RED Method

The RED method is a monitoring checklist for services that handle requests, such as web servers and APIs. It tracks Rate (requests per second), Errors (how many of them fail) and Duration (how long they take, usually as percentiles).

Why it matters

Three numbers cover most user-visible symptoms without a wall of charts. Rate shows traffic shifts, errors show breakage, and duration shows slowness. Because they describe what users experience, they make good alert and dashboard material, and they apply the same way to every service, which helps on-call engineers who are unfamiliar with a given component. The method is associated with Tom Wilkie. It complements the USE method, which looks at resources rather than requests.

Example

For one endpoint, per minute:

rate      1,200 requests/min
errors    18 responses with 5xx (1.5%)
duration  p50 80 ms, p95 420 ms, p99 1.9 s

If errors jump to 12% while rate is flat, something broke. If duration climbs and rate is flat, look for a slow dependency or resource saturation next.

Common mistakes

• Reporting average duration, which hides the slow tail that real users hit.
• Counting only 5xx and missing failures that arrive as 200 with an error body.
• Applying RED to batch jobs and queues, which need different signals.

Related

• USE method
• Golden signals
• Error rate
• Percentile

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.