Plans
What a plan includes, how the sizes differ, and where prices are published. No plan number here is a guess.
A plan is one service: one PostgreSQL instance with a defined vCPU, RAM, storage and I/O allocation, in one location, on one major version. Every price is per service, per month.
All four resource numbers are enforced by the kernel rather than advertised, and the same configuration feeds this page, the pricing page and the dashboard, so the three cannot disagree.
The sizes
| Plan | vCPU | RAM | Storage | Built for |
|---|---|---|---|---|
| Starter | 2 | 4 GiB | 80 GiB | Development, staging and early-stage apps. |
| Professional | 4 | 8 GiB | 160 GiB | Production SaaS and web applications. |
| Business | 8 | 16 GiB | 320 GiB | Growing products with real traffic. |
| Performance | 8 | 32 GiB | 500 GiB | Memory-intensive workloads and analytics. |
| High Performance | 16 | 64 GiB | 1 TiB | Large, business-critical databases. |
| Scale | 16 | 128 GiB | 2 TiB | Very large production databases. |
| Scale 256 | 256 GiB | Memory-bound workloads above 128 GB. |
The memory number includes page cache for your files, which is the honest definition. See what a memory allocation includes.
Prices
Prices live on the pricing page, which reads the same configuration as the table above.
Is there a free trial?
No. Every service reserves dedicated CPU, memory and storage the moment it’s created — capacity no other customer can use. That guarantee is why a payment method is required to start.
What is included in every plan
- Continuous backup with write-ahead log archiving to two repositories in two countries
- Restore verification on a size-based cadence
- Backup and archive alerts by email, on six conditions
- Pooled and direct connection endpoints
- Point-in-time recovery across the whole retention window
- Dashboard exports with a 24-hour download link
- Infrastructure monitoring, 24 hours a day
None of that is a paid add-on. A backup you have not verified is not a feature we would charge extra to make real.
Inside a plan, at no extra charge
Logical databases up to the published plan limit, plus schemas and roles. We do not charge per database.
The role count is a guideline rather than a limit, because
CREATEROLE is granted and a limit we cannot enforce is not a limit.
I/O and connection allocations
Each plan carries an I/O allocation as a pair of ceilings, IOPS and throughput, and a connection allocation. The I/O figures are set from a benchmark on the production host class and published as measured values. We do not print an estimate in their place.
max_connections is the connection number that stops you, and it is listed per plan on
plan limits.
Retention
Seven, 30, 90 or 365 days, chosen per service on the Protection step when you create it. Longer retention costs more because it consumes real storage across both repositories. See retention.
Changing plan
Plans go up, not down, because every smaller plan also has less storage and storage does not shrink. See scaling.
What you will see
The plan is on the service's Overview, and the resources it allocates are on Configuration. The charge for it appears on your organisation's invoice as a line of its own. See invoices.
Troubleshooting
The plan you want is disabled in the dialog. It has less storage than the service currently has. That is the one direction a plan change cannot go.
A price here differs from what you were quoted. The pricing page and this page read the same configuration. If they disagree with an invoice, tell us and quote the invoice number.
You need a size that is not listed. Ask, and say what the workload is.
Contact support any time by email or from your dashboard. Every service is monitored around the clock.