Public IP Planner
Plan and allocate public IP address requirements across cloud resources.
Inputs
e.g. load balancers, bastion hosts, VPN endpoints
One per AZ is typical for high availability
1 = no redundancy, 2 = active/standby or dual-AZ, etc.
Total Public IPs Needed
12
Base IPs (no redundancy)
6
IPs for NAT Gateways
3
Additional IPs for Redundancy
6
Step by step
Values used
Services Requiring a Direct Public IP = 3; NAT Gateways = 3; IPs per NAT Gateway = 1; Redundancy Multiplier = 2
Total public IPs needed
total_ips = (direct_services + nat_gateways × ips_per_gateway) × redundancy_factor
Total Public IPs Needed
= 12
Base IPs (no redundancy)
= 6
IPs for NAT Gateways
= 3
Additional IPs for Redundancy
= 6
How it works
Public IP planning sums every service that requires direct internet-routable addressing — load balancers, bastion hosts, VPN endpoints — plus the IPs consumed by NAT gateways (which need at least one public IP each, more if scaling for port capacity), then applies a redundancy multiplier to account for standby capacity, multi-AZ duplication, or failover requirements. Cloud providers often charge per allocated public IP and increasingly charge for idle ones too, so right-sizing this number avoids both under-provisioning (risking port exhaustion or availability gaps) and unnecessary ongoing cost.
Formula
Total public IPs needed
total_ips = (direct_services + nat_gateways × ips_per_gateway) × redundancy_factor
- s
- Services needing a direct public IP
- g
- Number of NAT gateways
- p
- IPs per NAT gateway
- r
- Redundancy multiplier
Frequently Asked Questions
Why does redundancy multiply IP count instead of just adding a fixed number?
Redundancy needs typically scale with the size of the base deployment — doubling capacity for active/standby or multi-AZ resilience means doubling every category of public-facing resource, not just adding a flat buffer, which is why a multiplier models it more accurately than a fixed addend.
Why might a NAT gateway need more than one public IP?
Additional IPs on a NAT gateway multiply its available port pool for connections to any single destination (see the NAT Gateway Capacity calculator) — high-connection-volume workloads to a small number of destinations sometimes need multiple IPs per gateway to avoid port exhaustion.
Do all cloud providers charge for public IPs the same way?
No — pricing models vary and change over time; some providers now charge hourly for any allocated public IPv4 address regardless of use, including idle ones. Check your specific provider's current pricing tools for exact costs rather than assuming a flat rate.
Should private services be included in this count?
No — this planner is specifically for internet-routable public IPs. Services accessible only within your VPC/VNet or over private connectivity (VPN, Direct Connect, ExpressRoute) don't need public IPs and should be planned using the Private IP Planner calculator instead.