How I cut my Datadog bill by 90% - the open-source observability stack I switched to

Описание и применение

Автор заменил платную систему Datadog на набор открытых инструментов (Prometheus, Grafana, Loki, SigNoz, Uptime Kuma) и сократил расходы с $6 026 до $31 в месяц, сохранив при этом видимость инфраструктуры, логов и трассировки.

Задача

High cost of Datadog for 40 hosts, especially metric pricing, leading to ~$6,000/month bill.

Результат

Reduced monthly cost from $6,026 to $31, a $5,995/month saving (~$72,000/year).

Рабочий процесс

• Set up Prometheus + Grafana for infra metrics (Week 1).
• Add Loki + Promtail for logs (Week 2).
• Migrate alerts to Alertmanager (Week 3).
• Add SigNoz for APM on key services (Week 4).
• Decommission Datadog (Week 5).

Доказательства

I was paying $1,800/month to Datadog for 40 hosts. Now I pay $0.

Источник

Total: ~$6,000/month

Источник

Setup: ~3 hours. Single VM, 2 GB RAM handles my 40 hosts comfortably.

Источник

Setup: ~4 hours including Promtail config for journald, docker logs, and a few specific app files.

Источник

Setup: ~2 hours (the OpenTelemetry SDK changes were the most painful part).

Источник

What I gave up: Auto-instrumentation for some libraries; Real-time log streaming; Some managed integrations.

Источник

What I gained: No per-metric pricing; Predictable cost; My data stays on my servers; Custom queries.

Источник

About 30 hours total over 5 weeks.

Источник