Infrastructure sizing recommendations
info
TL;DR : Simplicité does not provide fixed sizing recommendations for production apps. Use standard Java webapp practices and adapt to your workloads.
Factors
Sizing an infrastructure—RAM, disk, CPUs, etc.—depends mostly on the application itself rather than the platform.
A number of factors can impact resource requirements:
- Database type and location
- Volume and nature of data
- Backup operations
- Number of simultaneous users (both normal and peak)
- Usage patterns
- Number of business objects managed by the application
- Heavy processing tasks, such as batch loads or large computations
- High API traffic or unusual third-party usage
- High availability requirements
- Other environment-specific constraints
Development and test environments
danger
These are meant for dev/test workloads only and do not apply to production, especially where backups and scalability matter.
For small-scale environments with few users and limited datasets, rough VM sizing guidelines are:
| VM Size | RAM | Disk | vCPU | Approx. # of Dev/Test Instances |
|---|---|---|---|---|
| Small | 4 GB | 20 GB | 2 | ~5 instances + databases |
| Medium | 8 GB | 40 GB | 4 | ~10 instances + databases |
| Large | 16 GB | 80 GB | 8 | ~20 instances + databases |
Production environments
There’s no one-size-fits-all for production, nor is the data we have on our client's sizings sufficient to give recommendations.
Best practices:
- Reserve dev/test sizing for low-load environments only
- Use scalable infrastructure
- Document usage and growth expectations to support future adjustments
- Proceed with regular & realistic high-load benches (with expected maximal number of concurrent users + representative use case scenarii & think times) to define/maintain the appropriate sizing for your application