I had this question asked, if we need 3000/tps to be served by a spring/spring boot application running on tomcat(200 threads), how many servers I need to provision to have good performance. How many servers I need to provision so that my request are not queued. How do i calculate and provide a capacity estimation.
There's really no way to tell. It depends on many factors, and not the least of which is how fast your application code can chew through that kind of load.
The only realistic way to tell would be to benchmark successively heavy loads. This is obviously a system where an elastic services system would be useful, though.
I'm going to be a "small government" candidate. I'll be the government. Just me. No one else.
Heroic work plunger man. Please allow me to introduce you to this tiny ad:
The Low Tech Laboratory Movie Kickstarter is LIVE NOW!