Performance
How a system behaves under load, usually described by latency, throughput, and resource usage rather than a single speed number.
Definition
Performance is how a system behaves as demand increases.
It is not one number.
It is a mix of latency, throughput, and resource usage.
How to talk about it in system design
- Latency: use percentiles like p99 latency.
- Throughput: use rates like QPS.
- Load shape: call out peak-to-average.
Common mistake
Optimizing averages and ignoring the tail.