Serverless Containers — Scaling & Lifecycle Management
This guide explains how to scale, change, and retire Serverless Container applications in a safe, predictable, and production-ready way. It focuses on compute scaling, redeploy behavior, billing impact, and data lifecycle based on how the CloudRaya platform actually operates.
Scaling Model Overview
Serverless Containers use a package-based compute scaling model.
When you resize an application, you change:
- CPU allocation
- Memory (RAM) limits
Storage size does not change.
Each application includes a fixed-size, platform-managed storage volume (50GB). Resizing a package affects performance and concurrency, but it does not increase or decrease disk capacity.
This model keeps scaling simple and predictable:
- Scale up or down when performance changes
- Keep data stable across package changes
Compute vs Storage Model
It is important to understand which parts of your application can be scaled and which cannot.
Compute (CPU & Memory)
- Scaled by resizing the package
- Affects how many requests your app can handle and how fast it responds
- A short service interruption may occur
Storage (Disk)
- Fixed per application (50GB)
- Preserved across redeploys and package changes
- Deleted when the application is deleted
If your application slows down under load, scale compute.
If you need:
- Larger datasets
- Long-term backups
- Shared or external access to data
Consider integrating with StorageRaya (CloudRaya Object Storage) or backup services.
📄 See: StorageRaya
How to Resize an Application
- Open My Apps from the Marketplace
- Select the application you want to scale
- In the General tab, click Resize Package
- Choose a new package size
- Review the updated monthly cost
- Click Resize to confirm
The application will enter a redeploying / terminating state while the new compute limits are applied. Once complete, the status returns to Running.
What Happens When You Redeploy
Redeploying an application restarts its containers and applies the latest platform and configuration changes.
A redeploy may occur when you:
- Resize the package (CPU/RAM)
- Update environment variables
- Apply platform-level updates
During redeploy:
- Containers are restarted
- The application may be temporarily unavailable
- Attached storage is preserved
Use redeploy as a controlled change action, especially for production workloads.
Data Behavior and Persistence
Serverless Containers can support both lightweight and data-driven applications.
Data-Persistent Applications
Some applications store data on the platform-managed storage volume, such as:
- Content management systems
- Internal databases
- Learning platforms
This data:
- Persists across redeploys
- Is tied to the lifecycle of the application
Lightweight or Non-Persistent Applications
Some workloads do not rely on stored data, such as:
- APIs
- Automation tools
- Demo or test environments
These can be resized or redeployed with minimal risk.
If your workload requires:
- Large or growing datasets
- External backups
- Data sharing across systems
Consider integrating with StorageRaya (CloudRaya Object Storage) or backup services.
📄 See: StorageRaya
Billing Impact
Serverless Containers use a monthly, usage-based billing model.
Usage is tracked throughout the month, and charges are settled at the end of each billing period based on how long each package size is active.
What This Means for You
- You can resize at any time, and your cost will adjust automatically
- If you upgrade or downgrade during the month:
- The previous package is charged up to the day of the change
- The new package is charged from the change day to the end of the month
- If you resize multiple times, you are only charged for the days each package is active
Account Balance and Service Availability
At the end of the billing period, your account balance is used to settle Marketplace charges.
To avoid service interruptions, make sure your account has sufficient balance.
Example — How Prorated Billing Works
Here is a simple example for a 30-day month:
| Day Range | Package Size | Billed For |
|---|---|---|
| Day 1–10 | Nano | 10 days of Nano |
| Day 11–20 | Small | 10 days of Small |
| Day 21–30 | Nano | 10 days of Nano |
Monthly Total:
- Nano → 20 days total (Day 1–10 + Day 21–30)
- Small → 10 days total (Day 11–20)
At the end of the month, charges are calculated by adding up the total number of days each package size was active, even if you switch back to a previous size.
Data Retention and Deletion
Each Serverless Container application includes a platform-managed storage volume.
When you delete an application:
- The application and its storage are removed together
- Stored data is permanently deleted
Before deleting an app:
- Export important data
- Back up content to external storage or backup services
Once deleted, data recovery is not supported.
Safe Change Practices
For production environments, follow these best practices:
- Resize during low-traffic periods to reduce user impact
- Change one thing at a time (package size or configuration, not both)
- Verify application health after each redeploy
- Back up data before major changes
These practices help reduce risk and make troubleshooting easier if issues occur.
When to Use Serverless Containers
Serverless Containers are best suited for:
- Business applications and internal tools
- Lightweight production workloads
- Rapid deployment and scaling needs
For workloads that require:
- Large-scale persistent storage
- Advanced networking
- Fine-grained infrastructure control
Consider deploying directly on Virtual Machines or KubeRaya clusters.
Next Steps
-
Learn how to Monitor and Troubleshoot Applications
-
Review Domain & Access Management for custom domains and HTTPS
-
Explore Configuration & Environment Variables for advanced setup and secrets management
📄 Serverless Containers — Configuration & Environment Variables
These sections help you operate Serverless Containers safely and efficiently in production environments.