So, when do you actually want to focus on making things fast, versus just getting a lot of things done?
Diederik Van Der Velde
·15423 views
Hey everyone, I've been thinking about system performance lately. We always talk about making things fast, but sometimes, trying to shave off every last millisecond of latency might not be the best idea if it really tanks our overall throughput. Like, caching super aggressively could cut down latency for repeat requests, but it might also cause stale data or hog memory, which then impacts how many requests we can actually handle at the same time. What do you all think about this? When do you decide one is more important than the other, and what do you do when you need both
28 comments