How to Read Layer2C Assessments
Layer2C assessments are opinionated architectural analyses — not certifications, lab validations, or paid rankings. This page explains what the assessments evaluate, how they are scored, and how to interpret them correctly.
What the Assessments Evaluate
Each assessment analyzes a vendor's AI infrastructure offering across eight layers of the 4+1 AI Infrastructure Model:
| Layer | What Is Being Evaluated |
|---|---|
| Layer 0 — Compute & Network Fabric | Who owns the GPU/accelerator silicon, networking fabric, and interconnects? Does the vendor design its own silicon or brand a partner's? |
| Layer 1A — Data Storage & Governance | Who controls the data substrate, metadata layer, and governance catalog? Can the Layer 2C reasoning plane query it programmatically? |
| Layer 1B — Context Management & Retrieval | What retrieval infrastructure does the vendor own versus delegate? Who controls retrieval quality and observability? |
| Layer 1C — Data Movement & Pipelines | Who orchestrates data movement, KV cache tiering, and pipeline automation? Is lineage captured? |
| Layer 2A — Infrastructure Orchestration | Who controls GPU scheduling, workload placement, and the Kubernetes control plane? Is GPU-aware scheduling vendor-owned? |
| Layer 2B — Application Runtime & Execution | Who owns the model serving layer, inference engine, agent runtime, and guardrails? |
| Layer 2C — Agentic Infrastructure | Does the vendor have a policy-driven reasoning plane — one that makes placement decisions, not just routing decisions? Can it audit agent behavior at the infrastructure level? |
| Layer 3 (+1) — AI Application Layer | What AI applications and ISV ecosystem does the vendor enable? Is the application layer open or captive? |
The Decision Authority Placement Model (DAPM)
Every assessed component receives a DAPM classification. The classification answers one question: if the enterprise wants to change, audit, or override this capability, who has the authority to do it?
| Classification | Meaning | Practical implication |
|---|---|---|
| Retained | Enterprise owns and controls this capability | Can be audited, modified, or replaced without vendor involvement |
| Delegated | A substitutable partner provides this capability | Enterprise retains swap authority — authority is lent, not surrendered |
| Ceded | Vendor controls this capability | Changing it requires changing the vendor or accepting the vendor's roadmap |
| Absent | No capability exists at this layer | Enterprise must build, buy elsewhere, or operate without it |
DAPM is not a quality score. A Ceded capability can be excellent. The classification describes the governance relationship, not the capability's fitness for purpose.
Layer Status Indicators
Each layer also receives a status indicator summarizing the overall assessment at that layer:
| Indicator | Meaning |
|---|---|
| ● Strength | The vendor has genuine, owned capability at this layer |
| ◑ Moderate / Delegated | Partial coverage; meaningful capability exists but is shared with or depends on partners |
| ○ Gap | The vendor lacks expected coverage at this layer; notable structural weakness |
| ✕ Absent | No capability at this layer by design or omission |
| ◇ Partner Ecosystem | Coverage is delivered entirely through partners; vendor provides integration but not the capability itself |
What Assessments Do Not Claim
- They are not lab validations or benchmark results
- They are not exhaustive product reviews covering every feature
- They are not procurement recommendations — they do not tell you which vendor to buy
- They are not certifications — no vendor pays to be assessed or approved
- They are not permanent — architecture evolves; assessments reflect a point in time
Evidence Sources
Assessments draw from publicly available materials. Each assessment lists its primary sources, which may include:
- Vendor conference presentations and technical sessions (GTC, re:Invent, Google Cloud Next, HPE Discover, etc.)
- Official product documentation, white papers, and architecture guides
- Press releases and earnings call transcripts
- Analyst coverage and independent technical reporting
- Vendor briefings and technical conversations
- The CTO Advisor's independent architectural judgment
Where a vendor has reviewed an assessment prior to publication, this is noted on the assessment page. Vendor review does not imply vendor approval or endorsement — factual corrections are incorporated; editorial positions are retained.
The Borrowed Judgment Test
Each layer includes a "Borrowed Judgment" assessment. This answers: how much of the vendor's authority at this layer depends on a partner's roadmap, not their own?
Borrowed judgment is not inherently negative — every vendor has it at some layers. The question is whether the enterprise recognizes it. A vendor that says "we provide full GPU orchestration" while actually depending on NVIDIA Run:ai is exercising borrowed judgment. The enterprise buys what it thinks is the vendor's capability but is actually exposed to NVIDIA's roadmap decisions.
High borrowed judgment at Layer 2C is the most consequential form — it means the enterprise's governance plane depends on a third party's architectural choices.
Commercial Disclosure
The CTO Advisor is an independent analyst firm. Vendors do not pay to be assessed on Layer2C. Where The CTO Advisor has a commercial relationship with an assessed vendor (advisory, consulting, sponsored content), this is disclosed on the relevant assessment page.
Assessments are not influenced by vendor relationships. If you believe an assessment contains a factual error, contact The CTO Advisor at thectoadvisor.com.