Synthesis

The evolving thesis. Current best understanding of this wiki’s topic, updated on every ingest.

Current thesis

This wiki tracks cloud-hosting providers. The corpus now spans three models of how projects get hosted — raw IaaS, managed PaaS, and edge compute — anchored at the top by a hyperscaler and organized along a pricing/acquisition-model spectrum:

The hyperscaler anchor. aws is the high/complex end the whole spectrum implicitly defines against: 200+ à-la-carte services you assemble and separately meter. Its revamped 2025 free tier is a credit funnel ($200 credits, 6-month auto-close, 30+ always-free services) — the same “free has a shape” mechanic as oracle-cloud-free but time-boxed rather than permanent. Most of the Tier-2 PaaS pitches (“no idle markup,” git-push) are really pitches against assembling raw AWS yourself, so AWS is the baseline that explains why managed PaaS exists.

Tier 1 — raw IaaS (you rent the box):

• hetzner-cloud — low-cost paid IaaS. Wins on price-per-resource; fixed monthly VPS, bundled traffic, no free tier. “Cheap and simple.”
• oracle-cloud-free — free-tier funnel. A permanent Always Free allowance + a $300 / 30-day trial, converting to Pay-As-You-Go. The quotas are now pinned down (Arm A1: 4 OCPU / 24 GB, 200 GB block, 2 Autonomous DBs, 10 TB/mo egress) — unusually generous — gated behind funnel mechanics: card required, one account/person, idle-30-day suspension.
• digitalocean — the IaaS↔PaaS midpoint. Droplets and App Platform in one vendor — proof the tiers are a continuum, not a wall.

Tier 2 — managed developer PaaS (you push code, they run it):

• fly-io — pure pay-as-you-go, no base fee, per-second compute, metered egress ($0.02/GB), 18+ edge regions; no free tier. Bills “resources pro-rated by time.”
• render — free-but-it-sleeps: 750 free instance-hours/mo, spins down after 15 min idle, free static sites, free Postgres that expires in 30 days; Heroku-lineage managed PaaS.
• railway — subscription + usage hybrid: $5/mo Hobby + per-second metering, “no overprovisioning, no idle markup.”
• vercel — the frontend-specialized PaaS (maker of Next.js): pure usage-metered Managed Infrastructure (1M free function invocations / 4 CPU-hrs on Hobby, then it stops) with the box abstracted away entirely — you ship a framework build, not a sized machine. (Cross-spoke seam: Vercel hosts static-site-wiki’s frameworks.)

Tier 3 — edge compute (no instance to size): cloudflare is a third model entirely — Workers/Pages/R2/Workers AI, global and per-request, with R2’s no egress fees the sharp foil to fly-io‘s metered egress. Workers AI also directly addresses the “where do AI stacks run?” question. The official cloudflare-workers docs (T1) show this tier is no longer just stateless functions at the edge: it carries a full data plane — Durable Objects, D1 (SQL), KV, R2, Vectorize — so an entire stateful app can run end-to-end on the edge, not just request handlers in front of an origin.

A new shape — stateful suspendable sandboxes. aws-lambda-microvms adds a point the spectrum didn’t have: Firecracker-based isolated VM sandboxes with a serverless model and state-preserving suspend/resume (up to 8h; suspended = no charge). It’s render‘s scale-to-zero without losing the running state, at VM isolation, billed per-second — “warm scale-to-zero.” Notably it’s pitched first at agent sandboxes (running LLM-generated code), so it doubles as a partial answer to the open question below (where AI-agent stacks run): the hyperscaler is now selling the agent-execution substrate as a primitive (cross-spoke seam to agentic-tooling-wiki’s harness sandboxes / Signadot ephemeral envs).

So the real axis is a cost-model spectrum: hyperscaler à-la-carte (aws) → fixed monthly VPS (hetzner-cloud) → permanent free tier + PAYG (oracle-cloud-free) → IaaS/PaaS midpoint (digitalocean) → edge per-request (cloudflare) → low monthly floor + per-second usage (railway) → pure pay-as-you-go (fly-io/vercel) → free-with-spin-down (render). Tier 1 sells capacity; Tier 2 sells a deploy workflow (git push, zero-downtime deploys, preview envs) and bills actual usage; the edge tier sells global per-request execution with no machine to size. A recurring catch shows up across the free tiers — Oracle’s idle-suspension, Render’s spin-down + Postgres-expiry, Fly’s metered egress vs Hetzner’s bundled traffic — “free” always has a shape.

Sourcing caveat: these remain mostly vendor/official surfaces (pricing & docs pages) — good for what’s offered. The first neutral third-party benchmark now lands too: vpsbenchmarks-docean-hetzner grades providers A–F on a percentile curve from a reproducible suite (web/sysbench/geekbench/fio/iperf3), so vendor claims can finally be checked against independent reliability/price-perf data. (An earlier Oracle signup-SPA stub was removed 2026-05-31; see log.md.)

What the hyperscaler actually sells

azure and google-cloud join aws, and with three examples the spoke can stop generalizing from one. The three descriptions converge on the same product, and it is not compute.

What a hyperscaler sells above a rented machine is an explicit failure-domain hierarchy: zone → region → geography, each defined, each contractual. Google states it most precisely — a zone “should be considered a single failure domain,” a region is “three or more zones housed in three or more physical data centers.” Microsoft sells the same shape plus curated region pairs for geo-replication that customers cannot choose. On top of that sits the piece that is genuinely hard to replicate: managed multi-region state (BigQuery, Cloud Storage, Firestore, Spanner) “designed to be able to function following the loss of a single region.”

That reframes the spoke’s central comparison. hetzner-cloud and the fixed-price VPS tier compete on compute per euro, and they are not selling against zones and residency boundaries at all — which is why the price gap can be as large as it is without either side being wrong. The hyperscaler’s premium buys geography and managed replication; if a workload does not need them, the premium buys nothing.

One drift worth tracking: Azure’s newer regions are not paired, using zones as the primary redundancy instead. Redundancy is moving down a level, from region-to-region to zone-to-zone.

Open questions

• How does Hetzner compare with Render/Railway/Fly on price/limits for small projects? Now mapped (above): a fixed-VPS vs. usage-metered-PaaS split. Still open: independent reliability / price-performance benchmarks (all current data is vendor-stated). Now opened (2026-06-12): vpsbenchmarks-docean-hetzner is the first neutral benchmark — it corroborates the vendor-built story (hetzner-cloud best price-per-resource; digitalocean weak on raw synthetic CPU/RAM, strong as a platform). Still want the PaaS tier (Fly/Render/Railway) measured independently — VPSBenchmarks covers raw VPS, not managed PaaS. Searched 2026-06-15 (quality cycle): no neutral source found. The PaaS-comparison space is entirely affiliate-monetized review/comparison content (ExpressTech, SaaSPricePulse, The Software Scout, DevToolReviews, etc.) — T3/T4 with referral incentives, not reproducible-methodology benchmarks. So this gap stays open by scarcity, not oversight: closing it needs a “VPSBenchmarks-for-PaaS” (a methodology-disclosed cold-start/throughput/price suite) that does not appear to exist publicly. Per the floor-raise rule, no source added rather than padding the spoke with more affiliate comparisons.
• Catches of “free tier” offerings; the actual Oracle Always Free quotas. Answered: Oracle quotas pinned (oracle-cloud-free); free-tier catches catalogued (idle-suspension, spin-down, DB-expiry, metered egress). Remaining nuance: real-world egress overage bills at scale.
• Which providers do the AI-agent/brain stacks actually run on, and why? Partially: render/railway/fly-io are the common managed targets (git-push, per-second, edge); still want a concrete stack→provider mapping. New angle (2026-06-23): the question is no longer just where agents are hosted but where agent-written code is safely executed — aws-lambda-microvms is AWS selling that sandbox substrate directly (isolated VM + suspend/resume), competing with E2B/Modal and the harness-internal sandboxes agentic-tooling-wiki tracks.
• What the hyperscalers actually charge. Half answered 2026-08-10 (quality cycle). Azure is now priced from first-party billing data (azure-vm-retail-prices, the spoke’s second T1): D2s v5 (2 vCPU / 8 GB, eastus) at $0.096/hr on demand, $519/yr reserved 1-year, $997 reserved 3-year, $0.020266/hr spot. google-cloud is still unpriced — JS-rendered pages, the legacy public pricelist.json is 404, and the Billing Catalog API needs a key.
• (new) Tier 1 vs Tier 2 crossover: at what scale does raw IaaS (hetzner-cloud) beat usage-metered PaaS on cost — i.e. where does the “no idle markup” promise stop paying off?

The purchase-model axis (2026-08-10). The spoke has been comparing tiers by architecture — raw IaaS, managed PaaS, edge. Azure’s own price list adds an axis the other tiers don’t have: the same machine costs $35.04, $21.63, $13.85 or $7.40 per vCPU-month depending purely on how you buy it (on demand, 1-year reserved, 3-year reserved, spot). A 2.5× spread with no technical difference, 5× with eviction risk. Two consequences. First, any comparison quoting “the hyperscaler price” without a commitment term is not comparing anything. Second, the crossover question above is under-specified: hetzner-cloud beats Azure on demand easily and beats 3-year reserved by much less, so the answer depends on a contract, not a workload. railway‘s metered $20.02/vCPU-month, re-verified the same day, sits between Azure’s 1-year and 3-year reserved rates.

Growth edges

Ranked; each names the kind of source that would close it (see ../QUALITY.md → Growth edges).

1. A neutral PaaS benchmark. The spoke’s oldest gap and the one it has already hunted and failed: vpsbenchmarks-docean-hetzner covers raw VPS, nothing covers the managed tier (fly-io/render/railway). Searched 2026-06-15 — the space is entirely affiliate-monetized comparison content. Open by scarcity, not oversight. — needs: a methodology-disclosed cold-start / throughput / price suite for managed PaaS (T1/T2); do not close it with another affiliate roundup.
2. Where the Tier-1/Tier-2 crossover actually falls. The “no idle markup” promise has to stop paying off at some scale, and no source here names it. — needs: a worked cost comparison at stated load, or a first-party pricing model detailed enough to compute one.
3. Real egress overage bills. Free-tier catches are catalogued; what egress actually costs at scale is the remaining nuance. — needs: a first-party billing doc (T3 acceptable — this is a vendor fact) or a published post-mortem with figures.
4. A concrete agent-stack → provider mapping. The spoke knows which managed targets are common and not what runs where, and the question has moved on to where agent-written code is safely executed (aws-lambda-microvms vs E2B/Modal). — needs: a T2 survey or a system report naming its substrate.

Coverage edges (added 2026-08-08, at the curator’s request for a wider backlog). These widen what the spoke covers instead of answering an open question above. Any of them closes with one ordinary solid source, and a vendor’s own pricing page counts (T3 is allowed where the edge names a vendor fact).

5. Two of the three hyperscalers are missing. CLOSED 2026-08-09 (research pass) — azure and google-cloud from their own region/reliability documentation (T3, as the edge allowed). Successor: the price half. Both pages describe failure-domain geography and neither carries a rate, so the spectrum’s expensive end is still anchored on aws alone for cost. — needs: the two vendors’ pricing pages for a comparable VM, or one three-way priced comparison.
6. The European mid-market beyond Hetzner. OVHcloud, Scaleway, Netcup and IONOS sell the same fixed-price VPS tier the spoke’s thesis is built on, and none of them is written up. — needs: first-party pricing/spec pages, one page per provider.
7. Object storage and the egress fight. The spoke prices compute and treats storage as a footnote, while the loudest price war of the last three years (S3 versus R2 versus B2, and who charges for egress) happens in storage. — needs: the three first-party pricing pages plus one worked cost comparison at a stated volume.
8. GPU hours as a price tier. Inference has to run somewhere, and the spoke has no page on a GPU host. — needs: first-party GPU-hour pricing from Lambda, RunPod, CoreWeave or Hetzner’s GPU line. Cross-spoke: the workload side sits in llm-inference-wiki.

Contradictions

None yet. (Watch vendor-stated pricing against future independent benchmarks.)

Index

Catalog of every wiki page, grouped by schema.org @type. Read this first when answering a query, then drill into the relevant pages. Updated on every ingest.

WebPage (sources)

Tier 1 — raw IaaS

• hetzner-cloud — Hetzner Cloud hosting (German, GDPR, low-cost) product page · source
• oracle-cloud-free — Oracle Cloud Free Tier: $300/30-day trial + permanent Always Free (quotas pinned) · source

Tier 2 — managed developer PaaS

• fly-io — Fly.io: per-second pay-as-you-go, edge regions, metered egress; no free tier · source
• render — Render: managed PaaS; free-but-spins-down (750 hrs/mo), free static sites, 30-day free Postgres · source
• railway — Railway: usage-metered PaaS; $5/mo Hobby + per-second billing, “no idle markup” · source

Spans tiers / edge

• digitalocean — developer cloud; Droplets (IaaS) + App Platform (PaaS); the Hetzner↔hyperscaler midpoint · source
• cloudflare — edge compute: Workers / Pages / R2 (no-egress) / Workers AI; a third model (global edge, per-request) · source
• cloudflare-workers — official Workers docs (T1): the edge platform’s full data plane (Durable Objects, D1, KV, R2, Hyperdrive, Vectorize, Workflows, Workers AI) · source

Hyperscaler / frontend PaaS

• aws — the hyperscaler anchor; 200+ à-la-carte services; revamped 2025 credit-funnel free tier ($200 / 6-month) · source
• azure — Microsoft’s hyperscaler: 70+ regions inside data-residency geographies, availability zones with isolated power/cooling/network, and region pairs you cannot choose — with newer regions dropping pairing in favour of zones. First-party reliability docs · source · T3 · learn.microsoft.com
• google-cloud — the third hyperscaler: regions of “three or more zones housed in three or more physical data centers,” a zone defined as a single failure domain, and managed multi-region services (BigQuery, Cloud Storage, Firestore, Spanner) built to survive losing a region · source · T3 · cloud.google.com
• aws-lambda-microvms — AWS Firecracker-based isolated VM sandboxes; serverless + state-preserving suspend/resume (8h, suspended=free); new cost-model shape (per-second + warm scale-to-zero); pitched at agent/LLM-code sandboxes · source · T3 · aws.amazon.com
• vercel — frontend-specialized usage-metered PaaS (maker of Next.js); Hobby free allowances, box abstracted away · source

Pricing data (first-party, machine-readable)

• azure-vm-retail-prices — Azure Retail Prices API: unauthenticated JSON price feed; D2s v5 (2 vCPU/8 GB, eastus) at $0.096/hr on demand, $519/yr 1-yr reserved, $997 3-yr, $0.020266/hr spot — a 2.5× purchase-model spread on identical hardware · source · T1 · prices.azure.com

Independent benchmarks (third-party)

• vpsbenchmarks-docean-hetzner — VPSBenchmarks: neutral A–F price/perf grading; Hetzner beats DigitalOcean on value; first non-vendor source · source

Synthesis

• synthesis — the evolving thesis (open questions + flagged contradictions)