As our app has grown, the amount of initialization work to do in order to 
make an instance useful has increased to the point that GAE often terminates 
requests that start a new instance. We had a warmup servlet, but the work 
started to commonly exceed the deadline. We're now looking to pare down the 
warmup, but we're having trouble with tuning since the behavior seems pretty 
different on the dev server from GAE. On the dev server, the initialization 
(loading our initial servlets) will complete in the vicinity of 4sec while 
GAE instance startup is often more than 20sec. We often see inexplicable 
gaps in our logging of 15sec or more (seemingly hung up), but it doesn't 
seem repeatable.

Our focus to date has been to optimize the JDO/Datanucleus startup, but we 
believe that a significant contributor might be the number of classes 
loaded. Is it accurate to conclude that class loading carries a larger 
overhead on GAE than other environments? Are there any tools that will help 
us understand the processing during this startup? The Appstats tool shows 
long delays, but we don't know what is taking time besides the RPCs.

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine for Java" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-appengine-java/-/siKBgvUpsk4J.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine-java?hl=en.

Reply via email to