Skip to content
BENIBIX Learn

Throughput

Infrastructure🌐 NetworkingNetwork Performance

Also known as Requests per Second, RPS, Data Rate

How much work a system really gets done each second.

Think of it like

Think of it like the cars that really cross a bridge in an hour. It is not how many could.

Example

Lucy load tests her API. It handles 800 requests a second with quick replies, but past 1,000 a second the replies slow down and some fail.

Why it matters

It tells you how many users a system can serve at once. That decides how many servers you need.

Definition

Throughput is how much work a system really gets done in a set time. For a network it is the data that actually arrives each second. For a server it is often requests handled per second. It is always at or below the bandwidth or the most the system could do. Load tests measure it to find how much traffic a system can take before it slows down.

Common mistake

Looking only at throughput and not at delay. A system can handle lots of requests a second while each user still waits too long.

Key takeaways

  1. 1
    Work actually done. Not the limit.
  2. 2
    Per second. Data or requests.
  3. 3
    Watch latency too. Both tell the story.

Compare

Learn it properly