Open source alternatives to New Relic

What New Relic does

New Relic is a software-as-a-service observability platform. Language agents, infrastructure agents and OpenTelemetry send telemetry to a single data store, where APM, infrastructure, browser and mobile monitoring, logs, synthetic checks and error tracking share the same data, queried with the NRQL language and shown in dashboards and alerts. It accepts OpenTelemetry data through an OTLP endpoint. There is no self-hosted edition; organizations choose the US, EU or Japan region.

Why people look for an alternative

• Billing on data and users. New Relic bills on gigabytes ingested and on the number of billable users. Basic users are free but cannot use curated experiences such as the APM interface, so giving more engineers full access raises the bill. A perpetual free tier includes a monthly ingest allowance and one full platform user.
• Short default retention. With the default Original Data option, APM, APM error and browser data are kept for 8 days. The Data Plus option adds up to 90 days and compliance features at extra cost.
• Region fixed at creation. New Relic does not support moving or sharing customer data from existing accounts across regions, and some account data such as billing may also be stored in the US.
• Lock-in. Dashboards and alerts written in NRQL, and data collected with New Relic’s own agents, have to be rebuilt elsewhere.

What to compare in an alternative

1. APM depth. Distributed traces, service maps, error grouping, database and external call views, and how the tool links traces to logs.
2. Instrumentation. If you already use OpenTelemetry, switching is mostly a collector change; if you use New Relic agents, check the migration guides and the languages covered.
3. Query language. SQL, PromQL or a visual builder in place of NRQL, and whether dashboards can be converted.
4. Self-hosting cost. Most APM backends here need ClickHouse; check the documented hardware for your data volume.
5. Access control. SSO and fine-grained roles are often in paid editions, which matters when you replace a per-user license.

Which one to choose

SigNoz is the most direct replacement: it is built on OpenTelemetry, keeps traces, logs, metrics, exceptions and alerts in one interface on ClickHouse, and publishes a migration guide from New Relic. Its self-hosted community edition has no per-user charges, but SSO and RBAC are paid. Coroot instruments services with an eBPF node agent instead of language agents, so you get a service map, traces and profiles without code changes; it fits containerized and Kubernetes environments best. OneUptime covers APM, logs and traces through OpenTelemetry and adds uptime checks, status pages, incident management and on-call scheduling in the same product, at the cost of a heavier install.

For infrastructure metrics without tracing, Prometheus with Grafana is the common open source stack. The monitoring guide compares all the options.