Response timeレスポンスタイム is the time from when a user finishes entering a processing request until the system's first response begins to come back.
Finish entering request
Transmission and processing
First response starts
Timing stops when the first response starts to arrive, and this span is the response time.
3 / 3
Timing starts when input ends, so the time spent on the input itself is not included. Shorter values mean less waiting for the user, which matters most in online processingオンライン処理, where people work interactively with the screen.
AnalogyIt is like the gap between finishing your request at a service counter and the clerk's first words in reply. That is counted separately from the time needed to finish the whole procedure.
Exam points
ExamThroughputスループット is the amount of work finished within a given time, while response time レスポンスタイム is how long a single request takes to get a response.
Mix-upTurnaround timeターンアラウンドタイム runs from submitting a request until all results have been output, a longer span than response time, which stops when the response begins.
Response time written as 応答時間 is the kanji-based Japanese name for response time and refers to exactly the same interval.
Documents differ only in whether they use the katakana word or the kanji word, and neither the interval measured nor which way counts as better (shorter is better) changes. Whichever name appears, read it as the same metric.
How to rememberIn 応答時間, 応答 is the word used to render the English "response," so it means the same as the katakana レスポンス.
Exam points
ExamResponse time 応答時間 covers the span from the moment the user finishes input to the moment the system's response starts to come back.
A benchmarkベンチマーク is a method that actually runs a workload fixed as the standard on the system being evaluated and uses measured values such as processing time to evaluate and compare performance.
Fix the standard workload
Run on real machines under the same conditions
Compare measured values
Line up the measurements, such as time taken, and evaluate performance.
3 / 3
Because every system is measured with the same workload under the same conditions, results can be compared fairly even across different models and configurations. The workload is chosen to match what the system will be used for. It is also called a benchmark test (ベンチマークテスト).
AnalogyIt is like a test where everyone solves the same questions within the same time limit and ability is compared by score.
How to rememberThe word originally meant a mark cut by surveyors as a reference for height, and the sense of "a reference for comparison" carried over to performance evaluation.
Exam points
ExamThe values a benchmark ベンチマーク produces are actual measurements taken by really running the standard workload on the machine being evaluated.
ExamSPEC is a neutral organization that develops benchmarks for comparing performance and publishes the results.