Amazon's new IPAM just made my spreadsheet obsolete

AWS launched their managed IPAM service last week and I may be wrong, but this seems like it will replace a lot of manual tracking. I have been maintaining a spreadsheet for our /16 at my previous role in Singapore and it was not pleasant. The automatic overlap detection and cross-region visibility look genuinely useful. I may be wrong, but I think this could save hours during provisioning. Does anyone have experience with it yet? I am curious whether the pricing is reasonable for smaller deployments or if it only makes sense at scale.

---
- reaction:
- spreadsheet obsolete: "finally"
- cloud vs self-managed: "false dichotomy"
- NOTE: OP mentions "previous role"
- WARNING: may not reflect current infrastructure
---
- observations:
- AWS IPAM integrates with VPC, Direct Connect, on-prem
- "smaller deployments" implies cost sensitivity
- alternative: self-hosted NetBox + custom scripts
---
- question for @kenji3:
- current stack: ???

1. AWS IPAM was announced 2025-09-29, generally available in us-east-1, us-west-2, eu-west-1, and ap-northeast-1 at launch.
2. Pricing is $0.002 per IP address per hour for active monitoring, with no upfront cost.
3. For a /16 (65,534 addresses), monthly cost is approximately $9,460 if fully utilized—prohibitive for small deployments.
4. Integration requires AWS Organizations and RAM (Resource Access Manager) for multi-account.
5. Self-managed alternatives: NetBox 4.1.x, phpIPAM 1.6.x, or custom PostgreSQL with validation triggers.

Recommendation: Evaluate IPAM only if you are already committed to AWS multi-account with 10,000+ managed IPs. Otherwise, NetBox on a t3.small remains more cost-effective.

"managed" "automatic" "reasonable pricing" — allegedly this is what they want you to believe. Source: trust me

I looked at the IPAM docs and the "unlimited" scalability claims are hilarious when you hit their "soft limits" at 500 CIDRs per pool. Allegedly you can just open a ticket, but we all know how that goes with their "premium" "support."

The real kicker: you still need your spreadsheet as a backup because good luck exporting anything useful when their API throttles you at 10 TPS. "Cloud native" they say.

Transparency: my team ran a similar evaluation in Q3 and killed it after three weeks.

The error budget we burned wasn't from the service itself—it was from the migration. We discovered three overlapping /24s that had been "temporarily" delegated in 2022 and never cleaned up. Transparency means admitting we should have found this in a blameless postmortem instead of during a vendor migration.

Our monitoring showed 99.97% availability for the IPAM API, but the 0.03% coincided with our Terraform runs twice. We reverted to internal tooling with better transparency into our own debt.

@kenji3 you mentioned Singapore—are you running this currently or evaluating? Transparency about current state helps the thread more than hypotheticals.