Best Datadog Alternatives in 2026: Open-Source and Hosted Monitoring Tools

Teams replacing Datadog usually take one of three routes. SigNoz is a single open-source platform for metrics, logs and traces, Grafana with Prometheus is a self-hosted stack you assemble yourself, and New Relic is a hosted full-stack product for teams that would rather not run anything.

Datadog is several products under one login, so this guide sorts twelve Datadog alternatives by the job they do: all-in-one observability, metrics and dashboards, logs, tracing, error tracking and uptime. The table below shows live licence, self-hosting, pricing and GitHub data for the main picks.

Why people look for Datadog alternatives

Datadog is a cloud monitoring and observability platform that brings metrics, logs and traces from applications and services into one place, with dashboards, alerting and a large catalogue of integrations. Teams usually start comparing it with other tools for these reasons:

• Pricing has several dimensions. Datadog's pricing page prices each product separately. Infrastructure and APM are charged per host, log management per ingested or scanned gigabyte, and some tools per committer. The bill follows how many hosts you monitor and how much data you send.
• It is closed source and hosted only. Enlisted lists no self-hosted version, which rules it out where telemetry must stay on your own infrastructure.
• The free tier is small. Datadog lists a free tier limited to a handful of hosts for infrastructure monitoring, plus trials for other products, so a team monitoring more than a few hosts needs a paid plan.
• Instrumentation lock-in. OpenTelemetry is a vendor-neutral way to emit traces, metrics and logs, and several tools below are built on it. Choosing one makes it easier to change backends later.
• You may only need one job. If the real pain is error tracking or uptime checks, a focused tool can be simpler than a platform that does everything.

How we picked

A tool made this list if it replaces at least one job Datadog covers: metrics and dashboards, logs, distributed tracing, error tracking or uptime monitoring. Open-source projects needed a clear licence and recent GitHub activity, and Enlisted notes whether each can be self-hosted. Hosted products are included for their published pricing model and free tier. The write-ups draw on catalogue data and vendor documentation, and none of it is first-hand testing. See how Enlisted ranks tools.

Which job do you need to replace?

Observability is usually split into five jobs, and many open-source tools do only one of them well:

• Metrics and dashboards: numbers over time from servers, containers and applications, with alert rules. Prometheus, Grafana and Netdata live here.
• Logs: searchable text from applications and infrastructure. Grafana Loki and OpenObserve are log stores.
• Tracing and APM: following one request across services to find slow calls. Jaeger is a tracing system, and SigNoz covers APM.
• Error tracking: grouped exceptions with stack traces and release context. Sentry is one example.
• Uptime monitoring: checks from outside that a site or API responds, with status pages. Uptime Kuma fits here.

The all-in-one tools (SigNoz, OpenObserve, HyperDX, New Relic and Better Stack) cover several of these together, which is closest to the way Datadog is sold.

1. SigNoz: best open-source all-in-one for APM, logs and traces

SigNoz is an open-source observability platform built on OpenTelemetry that unifies APM, logs, metrics, traces, alerts and dashboards in one tool, self-hosted or in the cloud. That single interface is the main difference from stacks made of separate projects: Datadog sells one product across these jobs, and SigNoz takes the same single-product approach in open source.

It shows latency, error rate and Apdex for services, offers a visual query builder for logs, and builds dashboards from a query builder, PromQL or SQL. It also covers Kubernetes infrastructure monitoring and LLM and AI application observability. The self-hosted Community edition is free, with a paid cloud that bills by usage. Enlisted lists its licence as Other, so read the repository's licence files before you commit.

• Best for: teams replacing Datadog APM, logs and metrics together who already use or plan to adopt OpenTelemetry.
• Watch out for: if you self-host, you operate and scale the platform yourself, including storage for telemetry.

2. Grafana: best dashboards over data you already collect

Grafana is an open-source platform for visualizing metrics, logs and traces from many data sources, with dashboards, alerting and ad-hoc exploration. The open-source Grafana is a visualization layer that queries data held elsewhere, so it replaces the dashboards and alerting side of Datadog and relies on other systems underneath.

Dashboards support template variables, one graph can mix data sources, and alert rules can notify Slack or PagerDuty. It connects to Prometheus, Loki and SQL databases. Grafana Cloud has an always-free tier with usage limits, and self-hosting is possible under AGPL-3.0.

• Best for: teams that already have metrics in Prometheus or logs in Loki and want one place to see them.
• Watch out for: it needs data sources behind it, so a Grafana-based stack means choosing and running several components.

3. Prometheus: best open-source metrics and alerting for Kubernetes

Prometheus is a metrics monitoring system and time series database that scrapes targets, evaluates alert rules and offers the PromQL query language. It is a Cloud Native Computing Foundation project licensed under Apache-2.0, and its pull-based collection with service discovery makes it a common fit for container workloads.

Compared with Datadog Infrastructure, you gain control and lose convenience: there is no hosted version in Enlisted's data, and you decide how long to keep data and how to scale storage. VictoriaMetrics is a Prometheus-compatible database that teams use for long-term storage.

• Best for: engineering teams running Kubernetes or services that expose metrics endpoints.
• Watch out for: it handles metrics only, so logs, traces and dashboards come from other tools such as Loki, Jaeger and Grafana.

4. Netdata: best for quick per-second infrastructure monitoring

Netdata is an open-source infrastructure monitoring platform that collects per-second metrics with little configuration and adds machine-learning anomaly detection and alerting. Its agent is described as zero-configuration, with per-second charts for servers and containers, and distributed storage keeps data local.

It monitors Linux servers, containers, Docker, Kubernetes and databases, and is licensed under GPL-3.0. The agent is free, with a free community tier and paid plans in the cloud for more nodes.

• Best for: small ops teams and individuals who want fast, detailed infrastructure visibility without building a stack.
• Watch out for: it centres on infrastructure metrics, so application tracing and error tracking need separate tools.

5. OpenObserve: best for lower-cost log, metric and trace storage

OpenObserve is a cloud-native observability platform covering logs, metrics, traces, real user monitoring and LLM monitoring, with Parquet columnar storage on S3. Its own pitch is against Datadog and Splunk.

It is compatible with OpenTelemetry and Prometheus, supports pipelines and SLO tracking, and is licensed under AGPL-3.0. Self-hosting is free, and the cloud offers pay-as-you-go pricing by data ingested, with unlimited users.

• Best for: teams whose main cost is log volume and who want a Datadog or Splunk replacement for centralised search.
• Watch out for: test it against your own query patterns and retention needs before moving all of your telemetry.

6. HyperDX: best ClickHouse-based tool for debugging with logs, traces and replays

HyperDX is an open-source observability platform that unifies logs, traces, metrics and session replays on top of ClickHouse and OpenTelemetry, described as a Kibana-style experience for ClickHouse. It targets the moment of an incident: searching across schemas, tailing logs live and comparing events.

Search syntax covers full text and properties, alerts and event delta analysis are included, and dashboards handle high-cardinality events. It is MIT-licensed, available self-hosted or in a cloud with volume-based pricing and no per-seat or per-host billing.

• Best for: teams that already use or want ClickHouse and care about debugging production incidents quickly.
• Watch out for: its feature list centres on logs, traces, metrics and replays, so check that the infrastructure monitoring you rely on is covered before replacing Datadog entirely.

7. Grafana Loki: best for logs that sit next to Grafana

Grafana Loki is a horizontally scalable log aggregation system that indexes labels instead of log contents, inspired by Prometheus. That is a different model from full-text log engines, and it will feel familiar to teams that already label their metrics in Prometheus.

It indexes Kubernetes pod labels automatically, collects with the Grafana Alloy agent, supports multi-tenant setups and can run as a single binary. It is AGPL-3.0 licensed.

• Best for: Kubernetes teams that want logs in the same Grafana views as their metrics.
• Watch out for: queries that depend on searching log contents behave differently from a full-text index, so plan your labels up front.

8. Jaeger: best for open-source distributed tracing

Jaeger is a distributed tracing platform, originally from Uber and now a graduated CNCF project, for monitoring and troubleshooting microservices. It answers questions such as which service made a request slow, which is the tracing part of what Datadog APM does, and it is built on OpenTelemetry Collector components.

There is a Docker quick start for local use, and it is licensed under Apache-2.0. Teams often send traces from OpenTelemetry into Jaeger and view metrics elsewhere. Pinpoint is a related open-source APM for Java, PHP and Python apps, with topology maps.

• Best for: microservice teams that need request tracing without adopting a whole platform.
• Watch out for: it traces requests and nothing else, so metrics, logs and alerting live elsewhere.

9. Sentry: best for error tracking and performance issues

Sentry is a developer-focused platform for error tracking and performance monitoring, with SDKs for most programming languages, available self-hosted or as a hosted service. If your Datadog use is mostly finding exceptions after a release, a focused tool like this is often easier to adopt than a full observability platform.

It groups issues with debugging context, tracks crashes in web, mobile and game builds and finds slow transactions. A free Developer plan covers a single user, and paid plans bill with unlimited users plus usage. The catalogue lists the licence as Other. The hosted product has its own Sentry listing, and the Sentry alternatives page lists tools on the other side.

• Best for: application teams that mostly need error tracking, crash reporting and release health.
• Watch out for: it is not an infrastructure monitor, so you still need a metrics tool for servers and containers.

10. Uptime Kuma: best self-hosted uptime monitoring and status pages

Uptime Kuma is a self-hosted monitoring tool that checks websites, ports, DNS and containers, sends alerts through many notification services and publishes status pages. It covers the uptime-check job and is MIT licensed.

Monitor types include HTTP, TCP, DNS, ping and Docker containers, with multiple status pages on custom domains and certificate details. It has no vendor-hosted version, so you run it yourself. Hosted alternatives for the same job include UptimeRobot, which has a free plan, and Hyperping.

• Best for: homelabs, small server fleets and products that need a public status page.
• Watch out for: monitoring from the same network as your servers will not notice an outage that cuts you off, so consider a second location or an external service.

11. New Relic: best hosted full-stack alternative

New Relic is an observability platform that correlates application, infrastructure, log and user-experience data to find root causes. It is a hosted platform in the same category as Datadog: closed source, SaaS only, and covering APM, infrastructure and Kubernetes monitoring, log management and OpenTelemetry support.

It has a free tier with a data allowance and one full platform user. Paid editions charge per user and by data ingested, so the cost drivers differ from Datadog's host-based model. The New Relic alternatives page lists more routes.

• Best for: teams that want a hosted platform with a free starting point and a user-based pricing model.
• Watch out for: no self-hosting, and per-user charges add up for large teams.

12. Better Stack: best hosted all-in-one for small teams

Better Stack is an observability and incident platform that bundles uptime monitoring, on-call alerts, status pages, logs, tracing and error tracking in one product. Enlisted lists migrating off Datadog as one of its use cases, and it suits teams that also want incident response in the same place.

It uses OpenTelemetry-native tracing with eBPF, real user monitoring and AI-assisted root-cause analysis. A free plan covers personal projects, and paid pricing combines responder licences with telemetry and extra status pages.

• Best for: small and mid-size teams that want uptime, logs and on-call in one hosted tool.
• Watch out for: pricing has several parts, so estimate responders and telemetry volume together. There is no self-hosted version. See the Better Stack alternatives page for comparisons.

Other options worth a look

• Enterprise suites: Dynatrace offers APM, log analytics and AI-based root-cause analysis with host-based pricing, and Splunk is a data platform for security and observability. Both are hosted commercial products; see the Dynatrace alternatives and Splunk alternatives pages.
• Classic infrastructure monitors: Zabbix, Checkmk, Icinga 2 and Nagios Core are long-running open-source monitors for servers and networks.
• Log storage: Quickwit indexes logs and traces on object storage with an Elasticsearch-compatible API, and Kibana visualises data in Elasticsearch.
• Profiling: Pyroscope is continuous profiling that shows where code spends CPU and memory.
• Errors with replay: Highlight.io combines session replay, error monitoring, logging and tracing, and Bugsnag is a hosted error and stability monitor.
• User behaviour: teams that want replay and product analytics rather than infrastructure data should read the Google Analytics alternatives guide.

Which Datadog alternative should you choose?

Browse every Datadog alternative in the catalogue, or the wider monitoring category.

Self-hosting observability: what you take on

Moving off Datadog changes where the cost sits. The vendor bill shrinks, but you now run storage for metrics, logs and traces, decide retention, upgrade the platform and respond when the monitoring tool itself fails. Teams with platform engineers often find that trade worthwhile; a very small team may not.

Two habits reduce the risk. Instrument with OpenTelemetry so your code is not tied to one backend, and keep your monitoring outside the infrastructure it watches. If you want open-source software without the operating work, several tools above sell a hosted version: SigNoz, Grafana, OpenObserve, HyperDX, Sentry and Netdata all do.