API response time test
One measurement is an anecdote. This sends three requests and shows you the spread.
Three GET requests, sent at the same time, from the cloud.
Reading the numbers
- Min is roughly your endpoint at its best: warm connection, warm cache.
- Max is what some of your users actually get. It is the number worth watching.
- The spread between them says more than the average: a tight one means predictable, a wide one means something is queuing.
- These are measured from our cloud, including DNS and TLS. A browser in another country will see more.
Latency drifts slowly, which is why a single test tells you little. Monitoring it every 5 minutes gives you the average and the worst case over a week, in one email.
Related
HTTP status code checker — what it answers, and where it redirects.