Hosts, logs and APM — model a Datadog-style monitoring bill.
Per-host pricing punishes autoscaling and log ingestion is the silent budget-eater. Trim log volume first. Model the raw storage side on the log ingestion cost calculator and retention on the log retention calculator.
The Observability & Monitoring Cost Calculator estimates your total monthly monitoring bill by combining two separate charges. The first is host-based: your number of monitored hosts multiplied by the price per host per month. The second is log-based: your daily log volume in GB scaled across the month and multiplied by the cost per GB ingested and indexed. Adding an optional APM component, the tool shows how these pieces add up. The main drivers are straightforward — how many hosts you watch, your per-host rate, how many gigabytes of logs you generate each day, and the ingest-plus-index price you pay for each one.
The key trade-off to watch is log volume, because per-GB charges multiply across roughly thirty days, so a small daily increase compounds into a large monthly line item. Before scaling up host counts, filter or sample noisy debug logs and tighten retention — cutting daily gigabytes often lowers the bill more than reducing hosts.
Two reasons: host-based billing (every node running the agent counts, so autoscaling multiplies it) and log ingestion + indexing, often $1.50–2.00 per GB. A chatty app logging tens of GB a day quietly becomes the biggest line on the invoice.
Sample or filter logs before ingestion, index only what you query, cap the agent to real hosts (not every ephemeral pod), and move cold logs to cheap archive storage. Ingestion volume is the main lever.
Cost at different volumes:
| Volume | Monthly | Yearly |
|---|