Sao Paulo. How long a round trip from your machine to this region takes — and how much of that distance alone demands.
| Cloud | Vultr |
|---|---|
| Region code | sao-br |
| Region name | Sao Paulo |
| Datacentre city | Sao Paulo, BR |
| Coordinates | -23.55, -46.63 |
| Published IPv4 prefixes | 2 |
| Probe endpoint | https://sao-br-ping.vultr.com/vultr.com.100MB.bin |
| Probe style | a CORS request whose reply the page can read |
| Verification | Resolved to an address inside this region’s own published range, TCP 443 open. Checked 2026-08-21. |
Coordinates are the real datacentre city rather than the marketing region name, because they feed the distance maths below.
Light in fibre covers about 100 km per millisecond of round trip. These are the great-circle distances to Sao Paulo and the latency no network can beat — not a measurement, a limit. What you actually measure is this plus routing.
| From | Distance | Floor |
|---|---|---|
| New York | 7,685 km | 77 ms |
| Los Angeles | 9,906 km | 99 ms |
| São Paulo | 0 km | under 0.5 ms |
| London | 9,498 km | 95 ms |
| Frankfurt | 9,829 km | 98 ms |
| Johannesburg | 7,430 km | 74 ms |
| Dubai | 12,226 km | 122 ms |
| Mumbai | 13,774 km | 138 ms |
| Singapore | 15,987 km | 160 ms |
| Tokyo | 18,537 km | 185 ms |
| Sydney | 13,357 km | 134 ms |
Within 75 km of Sao Paulo — near enough that distance sets no meaningful floor between them. When several sit in one city, the fastest is the city’s real floor; a slower neighbour is usually the endpoint’s own server time.
These figures are geometry. Your own round trip to sao-br depends on your connection and the route it takes — open the instrument, pick Vultr, and sweep. Every region is timed from your browser against the endpoint listed above, and the result is split into the part physics demands and the part the network added.
See every Vultr region on the Vultr latency page.