Calculator
Formula used
| Raw ingest (GB/day) | metrics + logs + traces |
|---|---|
| Effective ingest (GB/day) | raw_daily × (1 + cardinality_overhead) × sampling |
| Billable ingest (GB/month) | max(0, effective_daily × days_in_month − free_tier) |
| Ingest cost | billable_ingest × ingest_rate |
| Resident storage (GB-month) | (effective_daily × (retention_days / days_in_month) ÷ compression) × (1 + index_overhead) |
| Storage cost | resident_storage × storage_rate |
| Scanned per month (GB) | queries_per_day × scan_per_query × days_in_month |
| Query cost | monthly_scanned × query_rate |
| Egress cost | egress_gb_month × egress_rate |
| Total cost | ingest_cost + storage_cost + query_cost + egress_cost |
How to use this calculator
- Enter daily volumes for metrics, logs, and traces in GB.
- Set retention days, sampling percent, and compression ratio.
- Adjust overheads to reflect label growth and indexing strategy.
- Provide your ingestion, storage, query-scan, and egress rates.
- Click Calculate cost to see totals and breakdown.
- Use Download CSV or Download PDF for sharing.
Example data table
Sample inputs and outputs for quick validation.
| Scenario | Raw ingest (GB/day) | Retention (days) | Sampling (%) | Total monthly cost |
|---|---|---|---|---|
| Starter stack | 26.5 | 30 | 100 | USD 0.00–250.00 |
| Scale-up team | 85.0 | 45 | 50 | USD 300.00–1,200.00 |
| Enterprise program | 450.0 | 90 | 30 | USD 3,000.00–12,000.00 |
Operational cost context
Ingestion drivers and volume baselines
Daily telemetry volume is the most controllable input, and small upstream changes compound quickly. For example, a 10 GB/day increase adds roughly 304 GB/month on a 30.4‑day period before overheads. If enrichment adds 12% overhead and you keep full sampling, that becomes 340 GB/month billable in this model. Treat baselines as measured averages from a representative week, not peak days.
Retention, compression, and resident storage
Retention affects storage linearly, but the stored footprint is reduced by compression and increased by indexing. With 30‑day retention and 3× compression, 30 GB/day effective ingest yields about 29.6 GB‑month of resident data before index overhead. If indexes add 20%, the resident figure becomes 35.5 GB‑month. Use separate retention policies for logs versus traces to match incident response needs.
Query scanning as a hidden multiplier
Query cost often surprises teams because it scales with scanned bytes, not returned rows. A modest 90 queries/day scanning 1.2 GB each is 3,283 GB scanned per month. At 0.002 per scanned GB, that is about 6.57 in query charges, before adding dashboards, ad‑hoc investigations, and automated alerts. Reducing scan size through filters, partitions, and time bounds is usually the fastest win.
Overhead control: labels, metadata, and indexes
High‑cardinality labels, verbose attributes, and aggressive indexing inflate both ingest and storage. If overhead rises from 12% to 30%, the same raw stream becomes 16% more expensive after sampling and free tiers. Index overhead is similar: increasing searchable fields can add 10–60% storage depending on the platform. Use allowlists for labels, cap attribute lengths, and index only fields with repeated analytical value.
Budgeting, governance, and change planning
Use this calculator to build three scenarios: baseline, growth, and incident surge. A practical growth assumption is 5–15% monthly telemetry increase for expanding services, plus step changes when adding new environments. Tie budgets to service ownership by tagging streams and tracking per‑team effective GB. When costs exceed targets, prioritize sampling for traces, shorter log retention, and query discipline before renegotiating rates. Review costs weekly during major launches, and set guardrails such as daily ingest limits, retention caps, and alerting on scan spikes. Document assumptions so budget owners can explain variance clearly.
FAQs
1) What does “effective ingest” mean?
It is your raw telemetry after overhead and sampling. This is the volume used to estimate ingestion and storage usage in the calculator.
2) Why does retention change storage but not ingestion?
Ingestion is billed when data arrives. Retention controls how long data remains stored, so it changes resident GB‑months and storage cost.
3) Should I set sampling below 100% for everything?
Not always. Traces often benefit most from sampling, while security or audit logs may need full fidelity. Use different policies per telemetry type when possible.
4) How can I estimate “GB scanned per query”?
Use platform query insights or billing reports that show scanned bytes. Start with an average from recent dashboards and incident investigations, then refine monthly.
5) What do overhead percentages represent?
Cardinality overhead models metadata expansion during ingest. Index overhead models additional stored structures that enable fast search and aggregation.
6) Will vendor bills match this exactly?
No. Billing can vary by tier, feature, region, and rounding. Use this as a consistent planning model, then calibrate rates using your actual invoices.