For every service, the RED method tracks the blank of requests, the blank and their blank. For resources like CPUs, disks and pools, the USE method tracks utilization, saturation and errors.

Answer

For every service, the RED method tracks the rate of requests, the errors and their duration. For resources like CPUs, disks and pools, the USE method tracks utilization, saturation and errors.

Also accepted: request rate, requests for rate; error rate, failures for errors; latency for duration.

Why

RED describes what users feel; USE explains why: a disk at 100% utilization with a growing queue (saturation) is the cause behind rising durations. The two together cover most dashboards.

Review this in your daily deck → All cards in Observability