> infrastructure
What we run for ourselves
We run our own fleet: 132 nodes across 12 countries, 99.982% uptime over the last 90 days. This isn't the product. It's the training ground. Every dashboard, runbook and on-call rotation you'd get from us is battle-tested on infra we pay for ourselves.
> fleet by the numbers
Current state. Refreshed from the registry, not from marketing claims.
| Metric | Value | Window |
|---|---|---|
| Active nodes | 132 | current |
| Countries | 12 | current |
| Fleet uptime | 99.982% | trailing 90 days |
| Baseline bandwidth per node | 100 Mbps | contractual minimum |
| Base OS | Ubuntu LTS | 22.04 / 24.04 |
| Days since last Sev-1 | 19 | at render time |
> fleet geography
12 countries across Europe, the Americas and APAC. The distribution is intentional: regulatory diversity, network-route diversity, and time-zone coverage for follow-the-sun on-call.
> node types
Not one type of workload. Each type has its own alert profile, runbooks and escalation order.
| Type | Count | Distribution | Profile |
|---|---|---|---|
| Validator | 38 | CA / DE / US / FI / UK | signing-critical, slashing-aware alerts, HSM/KMS |
| RPC | 27 | DE / US / BG / CZ | latency-critical, p95 SLO, caching |
| Archive | 14 | DE / CA / US | disk-heavy, multi-TB storage, slow reindex |
| GPU / inference | 22 | US / DE / AU | A100 / H100, vLLM / Triton, OOM watchdog |
| Storage / DePIN | 23 | RU / BR / BE / AT | Filecoin / Akash, payout reconciliation |
| Sequencer / prover | 8 | US / DE | deadline-aware, retry queues |
> public node registry
Live spreadsheet. Country, type, count. Refreshed regularly. What we do not publish: hostnames, IPs, client assignments, validator addresses. OpSec stays intact.
> why we keep our own fleet
You can read an SRE book and draw a dashboard. You can sign an SLA on paper. But the only way to know a tool works at 3 AM on day 90 is to run it at 3 AM on day 90. On your own dime.
Every runbook you'd get from us has been through our own fleet first. Every alert is tuned not to wake an engineer for nothing. Every dashboard shows what people actually look at during an incident, not what looks clean in a screenshot.
That's operator discipline. Our fleet is the training ground where we sharpen it before we bring it to clients.
> the same engineers run client infra
Same team, same runbooks, same on-call. Just under your workload.