Comment on Saving months of compute time with a single Grafana queryComments−zug_zug2yI'm really surprised that 300ms at startup would result in 25% fewer pods.... What % reduction in the total startup time is that?Is it possible the prior measurement happened during a high traffic period and the post measurement happened in a low traffic period?−serverlessmomOP2yIt’s a 50% reduction in startup time, and each “run” for a pod is fairly quick.
Comments
I'm really surprised that 300ms at startup would result in 25% fewer pods.... What % reduction in the total startup time is that?
Is it possible the prior measurement happened during a high traffic period and the post measurement happened in a low traffic period?
It’s a 50% reduction in startup time, and each “run” for a pod is fairly quick.