New: one-click Cloud Apps

Network & Datacenter

Where your apps actually run

Two US locations, chosen on purpose -- Detroit and Los Angeles -- with the big networks a fraction of a millisecond away on both coasts, and you pick which at checkout. Our own platform, measurements we took ourselves, and addresses you can test before you spend anything.

The network

Plain facts, stated once

Where the machines sit, what we run on them, and the numbers we measured ourselves. Nothing on this page is copied from a datacenter brochure.

Two US locations: Detroit and Los Angeles

Every Cloud App runs in one of two US facilities -- Detroit, Michigan or Los Angeles, California -- and you choose which at checkout. Two coasts, stated plainly, rather than a map of cities we keep no equipment in. Central for the middle and the east, coastal for the west.

The big networks are in the building

From Detroit, Cloudflare answers our pings in about a fifth of a millisecond; from Los Angeles, Google answers in about a third of one. On both coasts the networks your visitors ride are, for practical purposes, in the same room as the machines.

Our platform, not a reseller login

The panel, the API, the deployment recipes and the app catalog are ours, built and run by us. We lease server capacity the way nearly everyone in this industry does -- what you pay us for is everything that runs on it.

Your app and its database share a machine

A deployed stack keeps its app and its database on the same host, talking over a private network that never leaves the box. Queries take microseconds, not a round trip between buildings -- and that is measured, not estimated, on either coast.

IPv6 on every app, from day one

Every app gets a real public IPv6 address alongside IPv4, without asking and without an add-on -- in both locations. A surprising share of hosts still treat this as a paid feature; we treat it as table stakes.

Your address never moves

Apps are reached by hostname -- yours under our domain, or a domain you point at us. If we change hardware or providers underneath, your hostname and your custom domain keep working; DNS follows us, not the other way round. Hardware moves should be our problem, not yours.

Measured, not claimed

Latency from the machines your apps run on

Taken with ping from each app host itself, August 2026. Your own result will differ -- that is the point of publishing addresses you can test.

Detroit -> Cloudflare (1.1.1.1)

0.22 ms

they answer from inside the building, effectively

Detroit -> Google (8.8.8.8)

6.5 ms

public resolver, Chicago region

Los Angeles -> Google (8.8.8.8)

0.35 ms

they answer from inside the building, effectively

Los Angeles -> Cloudflare (1.1.1.1)

1.0 ms

public resolver, about a hop away

App -> its database

< 0.1 ms

same machine, private bridge -- either coast

Test it yourself

Addresses you can ping

Run ping, mtr or traceroute against these from wherever your users are. We would rather you measured than took our word for it.

Detroit app host

IPv4 -- ping, mtr or traceroute

66.163.115.158

Detroit app host

IPv6, if you want to test v6 reachability specifically

2604:86c0:b001:2::2

Los Angeles app host

IPv4 -- ping, mtr or traceroute

167.17.68.34

Los Angeles app host

IPv6, if you want to test v6 reachability specifically

2604:86c0:2001:20::2

These answer ICMP so you can measure latency and routing. For a throughput test, ask support and we will set one up rather than leave a permanent download target on a machine serving customers.

Questions

About the locations

Ready to deploy?

Deploy an app from the catalog in minutes. No contracts, no setup fees.