Resource library

Checklists and worksheets for messy compute decisions.

Use these when you need to turn a vague cloud, GPU, inference, or migration question into fields a team can compare.

Estimate onlyNo fake precision or invented pricing.
Source visibleProvider or official docs linked from the page.
Decision-readyBuilt to help compare categories, not sell one provider.
Infographic showing a cloud cost decision tree from confirming bill pain through classifying the driver, testing in-place fixes, and choosing a placement move.
A high cloud bill is not a migration plan. Move after diagnosis, not after surprise.

AI inference cost

AI inference cost assets

API, managed inference, self-hosted GPU, batch, realtime, and hybrid serving cost decisions.

AI inference costAI Inference Cost ChecklistChecklist / 8 sections / source-linked

A practical checklist for estimating AI inference cost across APIs, managed inference, self-hosted GPUs, batch jobs, realtime endpoints, and hybrid routing.

GPU pricing

GPU pricing assets

Quote review, useful GPU-hours, data movement, utilization, and provider tradeoffs.

GPU pricingGPU Cloud Quote ChecklistChecklist / 7 sections / source-linked

A practical checklist and visual worksheet for comparing GPU cloud quotes beyond the advertised hourly rate.

AWS bill shock

AWS bill shock assets

Line-item triage before assuming the whole cloud placement is wrong.

AWS bill shockAWS Bill Shock Triage ChecklistChecklist / 7 sections / source-linked

A first-pass checklist and visual triage flow for finding the AWS line items that usually make a bill jump.

Cloud migration

Cloud migration assets

Exit costs, payback windows, portability, and partial move decisions.

Cloud migrationCloud Exit Cost ChecklistChecklist / 7 sections / source-linked

A checklist and payback worksheet for pricing the real cost of leaving AWS, GCP, or Azure before migration starts.

Workload placement

Workload placement assets

The baseline worksheet for choosing a placement category before comparing vendors.

Workload placementWorkload Placement WorksheetChecklist / 7 sections / source-linked

A practical worksheet and decision map for deciding where a workload should run before provider choice hardens.

Frameworks

Definitions and formulas behind the worksheets

These framework pages support the worksheets with concise terms, examples, and decision tables.

Workload placementWorkload Placement Frameworkworkload placement

Choose workload placement by matching the workload's cost driver, data movement, performance needs, operational tolerance, and commitment horizon to the right infrastructure category.

 GPU pricingUseful GPU-Hour Frameworkuseful GPU-hour

Useful GPU-hour cost is the better comparison unit when GPU providers differ in utilization, queueing, reliability, storage behavior, or operational model.

 AWS bill shockCloud Bill Shock Taxonomycloud bill shock

Classify bill shock by driver class first: compute, network, storage, observability, managed services, support, marketplace, or commitment mismatch.

 Cloud migrationCloud Exit Payback Frameworkcloud exit payback

Cloud exit is financially serious only when steady savings repay migration work, data movement, service replacement, downtime risk, rollback planning, and new operations inside an acceptable window.

 Workload placementManaged Platform vs Infrastructure Control Frameworkinfrastructure control premium

A managed platform is better when it removes operational work the team should not own; direct infrastructure is better when control creates enough performance, cost, or compliance value to justify the burden.

 AI inference costAI Inference Cost ModelAI inference cost

AI inference cost should be compared as effective cost per successful request and monthly serving cost, not just token price or GPU hourly rate.

Worked examples

Useful GPU-hour math

Use this proof asset when a quote comparison needs concrete examples rather than definitions.

Worked examplesUseful GPU-Hour ExamplesHypothetical GPU cost scenarios

Five labeled examples showing how retries, idle time, data staging, and utilization can change effective GPU cost.