> ## Documentation Index
> Fetch the complete documentation index at: https://docs.repocloud.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Scaling Resources

> Learn how RepoCloud matches resources to your application's demand

## Scaling on RepoCloud

As your application's needs evolve, its resource allocation should keep up. RepoCloud's approach to scaling is chosen when you deploy: you either enable elastic autoscaling (the default) and let the platform adjust resources for you, or you pick a fixed server size that you can scale manually later.

<Tip>
  For most applications we recommend keeping autoscaling enabled. It matches your resource allocation to actual usage automatically, with no manual intervention required.
</Tip>

## Scaling Options

RepoCloud provides two resource allocation modes, selected on the deployment page:

<CardGroup cols={2}>
  <Card title="Elastic Autoscaling" icon="chart-line">
    The default. Automatically adjusts compute and storage resources based on actual usage, with hourly billing
  </Card>

  <Card title="Fixed Server Size" icon="lock">
    Disable the autoscale toggle at deployment to select a specific resource tier with a predictable monthly maximum
  </Card>
</CardGroup>

## Elastic Autoscaling

Elastic Autoscaling is RepoCloud's intelligent resource management system that continuously monitors your application's resource utilization and automatically adjusts resources as needed.

### How Autoscaling Works

<Steps>
  <Step title="Resource Monitoring">
    Our system continuously monitors the CPU, RAM, and storage utilization of your application.
  </Step>

  <Step title="Automatic Adjustment">
    When resource usage increases, your instance is moved to a larger resource tier to maintain performance.
  </Step>

  <Step title="Scaling Down">
    During periods of lower activity, resources scale down automatically to reduce costs.
  </Step>

  <Step title="Hourly Billing">
    You're billed on an hourly basis for the resource tier your instance actually used each hour, rather than a fixed monthly amount.
  </Step>
</Steps>

### Benefits of Autoscaling

* **Cost Efficiency**: Pay only for the resources you actually use
* **Performance Optimization**: Automatically handle traffic spikes without manual intervention
* **Resource Efficiency**: Reduce waste during low-traffic periods
* **Simplicity**: No need to predict exact resource requirements

### Enabling Autoscaling

Autoscaling is enabled with the "Autoscale compute and storage resources" toggle on the deployment page, and it's switched on by default:

<Steps>
  <Step title="Choose Your Application">
    Select an application from the marketplace and click "Deploy Now".
  </Step>

  <Step title="Keep the Autoscale Toggle Enabled">
    Leave the "Autoscale compute and storage resources" toggle switched on.
  </Step>

  <Step title="Deploy">
    Click "Deploy App". Your instance will scale between resource tiers automatically from then on.
  </Step>
</Steps>

### Tracking Autoscaling Activity

To see how your instance has been scaling:

<Steps>
  <Step title="Open Instance Management">
    Go to your application's management page in the RepoCloud dashboard. The "Server Size" field will show "Autoscaling".
  </Step>

  <Step title="View Usage Chart">
    Click "View Usage Chart" to see which resource tier your application occupied hour by hour. This is exactly what your hourly billing is based on.
  </Step>
</Steps>

## Fixed Server Sizes

If you prefer predictable, fixed resources, disable the autoscale toggle on the deployment page and select a specific server size.

### Available Server Sizes

| Tier | RAM   | vCPU | SSD Storage | Hourly Price | Monthly Price |
| ---- | ----- | ---- | ----------- | ------------ | ------------- |
| 1    | 1 GB  | 1    | 20 GB       | \$0.0042     | \$3.00        |
| 2    | 2 GB  | 1    | 30 GB       | \$0.0083     | \$6.00        |
| 3    | 2 GB  | 2    | 40 GB       | \$0.0125     | \$9.00        |
| 4    | 4 GB  | 2    | 40 GB       | \$0.0167     | \$12.00       |
| 5    | 8 GB  | 4    | 80 GB       | \$0.0333     | \$24.00       |
| 6    | 16 GB | 8    | 160 GB      | \$0.0667     | \$48.00       |
| 7    | 32 GB | 16   | 320 GB      | \$0.1333     | \$96.00       |

<Note>
  Some applications have minimum resource requirements, so the smallest tiers may be unavailable for them on the deployment page. Platform apps such as Coolify and Dokploy, for example, require at least Tier 4 (4 GB RAM / 2 vCPU).
</Note>

### Adding Extra Storage

If you need more storage than your tier includes, use the "Add Extra SSD" option on the deployment page. Additional NVMe SSD storage costs \$0.10 per GB per month.

### Changing a Fixed Server Size

Instances deployed with a fixed server size can be scaled manually: open your application's management page and change the server size to a different tier. You can also switch a fixed-size instance over to autoscaling at any time.

<Warning>
  Switching modes is one-directional: a fixed-size (manual scaling) instance can be moved to autoscaling, but an autoscaled instance cannot be switched back to a fixed server size. If you may want manual control later, deploy with a fixed size.
</Warning>

## Choosing the Right Scaling Method

<AccordionGroup>
  <Accordion title="When to Choose Autoscaling">
    Autoscaling is ideal for:

    * Applications with variable traffic patterns
    * Websites with occasional traffic spikes
    * Cost-sensitive deployments
    * Development and testing environments
    * When you're unsure about exact resource requirements
  </Accordion>

  <Accordion title="When to Choose a Fixed Size">
    A fixed server size works best for:

    * Applications with predictable, consistent traffic
    * Workloads requiring guaranteed resources at all times
    * Situations where budget predictability is important
    * Applications with specific performance requirements
  </Accordion>
</AccordionGroup>

## Scaling Limitations

Be aware of these scaling limitations:

* **Minimum Resources**: Each application type has minimum resource requirements, which may disable the smallest tiers at deployment
* **Maximum Tier**: Resources cannot exceed the highest available tier (32 GB RAM, 16 vCPU)
* **Platform Apps**: [Platform Apps](/platforms/introduction) (Coolify and Dokploy) cannot use autoscaling, because autoscaling is not available for backends that require a dedicated IPv4 address and root access

## Next Steps

<CardGroup>
  <Card title="Environment Variables" icon="gear" href="/user-guide/environment-variables">
    Learn how to configure your application through environment variables
  </Card>

  <Card title="Updates" icon="arrows-rotate" href="/user-guide/updates">
    Keep your application updated with the latest features and security patches
  </Card>

  <Card title="Troubleshooting" icon="wrench" href="/user-guide/troubleshooting">
    Resolve common issues with your RepoCloud applications
  </Card>
</CardGroup>