2 ms·
did I miss something or the methodology lacks a warm up step? What I mean is, I'd expect a first run of 5 mins to warm up the servers, than another one to actu
by riffraff 14y ago
did I miss something or the methodology lacks a warm up step?
What I mean is, I'd expect a first run of 5 mins to warm up the servers, than another one to actually get the data.
Otherwise, while averages will be mostly the same, numbers like the dropped connections _could_ be misleading.