WordPress RAM Calculator

Most guides say "2 GB is fine for WordPress." That's true for an empty site with one plugin. Add WooCommerce, a page builder, and 20 plugins, and 2 GB will crash under any real traffic. This calculator gives you a number based on what you're actually running.

Calculate your RAM requirement

How this is calculated

WordPress RAM usage isn't linear with traffic. It scales with concurrent requests, not monthly visitors. A site with 15,000 monthly visitors might only have 20–40 concurrent users at peak. Each PHP process needs 30–80 MB of RAM depending on what it's doing.

// Base RAM by stack type stackMultiplier = 1.0 (blog) to 2.0 (WooCommerce + membership) // Concurrent users estimate concurrentUsers = monthlyVisits / 500 // Per-process memory (MB) perProcess = 40 + (plugins × 2) // plugin overhead grows with count // Total before caching rawRam = (concurrentUsers × perProcess × stackMultiplier) / 1024 // Apply caching reduction finalRam = rawRam × cachingMultiplier

This gives a starting point. Real-world testing with htop or ps_mem.py on your actual server will always be more accurate. But this gets you within 1–2 GB, which is usually enough to choose the right plan.

Worked example: WooCommerce with 20 plugins

A client was running WooCommerce on a 2 GB VPS. The site was crashing every time they ran a sale — pages would hang, then the server would throw 502 errors. They had 22 active plugins and about 30,000 monthly visitors during normal periods.

  • Concurrent users: 30,000 / 500 = 60
  • Per-process memory: 40 + (22 × 2) = 84 MB
  • Raw RAM: (60 × 84 × 1.6) / 1024 = 7.9 GB
  • With page cache only (0.85): 6.7 GB

We moved them to a 8 GB VPS. The sale-day crashes stopped. They could have optimized plugins down to 10 and fit in 4 GB, but they didn't want to rebuild the site — so 8 GB was the honest requirement.

RAM by site type (quick reference)

Site typeMonthly visitorsPluginsRecommended RAM
Static HTML / portfolioAny0512 MB–1 GB
Lightweight WordPress blog<10,0005–101–2 GB
WordPress + page builder10,000–30,00010–202–4 GB
WooCommerce store10,000–50,00015–254–8 GB
WooCommerce + membership50,000+20+8–16 GB
Laravel / custom appVariesN/A2–8 GB

These match what we see running on real servers[citation:8]. If your site doesn't fit the pattern, the calculator above will give you a more specific answer.

Frequently asked questions

Is 2 GB RAM enough for WordPress?
2 GB is enough for a basic WordPress blog with under 10,000 monthly visitors and a handful of plugins. It is not enough for WooCommerce, a page builder, or 15+ plugins under any real traffic. You'll hit memory limits and 502 errors during spikes[citation:8].
Does more RAM make my site faster?
Only up to a point. If your site is slow because of CPU limits or database queries, more RAM won't help. But if it's slow because PHP is running out of memory and swapping to disk, adding RAM is the single biggest speed improvement you can make. Check free -m — if swap is being used, you need more RAM.
How many CPU cores do I need?
A basic WordPress site can run on 1–2 cores. WooCommerce or a site with high concurrency needs 2–4 cores[citation:8]. CPU and RAM work together: more cores without enough RAM doesn't help (each PHP process needs memory), and more RAM without enough cores means requests queue up.
Should I use object caching to reduce RAM needs?
Object caching (Redis or Memcached) reduces database load, which indirectly reduces RAM pressure on MySQL. But Redis itself uses RAM — usually 100–500 MB for a WordPress site. Net effect is usually positive: 500 MB for Redis can save you 1–2 GB of MySQL memory.
Can I upgrade RAM later?
On most VPS providers, yes — you can resize the plan and reboot. On some, it requires a migration. Check before you buy. If you're unsure whether you need 2 GB or 4 GB, start with 2 GB and upgrade when free -m shows you're consistently using more than 70% of available RAM.
MU
Mujahid — Runs cPanel, DirectAdmin, and CyberPanel stacks across 50+ client deployments. Every calculator on this site is checked against real server configurations.