Manage Compute Usage
What you are charged for
Each project in Platform API runs on dedicated or shared infrastructure depending on the tenancy model and deployment configuration.
Platform supports multiple SQL-compatible database engines (including PostgreSQL and other SQL-based systems), deployed across AWS-backed infrastructure tiers.
You are charged for Compute resources based on the infrastructure model assigned to your project:
- Silo – dedicated infrastructure per tenant (fully isolated compute + database)
- Bridge – shared infrastructure with logically isolated compute and resources
- Pool – fully shared multi-tenant infrastructure
Compute usage is independent of your database engine type and is billed based on underlying infrastructure consumption.
How charges are calculated
Compute is billed per hour based on the infrastructure tier assigned to your project.
You are charged for every hour that a project is active and consuming resources. Partial hours are billed as full hours.
Paused projects do not contribute to compute usage. Compute billing is based only on active runtime consumption.
Compute usage is not covered by spend caps and is billed according to AWS-based infrastructure pricing tiers.
Tenancy impact on billing
Billing varies depending on the tenancy model:
Silo (Dedicated)
- Dedicated compute per tenant
- Highest cost tier
- Full isolation of CPU, memory, storage, and network
- Predictable performance
Bridge (Hybrid)
- Shared infrastructure with logical isolation
- Medium cost tier
- Balanced performance and cost efficiency
Pool (Shared)
- Fully shared multi-tenant compute
- Lowest cost tier
- Optimized for scale and cost efficiency
Example
Your billing cycle runs from January 1 to January 31.
On January 10 at 4:30 PM, you upgrade your project from a Micro tier to a Small tier compute allocation.
At the end of the billing cycle you are billed for:
- 233 hours of Micro compute
- 512 hours of Small compute
| Time Window | Compute Tier | Hours Billed | Description |
|---|---|---|---|
| Jan 1, 00:00 – Jan 10, 04:00 | Micro | 232 | Active usage |
| Jan 10, 04:00 – Jan 10, 04:30 | Micro | 1 | Partial hour billed full |
| Jan 10, 04:30 – Jan 10, 05:00 | Small | 1 | Partial hour billed full |
| Jan 10, 05:00 – Jan 31 | Small | 511 | Active usage |
Usage on your invoice
Compute usage appears on your invoice as:
- Compute Hours
- Silo Compute Hours
- Bridge Compute Hours
- Pool Compute Hours
depending on the tenancy model used.
Compute credits
Platform API plans may include compute credits depending on your subscription tier.
Compute credits:
- Apply only to compute usage
- Do not apply to storage, networking, or logs
- Reset monthly and do not roll over
- Are shared at the organization level
Pricing model
Platform API pricing is based on:
- AWS infrastructure tier equivalents
- Compute size allocation
- Tenancy model (Silo / Bridge / Pool)
- Region selection and usage intensity
Pricing varies dynamically depending on infrastructure configuration and cloud region.
Billing examples
One project (Pool model)
| Line Item | Hours | Cost |
|---|---|---|
| Platform Plan | - | XXX |
| Compute Hours Project 1 | 730 | XXX |
| Subtotal | XXX | |
| Compute Credits | -XX | |
| Total | XXX |
Multiple projects (Bridge model)
| Line Item | Hours | Cost |
|---|---|---|
| Platform Plan | - | XXX |
| Compute Project 1 | 730 | XXX |
| Compute Project 2 | 730 | XXX |
| Compute Project 3 | 730 | XXX |
| Subtotal | XXX | |
| Compute Credits | -XX | |
| Total | XXX |
Mixed tenancy (Silo + Pool)
| Line Item | Hours | Cost |
|---|---|---|
| Platform Plan | - | XXX |
| Silo Compute (Project 1) | 400 | XXX |
| Pool Compute (Project 2) | 330 | XXX |
| Subtotal | XXX | |
| Compute Credits | -XX | |
| Total | XXX |
View usage
You can view compute usage in the Platform UI:
- Organization → Usage → Compute
The usage dashboard provides:
- Per-project breakdown
- Tenancy model breakdown (Silo / Bridge / Pool)
- Region-based usage
- Hourly and daily metrics
- Cost estimation and forecasting
Optimize usage
- Start with Pool or Bridge for early-stage projects
- Use Silo only for high-security or performance-critical tenants
- Monitor usage dashboards regularly
- Scale compute tiers based on real workload metrics
- Decommission unused projects
- Consolidate low-usage workloads into shared infrastructure
FAQ
Does compute pricing depend on database type?
No. Pricing is based on infrastructure compute usage, not database engine type.
Do all database engines cost the same?
All supported SQL engines are billed using the same compute model, but actual cost may vary depending on resource consumption.
Does tenancy affect performance?
Yes:
- Silo = highest performance consistency
- Bridge = balanced performance
- Pool = optimized cost efficiency with shared resources