Obviously what you say is true, however in the regular deployment scenario, the discrepancy is obviously minimized, as far as I observed, the instances that can shutdown, shutdown at once and new ones are born and I hope/assume the scheduler skips the alive/old instances and route requests to new ones until old requests at hand are completed, however there is no such(observed) routine for multi-module uploads, would be great if there was
I've noticed a very significant performance increase after I isolated the frontend requests in the default module, request delays that previously reached 60 seconds vanished, now the interfaces are really snappy, I'm really happy to observe such performances (frontend requests are pretty optimized to run mostly around 1 seconds, however pre-modules, the request times were around 20 seconds instead) I sincerely hope the performance increase is because of the modules/scheduling and not because I temporarily got into a nice datacenter :) -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/google-appengine. For more options, visit https://groups.google.com/d/optout.
