OFFSHORE VPS / GROW
Grow.
More headroom for a growing workload.
Several modest services or a busier API. Profile the actual bottleneck before moving to a larger instance.
Start with these resources.
- Compute
- 4 vCPU
- Memory
- 4 GB
- Storage
- 100 GB
- Transfer budget
- 4 TB
Adjust resources, image, network options and the service period in the configurator. Choose Malaysia, Romania or Switzerland as the server location, with no location surcharge. Backup countries remain unconfirmed.
Plan around the whole workload.
Memory is shared by every process on the instance. Include database caches, job queues and your deployment process, and test at an expected busy period.
Build a resource budget →Know what you own.
Plan for application security, updates, observability and an independent recovery process. If you select a backup option, review the restore procedure and test it against your application.
Prepare your first restore drill →Extra room for several modest processes.
Grow is a comparison point for several modest processes or a busier application with a known workload. Its larger starting allocation can make room for a web process, data service and bounded background work, but every process still shares one failure and maintenance boundary.
A worked starting point
An illustrative app serves an API and performs a daily data import. Measure the import’s peak memory and write activity alongside normal requests. Decide whether to schedule it at a quieter time, reduce concurrency, add a measured resource increment or place it on a separate service.
When this starting point is a poor fit
Separate services when a single restart, disk problem or access mistake would affect work that should be independent. A larger shared instance does not create that separation. If one query or job keeps growing without a limit, fix its behavior before relying on a larger allocation.
Give every option a job.
The base configuration contains 4 vCPU, 4 GB RAM, 100 GB storage, a 4 TB transfer budget and 1 IPv4 address. IPv6 is requested separately and no backup option is selected by default. These are catalogue allocations; confirm the actual supplier’s hardware, counting and restore rules.
- Memory: leave room for the operating system, runtime, database and observed peaks. Adding 1 GB to Grow gives a monthly subtotal of $29.50 before term savings.
- CPU: compare the processes doing work and their concurrency. It does not describe a guaranteed processor model or requests per second.
- Storage and transfer: budget retained data separately from the bytes you move. Off-server exports may use both.
- Backups and networking: define the data you need back and the addresses your application uses. Selecting an option does not document the operating procedure for you.
Make the first review concrete.
Record a normal operating period, one heavier task and one release. Note memory available, the largest growing directories and the result of a local application check. Keep a small change log so an unexpected result has a starting point. Then compare the selected period’s whole upfront total in the configurator; a longer term changes the price, not the resources.
Read the pressure signals before upgrading → · Build a resource worksheet · Check catalogue facts and service boundaries