Overview
An instance class defines the CPU and memory allocated to your database. Nearbase
offers two groups: General Purpose and Dedicated. Pricing is uniform
across every region — the same class costs the same everywhere.
Instance class codes follow this pattern:
pg.{tier}.{size}.{variant}
| Segment | Meaning |
|---|
pg | PostgreSQL engine |
{tier} | n2e / n1e = Economy (ARM) · n2 / n4 = General Purpose · x8 = Dedicated |
{size} | vCPU count, e.g. 2c, medium, xlarge |
{variant} | memory variant within the size class |
pg.n4.4c.1m is a General Purpose instance with 4 vCPU and 16 GB memory.
Catalog
General Purpose
Balanced CPU and memory for most workloads. Includes low-cost economy (ARM)
options for small apps and trials.
| Class Code | vCPU | Memory | Notes |
|---|
pg.n2e.1c.1m | 1 | 2 GB | Economy (ARM) |
pg.n1e.2c.1m | 2 | 2 GB | Economy (ARM) |
pg.n2e.2c.1m | 2 | 4 GB | Economy (ARM) |
pg.n4.2c.1m | 2 | 8 GB | Standard |
pg.n2.4c.1m | 4 | 8 GB | Standard |
pg.n4.4c.1m | 4 | 16 GB | Standard |
pg.n4.6c.1m | 6 | 24 GB | Standard |
pg.n4.8c.1m | 8 | 32 GB | Standard |
Economy instances (pg.n1e.* / pg.n2e.*) are ARM-based and may not be
available in every region — pg.n1e.* in particular is not available in
ap-northeast-1 (Tokyo), ap-southeast-3 (Kuala Lumpur), and
ap-southeast-7 (Bangkok). Availability is validated when you create an
instance; pick a Standard or Dedicated class in those regions.
Dedicated
Dedicated CPU instances with a high memory-to-CPU ratio (8 GB RAM per vCPU).
Designed for CPU-intensive queries, analytics, and latency-sensitive workloads
that cannot tolerate CPU sharing.
| Class Code | vCPU | Memory |
|---|
pg.x8.medium.1 | 2 | 16 GB |
pg.x8.large.1 | 4 | 32 GB |
pg.x8.xlarge.1 | 8 | 64 GB |
pg.x8.2xlarge.1 | 16 | 128 GB |
pg.x8.4xlarge.1 | 32 | 256 GB |
How to Choose
| Scenario | Recommended |
|---|
| Development / staging environment | General Purpose (Economy) |
| Low-traffic production app | General Purpose (Economy) |
| Standard web app or API backend | General Purpose (Standard) |
| High-concurrency or latency-sensitive | General Purpose n4 / Dedicated |
| Analytics / reporting queries | Dedicated |
| Large dataset, CPU-intensive workload | Dedicated |
Scaling
You can change your instance class at any time from the instance detail page. Scaling is handled by the platform with minimal interruption. Your billing is adjusted on a prorated basis for the remainder of the billing period.
Start with a smaller class and scale up as needed. Nearbase supports vertical
scaling without data migration.
