Meet Every Compliance Mandate Without Sacrificing Visibility

Every component runs within your infrastructure. The control plane (Netdata Cloud) operates in your Kubernetes cluster. All Agents and Parents store metrics and logs locally on your servers. Zero data ever leaves your premises. This isn’t a feature - it’s our fundamental architecture. We can prove it: our control plane queries your infrastructure in real-time, streaming results without persistence.

A Kubernetes cluster (version 1.23+) with sufficient resources. For 2,000 nodes: 20+ CPU cores, 45+ GB RAM, 500+ GB storage. For 10,000 nodes: 100+ cores, 200+ GB RAM, 2+ TB storage. You’ll need image mirroring capability for air-gapped deployments. We provide Helm charts with pre-configured RBAC, network policies, and security contexts. Deployment typically takes 2-3 weeks including full Agent rollout across infrastructure.

Yes. Mirror images from Netdata ECR to your internal registry. Deploy Helm charts within your isolated network. Configure Agents and Parents to connect only to on-premises Cloud. No external connectivity required. We support data diodes for unidirectional telemetry flow. Complete feature parity with SaaS including SSO, RBAC, ML, and AI (bring your own LLM via Model Context Protocol).

Enterprise On-Premises starts at 200 nodes with custom pricing based on scale. Contact our sales team for detailed quotes. Pricing includes all features: unlimited metrics, logs, users, dashboards, ML, AI, SSO, RBAC, and support. No per-metric or per-volume charges. Costs scale linearly with infrastructure size. Typical deployments achieve 90% cost reduction vs traditional platforms.

SOC 2 Type 2 certified (Security & Availability) with annual recertification. Architecture aligns with GDPR, HIPAA, PCI DSS, DORA, NIS2, and FedRAMP requirements. We provide 6-year audit log retention for HIPAA. Forward Secure Sealing for tamper detection. Complete audit trails of all user actions. Certificate available for vendor reviews. Our architecture satisfies compliance by design - not configuration.

Netdata provides what Prometheus/Grafana provides plus ML, AI, logs, and automation in one platform. Key differences: (1) Real-time: 1-second vs 10-30 seconds, (2) Zero config: Auto-discovery vs manual, (3) ML included: 18 models per metric vs add-ons, (4) No query language: Point-and-click vs PromQL, (5) Performance: 36% less CPU, 88% less RAM at scale in Netdata benchmark, (6) Logs: Direct journal access vs separate Loki. If you want DIY flexibility, Netdata Agent is open source - but with better out-of-box experience.

Active-active Parent clustering with 2+ nodes in circular topology. Automatic failover when Parents fail. Complete data replication between all Parents. Agents buffer locally during Parent outage with automatic replay. Zero data loss architecture. Cloud fails → Agents continue monitoring, local dashboards accessible. Network partition → Each side operates independently. HA is built-in, not an expensive add-on.

Yes. Your observability data already lives on your infrastructure (Agents and Parents). Migration involves deploying on-premises Cloud and reconfiguring Agents to connect to it instead of SaaS. No data migration required because data never centralized. We provide migration guides and professional services. Most organizations complete migration in 1-2 weeks. Zero downtime during transition.

Two options: (1) Model Context Protocol (MCP): Every Agent/Parent is MCP server. Connect your own LLM (Claude, ChatGPT, Gemini). AI queries your infrastructure in real-time via MCP. Data stays local. Free option. (2) Managed AI: Netdata-optimized playbooks with your choice of LLM. Queries stream through Cloud without persistence. Both approaches keep data on-premises during AI analysis.

24/7 availability with dedicated support team. Phone support for critical issues. Custom SLAs available. Implementation assistance including architecture design and deployment. Migration support from existing platforms. Training programs for administrators and users. Regular health checks and optimization recommendations. Direct access to engineering team for complex issues.

Typical timeline: Week 1: Design topology and size infrastructure. Week 2: Deploy Cloud in Kubernetes, install Agents on first 100 nodes. Week 3: Roll out to remaining infrastructure, configure SSO/RBAC. Most organizations achieve full production deployment in 2-3 weeks. We provide professional services to accelerate. Compare to months for traditional platforms requiring extensive configuration and dashboard building.