All topicsTopic 08 / 24
Beginner 1 minute
Latency vs Throughput
Latency is how long one operation takes from request to response. Throughput is how much work a system completes in a period, such as requests per second. A system can have high throughput but slow individual responses, or low latency for one request but poor total capacity. Naming both metrics prevents teams from optimizing the wrong experience.
Key idea
Latency asks 'how long?' Throughput asks 'how much?'.
See it in one picture
Follow the arrowsLatencyHow long is one request?
ThroughputHow much work per second?
Real-world example
A video encoder may process thousands of files per hour while each file takes minutes. A checkout API needs low latency for a shopper, while its batch analytics pipeline may prioritize throughput.
Quick check