The control layer for infrastructure that has to stay up.
Nodeloom unifies service signals, routing decisions and maintenance windows in one deliberate console, so your team changes less and knows more.
In production since 2021. Frankfurt · Helsinki · Singapore
| Region | Status | Load | p95 |
|---|---|---|---|
| fra-1 | operational | 34% | 121 ms |
| ams-2 | operational | 29% | 118 ms |
| hel-1 | operational | 22% | 142 ms |
| tyo-1 | operational | 27% | 168 ms |
- 09:41:02 ingest fra-1 backlog cleared (0.4s)
One deliberate console.
Every part of Nodeloom exists to make operational change slower to start and faster to finish: reviewed, attributable and reversible.
Signals
Service indicators collected straight from the edge — checked, deduplicated and scored before they ever reach someone’s phone.
Routes
Routing changes ship as reviewable plans. Preview the blast radius, apply in one step, roll back without ceremony.
Maintenance
Windows that respect timezones and dependencies. Notifications queue themselves; every override is logged.
Telemetry ingest
Forty-one thousand events per second per region, retained ten days hot and a full year cold, queryable with one syntax.
Access control
Scoped roles and short-lived credentials. Every session is attributable to a person, never to a shared login.
Audit trail
Every decision, signed and searchable. Export the complete history as evidence the moment auditors ask.
Built for people who live in the terminal.
The console is a view over the same operations your scripts run. Everything is scriptable end to end, with a full API and CLI reference.
$ nodeloom status fra-1 operational 121 ms load 34% ams-2 operational 118 ms load 29% hel-1 operational 142 ms load 22% $ nodeloom routes plan --apply planning routes across 3 regions… ok applied 14 changes · 0 conflicts · 2.1s $ nodeloom maintenance schedule --window "sun 02:00 UTC" window accepted · notifications queued for 12 subscribers
The same operations, reproducible in CI. Exit codes you can build on.
Six regions, one standard.
Every region runs the same build on the same runbooks. Figures below are measured at the edge, not advertised.
| Region | Location | Status | Load | p95 | 90-day uptime |
|---|---|---|---|---|---|
| fra-1 | Frankfurt · EU | operational | 34% | 121 ms | 99.994% |
| ams-2 | Amsterdam · EU | operational | 29% | 118 ms | 99.997% |
| hel-1 | Helsinki · EU | operational | 22% | 142 ms | 99.991% |
| waw-3 | Warsaw · EU | operational | 31% | 126 ms | 99.998% |
| tyo-1 | Tokyo · APAC | operational | 27% | 168 ms | 99.990% |
| iad-2 | Virginia · US | operational | 38% | 134 ms | 99.996% |
Live figures refresh every 30 seconds. Historical incident data is public and exportable.
Run calmer operations.
Access is granted in small cohorts while the platform scales. Request a slot for your team, or read the documentation first; it is complete and public.