What shipping and retaining logs costs per month.
Logging charges when data lands and every month it lingers. Sample the noise and tier old data to archive. For raw archive pricing see the egress and storage calculators.
The Log Ingestion Cost Calculator estimates your monthly logging bill from four inputs: how many gigabytes of logs you ingest per month, the price per gigabyte to ingest, the price per gigabyte to retain each month, and how many months you keep that data. It combines these into a single figure by multiplying your monthly log volume by the ingest rate, then adding retention charges, which accrue per gigabyte for every month the logs are stored. The two main drivers are therefore volume and retention length: doubling either roughly doubles the corresponding portion of the cost, and long retention windows can quietly dominate the total even when the per-gigabyte rate looks small.
This matters because logging spend is often billed on a per-gigabyte basis that scales invisibly with traffic, verbose debug output, or noisy services, so a bill can grow faster than expected. Running the numbers before committing to a plan lets you compare providers, test how a shorter retention window or reduced log volume changes the total, and decide whether to filter or sample logs at the source. The key trade-off to watch is ingest versus retention: keeping everything for a long time is convenient for audits and debugging, but it steadily multiplies retention charges, so it is usually cheaper to ingest selectively and retain only what you genuinely need.
Two meters: a per-GB ingest charge when logs arrive, and a per-GB-per-month retention charge for keeping them. Long retention on high-volume logs multiplies quickly β retention often costs more than ingest over time.
Sample or drop noisy debug logs, set sensible retention (hot for days, cold archive for the rest), and avoid logging full payloads. Most teams ingest far more than they ever query β that gap is pure waste.
Cost at different volumes:
| Volume | Monthly | Yearly |
|---|