Logging calculators

Model ingestion + retention first, then add query/scan and metrics cardinality where it matters.

A practical model

  1. Estimate GB/day ingested and retention days.
  2. Convert to GB-month stored and apply $/GB-month.
  3. Add scan/query cost if you run frequent or wide queries.
  4. For metrics, model active series (cardinality) and retention.

What spikes bills

  • High-cardinality labels (metrics) and noisy debug logs.
  • Wide queries that scan large time ranges or many log streams.
  • Incident retries that multiply log volume without increasing useful signal.

How to get your inputs

  • Ingestion GB/day: use log platform metrics or billing exports by log group.
  • Retention days: check policies for each log group or bucket.
  • Scan GB: query history or dashboards (range length x data scanned).
  • Metrics series: active series counts and cardinality reports.
12 logging tools

Result interpretation

  • If ingestion dominates, focus on log volume controls before tuning pricing tiers.
  • If query costs dominate, add guardrails: shorter ranges, fewer streams, and sampled queries.
  • Retention changes often have a bigger impact than storage class tweaks.
  • Use a peak scenario; incident bursts and retries are the real cost multipliers.

Scenario planning

  • Baseline: average GB/day ingested and normal query frequency.
  • Peak: 2x-3x ingestion for incident months and wider query windows.
  • Retention shift: model shorter retention as a separate scenario.
  • Cardinality growth: add a growth rate for metrics series counts.

Validate after changes

  • Compare ingestion GB/day to billing usage types after policy changes.
  • Check retention policies for each log group or bucket.
  • Review query scan volume after dashboard or alert updates.
Advertisement