← all SaaS

Self-host New Relic

APM / observability platform · Category: observability & analytics

New Relic is a hosted observability suite — APM traces, infrastructure metrics, logs, browser RUM, synthetics — billed per ingested GB. The self-hostable replacements all align around OpenTelemetry: ship OTLP from your apps to one of these backends, get traces + metrics + logs in one UI without per-GB pricing.

New Relic pricing anchor: Free 100GB/mo + 1 user; Standard $0.30/GB ingested; Enterprise from $0.50/GB.

SigNoz SigNoz/signoz alive

GitHub
★ 26.9k · last commit 1d ago · 1531 open issues
License
MIT
Core is MIT; some Enterprise features (SSO, RBAC) are gated.
Setup time
30min docker-compose (SigNoz + ClickHouse + Zookeeper + OTel collector)
Monthly cost
$30-60/mo VPS for production traffic — ClickHouse + collector are the compute load.
Migration sketch. Use the official compose at signoz.io/docs/install/docker. Repoint your apps' OTLP exporter from the New Relic OTLP endpoint to your SigNoz collector (`OTEL_EXPORTER_OTLP_ENDPOINT=https://signoz.example.com:4318`). Traces, metrics, and logs all show up under one app namespace; SigNoz's UI is the closest spiritual match to New Relic APM with services view, error tracking, and SLO dashboards.
Good fit forTeams running OpenTelemetry-instrumented apps that want a single product (not Grafana + Tempo + Loki + Mimir glued together).
Weak atSynthetic monitoring and browser RUM are thinner than New Relic's; bring Uptime Kuma + a separate RUM tool.

OpenObserve openobserve/openobserve alive

GitHub
★ 18.8k · last commit today · 544 open issues
License
AGPL-3.0
Setup time
15min single-binary or docker run
Monthly cost
$15-40/mo VPS; storage backed by S3-compatible object storage keeps cost predictable.
Migration sketch. Run `docker run -d -v $HOME/openobserve:/data -p 5080:5080 -e ZO_ROOT_USER_EMAIL=admin@example.com -e ZO_ROOT_USER_PASSWORD=admin public.ecr.aws/zinclabs/openobserve:latest`. Point OTLP from your apps at the OpenObserve collector endpoint; logs, traces, metrics ingest through the same OTel pipeline. UI focuses on log search + trace correlation; Lucene-style query language.
Good fit forLog-heavy stacks moving off Datadog/New Relic that want predictable S3-backed storage costs.
Weak atAPM/services view is less polished than SigNoz's; better at logs than at deep distributed-trace exploration.

Grafana stack (Tempo + Loki + Mimir + Grafana) grafana/grafana alive

GitHub
★ 73.7k · last commit 1d ago · 3663 open issues
License
AGPL-3.0
Grafana, Loki, Tempo, Mimir all moved to AGPL-3.0 in 2021; Enterprise versions remain commercial.
Setup time
1h docker-compose for the full LGTM stack
Monthly cost
$30-100/mo VPS depending on retention; S3-backed for traces and logs.
Migration sketch. Use the `docker-otel-lgtm` reference stack from grafana.com/blog. Apps export OTLP to Grafana Alloy (the collector); Alloy fans out traces to Tempo, logs to Loki, metrics to Mimir. Build dashboards in Grafana — the New Relic NRQL queries don't port, but standard PromQL + LogQL + TraceQL gets you to similar visualizations.
Good fit forTeams that already have Grafana expertise and want best-of-breed components rather than a single product.
Weak atOperational complexity — four components to run instead of one; gluing dashboards takes a week vs SigNoz's defaults.

In a terminal? npx os-alt new-relic prints this table — how the CLI works →