperscaler HPC]
    B --> B3[Armada Bridge<br/>gap matrix]

    C --> C1[Kubernetes platform options<br/>self-managed / Rancher / Kamaji]
    C --> C2[WEKA storage<br/>capability assessment]
    C --> C3[Slurm UI options]
    C --> C4[Embedded UI gateway<br/>iframe vs subdomain vs link-out]
    C --> C5[H200 MAAS<br/>bundle fit analysis]
    C --> C6[OpenClaw integration]

    D --> D1[Prototype → Production<br/>UX parity report]
    D --> D2[V3 → V1<br/>workflow parity audit]
    D --> D3[V3 mock → V3 production<br/>data parity]
    D --> D4[Prototype gap analysis<br/>historical baseline]

    E --> E1[Allocation experience gaps]
    E --> E2[App platform gap tracker]
    E --> E3[Clustered app gap table]
    E --> E4[Slurm product workflow gaps]
    E --> E5[Login UX & IdP gap]

    F --> F1[External architectural<br/>& product review 2026-04]

Pages

Category	Page	Coverage
External	External clouds & products	AWS/GCP/Azure/Nebius hierarchy, GPU cloud landscape, Armada Bridge gap matrix
Build-vs-buy	Build-vs-buy evaluations	Kubernetes options, WEKA assessment, Slurm UI options, Embedded UI gateway, H200 MAAS fit, OpenClaw integration
Parity audits	Internal parity audits	Prototype→Production, V3→V1 workflow, V3 mock→production data
Gap analyses	Point-in-time gap analyses	Allocation gaps, app platform gaps, Slurm workflow gaps, login UX gap, clustered app gap
External review	External review (2026-04)	Independent 5-layer audit with letter grades

Why this surface exists

flowchart LR
    A[Comparison done] --> B[Embedded in<br/>some other doc]
    B --> C[Becomes a requirement<br/>or design decision]
    C --> D[Implemented or<br/>tracked in backlog]
    A -.surfaced here.-> H[Searchable<br/>cross-comparable<br/>one place]
    D -.linked here.-> H

Reviewers asking "what have you compared yourselves to?" or "where did this requirement come from?" can answer both questions from one section instead of grep'ing the source docs.

Reading order by interest

If you care about…	Start with
Whether GPUaaS is differentiated	External clouds & products
Why specific build/buy choices were made	Build-vs-buy evaluations
How the product has evolved internally	Internal parity audits
What's currently weak	Point-in-time gap analyses → External review