Choosing the right resource package
How pooled RAM, CPU and disk packages work on TheCodexApps, what each size comfortably runs, when to upgrade, and how to watch your usage so you never pay for idle headroom.
TheCodexApps Team
Most hosting bills people remember are per-app: every service gets its own plan, and every plan charges you for headroom you only use on the busiest day of the month. TheCodexApps does it differently: you buy a pool of resources and run as many apps as fit inside it. This post explains how the pools work, what each size comfortably runs, and how to decide when it is time to move up a step.
How pooled packages work
A package is a fixed amount of RAM, CPU and disk. Every app, database and one-click service you deploy draws from that shared pool, and the panel shows you exactly how much each service uses. There is no per-app fee and no artificial limit on the number of apps — the honest limit is the resources themselves.
The current lineup:
- 1 GB — $6/month — a serious starter: one small app plus a tiny database.
- 2 GB — $11/month — the sweet spot for a small production stack.
- 4 GB — $20/month — several production apps or a heavier database workload.
- 8 GB — $36/month — a full small-business stack with room to experiment.
Yearly billing takes roughly two months off the price. Upgrades are prorated from the wallet balance, and downgrades are just as easy when a project ends.
What fits in each size
Sizing is easier with concrete examples. Based on how real workloads behave:
1 GB
A static site or small API, a lightweight service like Uptime Kuma, or a single small app with SQLite. Fine for hobbies and internal tools with one or two users.
2 GB
A production Node.js or Next.js app, a small Postgres or MySQL instance, or n8n with a handful of workflows. This is where most side projects and early startups live comfortably.
4 GB
A small product stack: the app, a Postgres with a few gigabytes of data, a Redis for queues, and a status page — all in one pool. Also the right size for WordPress sites with real traffic.
8 GB
Multiple production stacks, a database with heavier queries, or a team that wants a staging environment running next to production without micromanaging resources.
When to upgrade
The panel's meters make this visible, but the signs are usually the same everywhere:
- memory usage sits above 80% for days, not minutes;
- deploys get slower because builds compete with the running app for memory;
- a database starts caching less because the working set no longer fits;
- you catch yourself stopping one service to run another.
None of these are emergencies — they are the signal that the pool is genuinely full. Because upgrades are prorated, moving up mid-month costs only the difference.
Monitoring usage
Every service page shows live RAM, CPU and disk usage, and the package page shows the pool as a whole. A practical routine: check the meters once a week for the first month, note the peak, and size for roughly 30% headroom above it. Peaks matter more than averages — a database that spikes during a nightly backup needs room for the spike, not just the baseline.
Right-sizing tips
A few habits keep the smallest possible package honest:
- Start one size below what you think you need; upgrading takes minutes, downgrading is a chore you will postpone.
- Give each database a soft disk limit so one runaway log cannot eat the pool.
- Put a low-traffic marketing site and a high-traffic app in the same pool — the site costs almost nothing and smooths your utilization.
- Re-check your package after deleting a big experiment; unused resources are just invisible cost.
The bottom line
Pooled packages reward you for consolidating: the more you run in one place, the better the price per service gets. Start at 1 or 2 GB, watch the meters for a couple of weeks, and let the graphs — not a guess — tell you when to move up.