Hello Google App Engine Team,
I am currently trying to understand the future consequences of the new pricing model. One of our typical request types has an average latency of 0.75 seconds (waiting for an URL Fetch most of the time) and needs around 0.10 seconds CPU. - With the current pricing, CPU costs (dynamic instance costs) are $2.78 per million requests: 10^6 * 0.10 CPU seconds/request / 3600 seconds/hour * 0.10 $/CPU hour - With the new model, dynamic instance costs seem to be $16.67 per million requests (without any idle time): 10^6 * 0.75 instance seconds/request / 3600 seconds/hour * 0.08$/instance hour That's an 6x increase without even considering idle time for the instances. If the scheduler can make the instances answer requests 60% of the time, that would mean a 10x increase. The current scheduler seems to be around 15-20%, which would mean a 30x-40x increase. Do these numbers sound about right? If not, could you please explain which of the assumptions I made are wrong? I hope you all have a great time at Google I/O! Tammo Freese -------------------------------- Tammo Freese FlockOfBirds UG (haftungsbeschränkt) Marie-Curie-Straße 1, 26129 Oldenburg, Germany Amtsgericht Hamburg, HRB 114558 Sitz Hamburg Geschäftsführer: Tammo Freese -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. 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?hl=en.
