https://gianarb.it/blog/server-time-vs-response-time

If you find yourself in San Francisco walking nearby Market Street, you should consider stopping at the Jewish Museum. There is a charming Pastrami place just next to it. It is a sandwich place with good lemonade. It only takes 3-4 minutes to get your meal, and from there it takes no more than 15 minutes walking to be in front of the Ocean. Very nice! Now, let’s consider this other scenario. It is lunchtime, and you are starving. You rush outside your office, and you run to the Pastrami place close to the Jewish Museum. After 35 minutes of wait, you get your sandwich and start eating it asking yourself: why it took so long this time? Shall I probably have walked to the next place to get a faster meal?

Something similar can happen to your Services as well! And that’s precisely the phenomena in computer science we try to capture using the concepts of server time and response time. Server time aims to measures how much a server takes to run a specific action. Let’s say consider an example operation the generation of a monthly report: it usually takes 2ms, but if a lot of customers require the same kind of report at the same time and your system saturates? This situation might very quickly end up in having a subset of them getting the report in more than 1 minute or actually in the timeout of the operation. The time it takes for a customer to get his report is what is typically called response time.