Decision library

Workload placement research, grouped by the decision that triggered it.

RunPlacement pages are built for quick scanning: short answers, rough math, tradeoffs, sources, and a decision rule. Start with the confusion closest to the workload.

Decision pages

Start with the durable decision library.

Admin-approved drafts can still publish here, but the main problem pages live at the decision index.

GPU pricingH100 Quote Checklist: What to Ask Before Choosing GPU CloudCommercial investigation

An H100 quote is worth comparing only after the provider exposes the GPU shape, minimum rental window, storage, data transfer, capacity model, retry risk, and support terms.

AWS bill shockAWS NAT Gateway Bill Shock: What to Check FirstProblem diagnosis

NAT Gateway bill shock usually means private subnet traffic is taking an expensive path. Start by finding which workload, route table, availability zone, or transfer pattern created the processed-data spike.

GPU pricingGPU Cloud Idle Cost: How to Price Wasted Accelerator TimeCost estimation

GPU cloud idle cost is the gap between paid accelerator time and useful workload progress. It matters most for training retries, batch queues, and inference fleets with low baseline utilization.

Cloud migrationCloud Egress and Exit Cost: What to Price Before MovingMigration planning

Cloud egress is only one part of exit cost. A serious migration estimate also prices data export, recurring transfer, storage retrieval, rewrites, testing, downtime, rollback, and new operations.

Cloud migrationBare Metal vs Cloud Break-Even: When Dedicated Servers WinCommercial comparison

Bare metal can win when a workload is steady, portable, highly utilized, and operationally owned. Cloud usually wins when flexibility, managed services, or variable demand matter more than unit cost.

GPU pricingRunPod vs Lambda GPU Cloud: How to Compare the FitProvider comparison

RunPod vs Lambda is less about one universal winner and more about workload fit. Compare GPU availability, storage behavior, operational model, support needs, and total job cost for your actual workload.

Start with GPU pricingH100 quotes, hidden GPU fees, provider comparisons, and utilization. Start with AWS bill shockNAT, data transfer, CloudWatch, S3, and AWS-vs-smaller-cloud decisions. Start with cloud migrationAWS exit, bare metal, portability, and when staying put is smarter. Use the resource libraryChecklists and worksheets designed to be shared and reused.

AI inference cost

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

AI inference costGPU Inference Cost Breakdown: The Numbers To Estimate Firstcost_breakdown

A practical breakdown of GPU inference cost drivers, including useful GPU-hours, batching, idle time, traffic shape, storage, and data movement.

AI inference costRunPod vs Lambda vs AWS: Which Fits GPU Inference?comparison

Compare RunPod, Lambda, and AWS for GPU inference by cost sensitivity, data gravity, reliability, operations, and production requirements.

AI inference costA100 vs H100: When the Cheaper GPU Is the Better Placementdecision

A practical decision page for choosing A100 or H100 based on workload shape, memory, throughput, price, and availability.

AI inference costH100 On-Demand vs Reserved Capacity vs Spot: Which Should You Use?decision

A decision page for choosing between on-demand H100, reserved GPU capacity, and spot or marketplace GPUs.

AI inference costAWS vs Specialized GPU Cloud for H100 Inferencecomparison

A practical decision page for comparing AWS H100 capacity against specialized GPU clouds for inference workloads.

AWS bill shock

Start here when the bill jumped and the expensive line item is not obvious.

AWS bill shockAWS vs Smaller Cloud For Simple Workloads: When Default Cloud Is Too Muchcomparison

A practical decision page for deciding when a simple workload should stay on AWS or move to a smaller cloud, managed platform, or bare metal.

AWS bill shockS3 Cost Surprise: Storage Is Only Part Of The AWS Billcost_breakdown

A practical S3 cost breakdown covering storage, requests, retrieval, replication, lifecycle rules, and data transfer surprises.

AWS bill shockCloudWatch Cost Surprise: Logs, Metrics, And The Observability Taxcost_breakdown

A practical AWS CloudWatch cost breakdown for logs, metrics, retention, dashboards, and workload observability tradeoffs.

AWS bill shockAWS Data Transfer Cost Confusion: Egress, Cross-AZ, And Region Mistakescost_breakdown

A practical page for understanding AWS data transfer cost surprises across egress, cross-AZ traffic, regions, and workload placement.

AWS bill shockAWS NAT Gateway Surprise Bills: When Private Subnet Traffic Gets Expensivecost_breakdown

A practical decision page for understanding AWS NAT Gateway cost surprises and when architecture, endpoints, or placement may need review.

AWS bill shockWhy Is My AWS Bill So High? The Usual Places To Look Firstcost_breakdown

A practical AWS bill shock checklist for finding common cost drivers before moving workloads or blaming EC2.

Provider comparisons

When the hard part is choosing between provider categories, not reading another pricing page.

Provider comparisonsCloud GPU Quote Comparison: The Questions To Ask Every Providercost_breakdown

A practical checklist for comparing cloud GPU quotes across hourly rate, billing unit, storage, bandwidth, availability, support, and commitments.

Provider comparisonsCheapest H100 Cloud: Why The Lowest Price Can Be The Wrong Answerdecision

A practical decision page explaining why the cheapest H100 cloud listing may not be the cheapest workload placement.

Provider comparisonsVast.ai vs Managed GPU Cloud: When Marketplace Pricing Is Worth Itcomparison

A practical decision page for comparing Vast.ai marketplace GPU pricing with managed GPU clouds for experiments, inference, and training.

Provider comparisonsH100 Cloud Pricing Comparison: What To Compare Before The Hourly Ratecomparison

A practical H100 cloud pricing comparison checklist focused on useful GPU-hours, availability, storage, bandwidth, and operational tradeoffs.

Provider comparisonsRunPod vs Lambda vs Vast.ai: Which GPU Cloud Fits Your Workload?comparison

Compare RunPod, Lambda, and Vast.ai by workload shape, reliability needs, pricing model, and operational tolerance.

Provider comparisonsHow to Systematically Compare Cloud GPU Prices Across 20+ Providerscomparison

The real approach to comparing GPU prices on AWS, Google, Oracle, and 20+ providers—when spot/on-demand, regions, and volatility can drive 2x–8x price swings monthly. Shortcuts, tradeoffs, and decision tools.

Capacity decisions

Commitment, reservation, on-demand, and utilization tradeoffs.

Capacity decisionsGPU Utilization Break-Even: When A Cheap GPU Cloud Actually Saves Moneycost_breakdown

A practical GPU utilization break-even page for deciding when lower hourly rates outweigh idle time, retries, and operational overhead.

Capacity decisionsGPU Training Cost Breakdown: Before You Rent The Biggest GPUcost_breakdown

A practical breakdown of GPU training cost drivers, including runtime, checkpointing, failed runs, storage, data movement, and capacity planning.

Cost breakdowns

Line-item checklists for finding what the hourly rate leaves out.

Cost breakdownsGPU Cloud Hidden Fees: The Costs Missing From The Hourly GPU Ratecost_breakdown

A checklist of GPU cloud costs that are easy to miss, including storage, bandwidth, idle time, retries, support, and commitment waste.

Cost breakdownsGPU Cloud Pricing Checklist: What the Hourly Rate Leaves Outcost_breakdown

A checklist for comparing GPU cloud quotes beyond the hourly GPU price, including storage, bandwidth, idle time, availability, and ops.

Cloud migration

AWS exit and workload portability decisions.

Cloud migrationCloud Exit Cost Checklist: What To Price Before You Leave AWScost_breakdown

A practical checklist for cloud exit costs, including data transfer, rewrites, managed service replacement, downtime risk, and operations.

Cloud migrationAWS vs Bare Metal: When Owning The Machine Makes Sensecomparison

A practical comparison of AWS and bare metal for steady workloads, predictable utilization, operations, and cost control.

Cloud migrationWhen Not To Leave AWS Even If The Bill Looks Highdecision

A practical decision page for knowing when a high AWS bill should be optimized inside AWS instead of triggering a migration.

Cloud migrationShould You Move From AWS To A Cheaper Cloud?decision

A practical decision page for deciding whether AWS savings justify migration work, data movement, and operational risk.

Resources

Linkable assets

Checklists and worksheets built for practical sharing, not promotional posting.

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 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.

AWS bill shockAWS Bill Shock Evidence ChecklistResearch checklist / 4 sections / source-linked

A source-backed checklist for collecting AWS Cost Explorer, NAT Gateway, transfer, CloudWatch, storage, and routing evidence before changing architecture.

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.

Cloud migrationCloud Exit Assumptions IndexResearch index / 4 sections / source-linked

A source-backed index of the assumptions to collect before estimating cloud exit payback, partial migration, or workload re-placement.

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.

Workload placementWorkload Placement Assumptions IndexResearch index / 4 sections / source-linked

A source-backed index of the assumptions to collect before choosing cloud, GPU cloud, bare metal, managed platform, or hybrid placement.

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.

AI inference costAI Inference Cost Assumptions IndexResearch index / 4 sections / source-linked

A source-backed index of the workload assumptions to collect before estimating API, managed inference, batch, GPU cloud, or self-hosted GPU cost.

AI inference costProvider Pricing Page Field AuditResearch audit / 4 sections / source-linked

A provider-neutral audit of the fields to verify on official pricing and deployment pages before comparing AI inference serving options.

AI inference costRealtime vs Batch Inference Cost Research GuideResearch guide / 7 sections / source-linked

A source-backed guide to deciding when realtime, asynchronous, batch, or hybrid inference changes effective AI serving cost.