Here's why GAE is (was) better than Heroku: auto scaling.

Here's how you broke it:

You instruct me to save money by telling the system that the maximum idle 
units should be 1. Here's a question: if I do get slashdotted, and I start 
needing 20 active instances (which I dont mind paying for if that happens), 
is "1" still the proper number of idle instances? No! So I have to log in, 
and up the maximum number of idle instances.

If I have to have a pager to tell me when I'm being slashdotted so I can log 
in to modify scaling settings, *then your system offer no advantages over 
Heroku*.


If my load is currently served by 1 instance, then having 20 instances alive 
is a waste of money. I have an app that in one day used 0.02 cpu-hours. 
Under your new system, it says it used 2.8 instance-hours. If your 
explanation is that its because I'm running 20 instances idle, my question 
is "What is wrong with your automated system that thinks I need 20 
instances?"

What I Want:

I want to be able to leave the system on "Automatic". I want it to pick zero 
idle instances for my app that serves 14 requests per day. I want it to pick 
1 idle instance for my app that has 5 instances running steady state. I want 
it to *notice* when I'm being slashdotted and start spinning up idle 
instances *in response* to the escalating traffic. YOU ARE GOOGLE. 
CORRELATION AND PREDICTION IS WHAT YOU DO.

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-appengine/-/NUoyckUf8tYJ.
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.

Reply via email to