How much RAM does a VPS need?

"How much RAM do I need?" is the most common question when buying a VPS — and the honest answer is: it depends entirely on what you're running. RAM is usually the first resource you run out of, so getting it right matters. Here's a practical guide by workload.

Why RAM is the resource that bites first

Every running program holds memory: your web server, database, language runtime, control panel, and the OS itself. When RAM fills up, the server either swaps to slow disk or the kernel starts killing processes. Unlike CPU — which just slows things down when busy — running out of memory causes crashes. That's why sizing RAM correctly is the priority.

Quick reference by workload

• 1 GB — a static website, a small bot, a single lightweight service, a personal VPN. Fine for low-traffic, single-purpose boxes.
• 2 GB — a WordPress site, a small Node.js or Python app, a personal Nextcloud, or a lean control panel. The comfortable minimum for most real projects.
• 4 GB — a busier app plus its database, several small sites, a full control panel like cPanel or CyberPanel with a couple of live sites, light Docker use.
• 8 GB — multiple sites, heavier databases, many containers, or a memory-hungry application. Room to grow without constant tuning.
• 16 GB+ — high-traffic sites, large databases, data processing, or many services on one box.

What eats RAM

Understanding the big consumers helps you estimate:

• Databases are usually the largest. MySQL/MariaDB and PostgreSQL cache data in memory for speed; the more data and connections, the more they want.
• PHP-FPM / app workers each hold memory, and you run several. More concurrent visitors means more workers means more RAM.
• Control panels (cPanel, Plesk) add a baseline of several hundred MB to a couple of GB before you host anything.
• In-memory caches like Redis and Memcached trade RAM for speed by design.

Real examples

• A brochure site on Nginx: happily runs on 1 GB.
• A single WordPress site with moderate traffic: 2 GB is comfortable; 4 GB gives caching headroom.
• A Node.js app with a PostgreSQL database: 4 GB is a solid starting point.
• cPanel hosting a handful of client sites: 4 GB minimum, 8 GB to breathe.
• Nextcloud for a family with real files: 2–4 GB, more if many people sync at once.

Check what you're actually using

Guessing is fine to start, but measure once you're live:

free -h

Watch the available column. If it's chronically near zero and swap is heavily used, you're short on RAM. htop shows which processes are the biggest consumers:

htop

Don't over-buy either

It's tempting to grab the biggest plan "to be safe," but idle RAM is wasted money. A better strategy is to size for today plus modest headroom, then watch the numbers. Because Nxeon lets you resize a VPS to add RAM without rebuilding, there's no penalty for starting lean — you upgrade the moment the data says you need to, not before.

A simple decision rule

1. Identify your heaviest component (usually the database or app).
2. Start one tier above the bare minimum it needs.
3. Add a small buffer for the OS and traffic spikes.
4. Monitor free -h for a week and adjust.

The short answer

• Just a website? 1–2 GB.
• WordPress or a small app? 2–4 GB.
• App plus a real database, or a control panel? 4 GB.
• Multiple sites, heavy databases, or containers? 8 GB+.

Pick the tier that fits your heaviest workload with a little room to spare, measure real usage, and resize when the numbers tell you to. That's how you get performance without overpaying.