I'll take a shot at this... The latency graph (milliseconds/request) shows you the average of the 'ms' field from the logs. The wall clock time of the request including the startup time for the instance. The logs also indicate if the request was a new/loading request.
If I understand your question correctly... There isn't really any time waiting for a request. No matter what the scheduler decides it's going to make a request, new or existing it's your code that is using the time to 'startup'... python imports, framework, loading that stuff from disk... that is what takes the startup time. Is that what you mean or are talking about pending_ms/throttle_code? -mike On Sep 5, 8:42 pm, GAEfan <ken...@gmail.com> wrote: > Do those request latency values include the time waiting for a new > instance, or does the clock start after loading into and running in > the instance? If the former, where it represents the total latency of > every request, then we can properly measure the effect of changing the > instance sliders. Please clarify. -- 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.
