3 minutes readKey takeaways Header presets stop you re-entering auth and content-type headers on every test. Keep a preset per environment so a test cannot accidentally target the wrong one. Update the preset rather than individual tests when an API changes. We’ve implemented the Header Presets functionality in order to save time and reused same set of… Read more »
4 minutes readKey takeaways Decide what you are measuring and why before configuring anything. Choose the metrics up front, or you will report whatever looks best afterwards. Change one variable at a time if you want the comparison to mean anything. Below we listed what we consider 10 tips and tricks related to load and performance testing:… Read more »
< 1 minute readKey takeaways Basic Auth sends base64-encoded credentials in a header, so the load test just needs that header set. Use test credentials, never production ones, since headers end up in logs and reports. Confirm the endpoint returns 200 rather than 401 before starting the real run. In order to create load tests for URLs that… Read more »
4 minutes readKey takeaways Cloud generation removes the capacity limit that caps most local API load tests. Weight endpoints by real traffic rather than testing them evenly. Assert on payloads, because an API can return errors quickly and look fast. Alright, folks, let’s dive into the nitty-gritty of load testing RESTful APIs in the cloud, but let’s… Read more »
2 minutes readKey takeaways Scheduled API load tests catch regressions that only appear after a deploy. Run them off-peak so the test does not compete with real traffic. Alert on threshold breaches; nobody reads a scheduled report. Scheduling Load Tests feature is now available on LoadFocus.com. Basically you can create new load tests and schedule them to… Read more »
2 minutes readKey takeaways Latency varies enormously by region, so a single measurement point tells you very little. Measure from where your users are, then decide whether a CDN or an edge deployment is warranted. Compare regions over time; a one-off reading catches routing noise. During the development of your web application you will encounter the need… Read more »
4 minutes readKey takeaways UI-heavy apps spend their time rendering, which HTTP-level load tests never measure. Combine a large HTTP load with a small number of real browser sessions. Watch client-side metrics, since the server can be fast while the page is not. Create automated website tests for free with LoadFocus.com. Testing the UI (user interface) of web… Read more »
