I'd like to point out that startup latency isn't an issue for all apps. Appengine used to aggressively kill idle instances. When I wrote my app almost every request it served was a startup request. As a result, I optimized for startup latency (used python, avoided django, import almost nothing), and my app can serve startup requests in <300ms.
As such, I'd rather the scheduler start a new instance than bill me for an idle one. On Nov 13, 12:59 pm, Marcel Manz <marcel.m...@gmail.com> wrote: > Simple answer: unwanted latency > > I'm also noticing this problem in one of my apps (see my > posthttps://groups.google.com/forum/?hl=en#!topic/google-appengine/dQQ2y0...) > > If the scheduler decides to spin up a new instance, this will add unwanted > warmup latency to the current request. If you look at my post and attached > screenshot, you'll see that obviousely there's really no need for the > scheduler to startup a third instance, while the second resident one is > just sitting there and doing nothing. > > Could someone from Google please look into this issue and explain why this > is happening? Thanks! -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To post to this group, send email to google-appengine@googlegroups.com. To unsubscribe from this group, send email to google-appengine+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.