Grafana Hosting

How Grafana works

Grafana connects to supported data sources and sends queries when users open dashboards, explore data, or evaluate alert rules. Panels transform query results into graphs, tables, gauges, logs, traces, and other visualisations. Dashboards group related panels, variables change query context, and permissions control who can view or edit resources.

The application does not create useful observability data by itself. Metrics, logs, traces, profiles, database records, and business data must exist in a connected source such as a monitoring backend or database. Query speed, retention, sampling, data quality, and ingestion cost remain properties of those systems, even when Grafana presents the results clearly.

Key Grafana features

Explore provides an ad hoc workspace for querying and correlating supported data sources outside a fixed dashboard. Alerting can evaluate rules and route notifications through configured contact points, but those delivery systems and credentials must be supplied separately. Platform application email is disabled, so email contact points cannot send from this installation.

The catalogue generates an administrator password, disables public user sign-up, and persists /var/lib/grafana. No metrics backend, log store, trace system, cloud account, agent fleet, data-source credentials, or commercial Grafana service is included. Administrators must add trusted data sources and use least-privilege accounts for each connection.

Grafana vs Datadog

Grafana in this package is a self-hosted interface that queries external sources and stores dashboards, users, alert definitions, and configuration locally. Datadog is a provider-operated observability platform that ingests and stores telemetry, supplies managed agents and integrations, and sells monitoring, security, incident, and analytics products through commercial plans.

Datadog may suit organisations that want a maintained telemetry platform, broad vendor integrations, unified billing, and less responsibility for operating the observability stack. Grafana is attractive when teams already control their data sources, prefer their own visualisation server, and accept responsibility for integrations, retention backends, and alert delivery channels.

Who uses Grafana

Site-reliability teams build service dashboards, developers investigate logs and traces, database administrators chart query statistics, and product teams visualise business metrics. A single Grafana installation can present several approved data sources while keeping editing access separate from read-only viewers.

Grafana is not a replacement for instrumentation, telemetry collection, data retention, incident response, or source-system permissions. A polished dashboard can still be misleading when queries, labels, sampling, or source data are wrong.

Self-hosting Grafana: requirements and cost

Grafana demand depends on active users, dashboard and panel count, query frequency, alert evaluation, plug-ins, transformations, and the latency and size of data-source responses. The catalogue records PostgreSQL and MariaDB as Not required and persists Grafana’s internal data directory. External telemetry storage usually drives the broader observability cost rather than this interface alone.

On AvaHost, Grafana uses Plan 1 at €2. The hosted Grafana package includes one-click deployment, a custom domain with automated HTTPS, automatic application updates, and scheduled backups. No metrics, log, trace, or profile backend is bundled. Customers configure their own permitted data sources, agents, credentials, and contact points. Application email is disabled, so email alert delivery is unavailable. Use least-privilege source accounts, review plug-ins before enabling them, and validate important dashboards and alert rules after data-source or application changes.