Traditional agents are data forwarders that ship metrics to centralized systems. Each Netdata Agent is a complete observability engine with collection, storage, ML, alerts, and dashboards working autonomously at the edge. This distributed architecture eliminates centralized bottlenecks, reduces bandwidth costs, ensures data sovereignty, and scales linearly to any size. Plus, Netdata provides per-second granularity (10-60× more granular than competitors) and zero-configuration deployment.

Netdata Agents are extremely lightweight: <5% CPU and 150MB RAM (standalone installation, machine learning enabled). Machine learning and alerting can be offloaded to Parents. The University of Amsterdam study (ICSOC 2023) independently validated Netdata as the most energy-efficient monitoring solution with the lowest overhead - even while collecting per-second data and running ML at the edge.

Dashboards as soon as data is collected. One-line install. Dashboards appear immediately with all auto-discovered services. 800+ integrations work out-of-the-box with zero configuration. ML-based anomaly detection starts after about 15 minutes of data (no tuning required). All 18 models per metric are in place after about 54 hours.

Yes. Netdata runs natively on Linux (all major distributions), Windows (Server 2016/2019/2022/2025, Windows 10+), macOS (10.12+), and FreeBSD (11+). It monitors containers (Docker, Kubernetes, Podman, LXC), cloud platforms (AWS and Azure through their native monitoring APIs; Agents on any cloud, including GCP and OCI), and network devices via SNMP. The same agent software works across all platforms with platform-specific optimizations (e.g., systemd-journal on Linux, Windows Event Log on Windows).

Netdata trains 18 unsupervised k-means models per metric automatically at the edge - no configuration, no tuning, no thresholds to set. Models use different time windows (6-hour training with 3-hour staggered intervals) and require unanimous agreement before flagging anomalies, achieving 99% false positive reduction in anomaly detection. Models retrain every 3 hours, adapting continuously to infrastructure changes. The Anomaly Advisor correlation engine cuts through thousands of metrics to surface root cause in the top 30-50 results. Note: ML anomaly detection is a separate signal and does not directly influence alert notifications.

Yes. Each Netdata Agent keeps collecting, storing, detecting anomalies and alerting with no internet connectivity required, with local metrics dashboards; viewing logs and traces needs a Netdata Cloud sign-in. For centralized dashboards, logs and traces in air-gapped environments, deploy Netdata Cloud On-Premises within your datacenter. Agents can also stream to Netdata Parents for centralized storage and multi-node dashboards without external connectivity.

Netdata queries systemd-journal (Linux), Windows Event Log (Windows) and macOS unified logs in place with no pipeline or log shipping, and Logs Management collects logs from any other source through Netdata’s OpenTelemetry logs pipeline. No Elasticsearch clusters, no Splunk licenses, and System Logs Monitoring (querying logs in place) is not billed. Every field in every log entry is automatically indexed by the operating system - Netdata just makes it queryable at scale. This delivers 90% cost reduction versus traditional log management while fully evaluating the first 500,000 entries of a query, then sampling up to 1M (vs 5K for traditional tools). Learn more about Logs Management.

Netdata uses transparent per-node pricing with no charges for metrics volume, System Logs Monitoring (querying logs in place), users, data retention or containers. The open-source agent is free forever with community support. Netdata Cloud Business plan adds Netdata AI, all roles, enterprise SSO and SCIM, enterprise notification integrations and Netdata Cloud MCP; machine learning anomaly detection runs on every Agent. This delivers 90% cost reduction versus volume-based platforms like Datadog or New Relic. No surprise bills - costs scale linearly with infrastructure size.

Netdata embraces open standards and interoperability. It ingests metrics via Prometheus, OpenMetrics, StatsD, and OpenTelemetry (metrics, logs and traces). It exports to Prometheus Remote Write, InfluxDB, Graphite, OpenTSDB, and TimescaleDB. Native Grafana plugin enables visualization in existing workflows. For logs, Netdata queries standard systemd-journal, Windows Event Log and macOS unified logs in place and ingests standard OTLP through its OpenTelemetry logs pipeline - no proprietary lock-in. You can run Netdata alongside existing tools initially, then migrate gradually as ROI becomes clear.

Multiple support tiers: Community (free via GitHub Discussions, Discord, forums), Business (email/ticket support during business hours with SLA), and Enterprise (24/7 availability, dedicated support team, phone support). Professional services include implementation assistance, architecture design, migration support, and training programs. The 80,000+ star GitHub community and active Discord provide peer support. Learn more about support options.