Course contentsShow
← Software Engineering
Lesson 1239 of 1,93337. Software Architecture FundamentalsFree lesson

Performance: Response Time and Throughput

Measuring how quickly systems respond to requests and how many operations they can handle per unit time.

Performance: Response Time and Throughput

What You'll Learn

You'll learn the two fundamental measurements that define how well a system performs under load: how fast it responds and how much work it can handle.

The Core Measurements of Performance

When evaluating system performance, two metrics dominate the conversation: response time and throughput. These are the quantitative answers to "how fast?" and "how much?"

Response time measures the elapsed time between sending a request and receiving a complete response. It's what users feel directly—the delay between clicking a button and seeing results. Response time is typically measured in milliseconds or seconds and includes network latency, processing time, and any queuing delays.

Throughput measures how many operations a system can complete per unit of time. Common units include requests per second (RPS), transactions per minute (TPM), or messages per hour. Throughput tells you the system's capacity—its ability to handle volume.

Why Both Matter

Think of a highway: response time is like your individual travel time from entrance to exit, while throughput is the number of cars that can pass through per hour. A wide highway (high throughput) doesn't guarantee a fast trip (low response time) if there's congestion. Similarly, a fast empty road (low response time) might have low throughput if it's only one lane wide.

These metrics often trade off against each other. Pushing for maximum throughput usually increases response time due to queuing and resource contention. Optimizing for minimal response time might mean running below capacity, sacrificing potential throughput.

When measuring, you must specify conditions: "50ms response time at 1000 RPS" is meaningful; "fast performance" is not. Always measure both metrics together with their relationship clearly stated.

Key Takeaway: Response time measures how quickly individual requests complete, while throughput measures system capacity; both must be measured together under specific load conditions to meaningfully characterize performance.