0

How could I dispatch the request by instances' type dynamically

I have a ELB contains an auto scaling group and an indenpendent ec2 machine

However the indenpendent ec2 machine is t2.large

and the instances in the auto scaling group are t2.micro

How could I dispatch the users requests in reasonable way.

From my point, If there are 100 request, I think 90% will go to the t2.large and the other 10% go to the t2.micro (suppose there is only one instance in the scaling group).

But if there are 2 instances in the scaling group, I think the scaling group should take about 15~20% requests.

How could I set the kind of policy with ELB and route 53 thanks so much

https://i.stack.imgur.com/SkWn2.png

https://i.stack.imgur.com/zmFIK.png

user3675188
  • 99
  • 1
  • 1
    Why don't you just keep all machines in the ELB the same type, and avoid this problem altogether? – womble Sep 27 '15 at 07:15
  • afaik ELB has only two working modes. dispatch requests evenly between availability zones or *try* to dispatch requests evenly betwen machines – theist Sep 28 '15 at 16:59

0 Answers0