For my fairly modest web application, I seem to be hitting some of the AWS VPC limits, which made wonder if I'm maybe structuring things incorrectly:
I have an application for which we have three environments:
Staging
,QA
,Production
. So I have a VPC for each.These three VPCs exist in the US-East-1 Region
Each Application is available in two availability zones (
us-east-1b
&us-east-1c
)
An example of a limit I'm hitting is for the following circumstance:
I have Public and Private facing subnets. And because I'm using two Availability Zones, I have Public and Private Subnets for each (4 subnets in total, per VPC)
The Private Subnets need internet access, so their respective Route Tables have entries for NAT Gateways.
So in the end, this means each Private Subnet has a NAT Gateway, and therefore each NAT Gateway has an Elastic IP. And because I have three environments, and two Availability Zones for each, that's already 6 Elastic IP addresses, and the Elastic IP limit per Region is 5.
So obviously, I can request an increase, but it made me wonder if I'm doing something wrong. Could some of these resources in fact share NAT Gateways, or Elastic IPs, or something to that effect?