How to Work Within Memory and Process Limits for Node.js on Shared Hosting

Shared hosting gives each account limited memory and a limited number of processes, so a lean, efficient Node.js app runs happily while a heavy one gets throttled or stopped. The trick is building within those limits — and recognising honestly when your app has outgrown them.

Understand your limits

Your plan has caps on memory and concurrent processes (and often CPU). Exceeding them can slow or stop your app. If you are unsure of your limits, check your plan details or ask via a support ticket. Understanding resource usage helps.

Keep memory use low

• Stream large data instead of loading it all into memory — see Node.js streams.
• Process in batches rather than huge arrays at once.
• Bound your caches so they cannot grow without limit.
• Fix memory leaks, which slowly push you over the limit.

Keep processes lean

On shared hosting, do not try to run your own clusters or many background processes — Passenger manages your app, and spawning extra processes competes for a limited pool. Keep to a single, efficient web app.

Offload heavy work

Move demanding tasks off the shared server: use a CDN for static assets, a cloud database where relevant, and build front-ends locally rather than on the server.

Knowing when you've outgrown shared hosting

If your app needs always-on background processing, clustering across cores, high memory, or native builds the server will not allow, you have outgrown shared hosting. That is a normal milestone — the next step is a VPS (or Webuzo) where you control the resources. See shared vs VPS vs dedicated.

Frequently asked questions

How much memory does my Node app get on shared hosting?

It depends on your plan's limits. Check your plan details or ask support. The practical answer is to keep your app lean and avoid loading large data into memory.

Can I run clustering on shared hosting?

Not really — process limits and Passenger's management make manual clustering impractical. Clustering belongs on a VPS or Webuzo where you control processes with PM2.

How do I know it's time to move to a VPS?

When you consistently hit memory or process limits, need always-on background work, or require software the shared server will not allow. Those are clear signs your app needs a VPS.