Measured latency to Luno around the world
REST ticker round-trip from eight AWS regions
Luno latency from eight AWS regions
| AWS region | Round-trip |
|---|---|
| London eu-west-1 | ~77 ms |
| Frankfurt eu-central-1 | ~102 ms |
| N. Virginia us-east-1 | ~275 ms |
| N. California us-west-1 | ~412 ms |
| Singapore ap-southeast-1 | ~609 ms |
| Hong Kong ap-east-1 | ~729 ms |
| Seoul ap-northeast-2 | ~884 ms |
| Tokyo ap-northeast-1 | ~1285 ms |
Indicative REST round-trips from our multi-region measurements — for comparing regions, not an SLA. WebSocket feeds are faster in absolute terms, but the regional proportions hold.
Best server location for trading on Luno
Host your trading server or bot in AWS London (eu-west-1): it reaches Luno in about 77 ms, fast enough that order placement and market data stay effectively real-time.
Avoid running latency-sensitive strategies against Luno from Tokyo or other distant regions. At ~1285 ms (17× the co-located figure) every price you see is stale before you can act on it, and that staleness turns into slippage on every market order.
Compare with other exchanges
Deribit server location
~10 ms
BitMEX server location
~26 ms
BEQUANT server location
~54 ms
NovaDAX server location
~58 msFrequently asked questions
Where are Luno's servers located?
Luno responds fastest from AWS London (eu-west-1) — about 77 ms round-trip — which places its API infrastructure in or near London.
What is the best server location for Luno?
AWS London (eu-west-1). A server there reaches Luno in ~77 ms, versus ~1285 ms from Tokyo — a 17× difference on every API call.
How much does server location change latency to Luno?
In our measurements: ~77 ms from London versus ~1285 ms from Tokyo — a 17× spread. Every order and every orderbook update pays this tax in both directions.