Each charger runs a lightweight Netdata Agent that collects, stores, and analyzes metrics locally at per-second granularity. Agents stream a continuous metrics feed to regional Netdata Parents for aggregation and longer retention. Netdata Cloud receives metadata only and runs distributed queries on demand, so bulk telemetry stays at the edge. Learn more about edge monitoring.

Netdata Agents continue collecting and storing per-second data locally during network partitions. When connectivity returns, buffered data is automatically streamed to the regional Parent, filling any gaps. There is no data loss and no blind period during WAN or cellular outages.

Netdata monitors the Linux controller that runs the charging-management software, including the process health, resource usage, network connectivity, and OS-level logs of that process. Netdata does not decode OCPP or other application-layer charging protocols directly.

The Netdata Agent typically uses less than 5% of a single CPU core and 150-200 MB of RAM with ML enabled. Agents can be further reduced by offloading ML training, alerting, and storage to a regional Parent, dropping to under 2% CPU with zero disk I/O.

Yes. Netdata Functions surface processes, network connections, systemd units, disk usage, and other runtime diagnostics directly in the browser dashboard. Operators get the same information as console tools like top, ss, and journalctl, with full history and anomaly context.

Netdata uses simple per-node pricing that scales predictably with fleet size. Agents are free and open source. There are no charges for metrics volume, Logs Monitoring (querying logs in place), users, data retention or containers. P90 billing protection excludes temporary spikes from your bill. See the pricing page for details.

The Netdata Agent installs in seconds with a single command and begins auto-discovering metrics immediately. Dashboards, ML models, and 400+ pre-configured alerts activate automatically. A full fleet deployment with regional Parents and Netdata Cloud is typically completed in days, not months.

Yes. Netdata queries systemd-journal in place on Linux charger controllers, with full field indexing, histogram visualization, and full-text search, and no separate log pipeline, ingestion cluster, or storage backend. Logs Management collects application log files and other sources through Netdata’s OpenTelemetry logs pipeline and stores them with every field indexed. Learn more about zero-pipeline logs.

Yes. Per-second metrics and ML anomaly detection surface early warning signs such as memory saturation, storage exhaustion, process instability, and backhaul degradation. Component-level alerts monitor each instance individually, so failures are caught before they cascade into an offline charger.

Netdata Parents are aggregation nodes placed near your fleet that store replicated metrics, run queries, and provide high availability. Netdata Cloud is the control plane that unifies dashboards, alerting, and access control across all Parents and Agents. Parents hold the data; Cloud coordinates the view. Read about the architecture.

Yes. Netdata monitors any Linux, Windows, macOS, or FreeBSD endpoint. A single deployment can cover charger controllers, site gateways, depot servers, back-office systems, and network devices through SNMP. Every node gets the same per-second granularity and anomaly detection.

Centralized SaaS platforms require shipping all telemetry to a remote backend, which becomes prohibitively expensive at per-second granularity across thousands of endpoints. Netdata processes and stores data at the edge, keeping costs predictable and tied to node count. The architecture is specifically designed for distributed fleets where connectivity is unreliable and scale is large. Explore our edge fleet monitoring solutions.

Yes. Netdata supports Model Context Protocol (MCP) on Agents, Parents, and Cloud, so you can connect your preferred LLM (Claude, ChatGPT, Gemini, and others) to query infrastructure data and run AI-assisted root cause analysis. Learn about MCP.