Projects

Usage & Cost

Understand how a Skytells project consumes resources, review cost breakdowns, and identify optimizations to keep your project efficient.

The Usage page shows you the resource footprint of a single project — how much compute, storage, and network it is consuming, what it is costing, and which services are the biggest contributors.

What Usage Tracks

Compute

CPU and memory consumption for every running app and database in the project. Aggregated at the project level with per-service breakdowns so you can identify which container is the main driver.

Storage

Persistent storage consumed by databases and app volumes. Tracks both current usage and growth over the selected time period.

Network

Inbound and outbound traffic for the project. Outbound traffic from your services to the internet is where egress costs typically accumulate.

Cost Breakdown

A line-item breakdown of what each resource type is contributing to the total project cost. Helps you separate compute costs from storage and network costs.

How to Read the Usage View

The usage page presents data for a selected time range — use the period selector to view the last 24 hours, 7 days, 30 days, or a custom range.

The top of the page shows total project cost for the period. Below it, the breakdown splits cost across:

  • Apps — each app's compute and network contribution.
  • Databases — each database's compute and storage contribution.
  • Shared infrastructure — project-level networking and reverse proxy overhead.

Click any service in the breakdown to jump to that service's own monitoring or settings page.

Comparing Projects

The usage page shows data for one project at a time. To compare resource consumption across multiple projects:

  • Open the account-level usage view from the Console navigation.
  • Each project appears as a row with its total cost and primary resource drivers.
  • Use this to identify which project is consuming the most resources or growing fastest.

Identifying Cost Drivers

Common patterns that cause unexpected usage spikes:

Over-provisioned services — a worker or background job running with excess CPU and memory reservations that it rarely uses. Check per-service metrics and reduce the instance size if utilization is consistently low.

Idle development apps — apps left running in a project used for testing or development. Pause or remove services you are not actively using.

Excessive outbound traffic — a service sending large payloads (images, files, AI outputs) directly to clients rather than through a CDN. Consider offloading asset delivery.

Database growth — a database accumulating data over time without a retention policy. Review whether old records or logs stored in the database need pruning.

Plan limits and quotas are enforced at the account level, not the project level. A single large project can consume a significant portion of your account quota. Monitor usage regularly if you are on a plan with compute or bandwidth caps.

How is this guide?

On this page