Be sure you include some urlfetch calls. The cpu cycles your code burns through are likely your own. However, you can see the variability of the infrastructure clearly -- Google provides a nice graph of such for urlfetch.
I see variability with urlfetch. Nothing too huge, but it is there, and if my app is at the margin for a new instance, urlfetch slowing down would absolutely result in new instances. So what to do if your GAE Finance: 1) Invest $10 million for infrastructure to halve urlfetch response times >> makes app latency go down >> fewer instances >> lots less revenue. vs. 2) Make no investment >> let app latency creep up >> more instances >> lots more revenue. This is an incredibly perverse setup requiring huge amounts of trust. However, if you look at one of Google's core competency which is algorithmically driving revenue it is a perfect fit. There are very likely ways that Google can put some skin the game and resolve these trust issues. I'll be interested to see what happens. cheers, stevep On May 22, 1:11 pm, "Brandon Wirtz" <drak...@digerat.com> wrote: > I am tempted to put up an app just to prove this doesn't happen. I have > never seen my Instance get slow because of someone else's. What do you want > for a test? Calculate pi? Hash a file? -- 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.
