Grafana Assistant Now Queries 30+ Data Sources via Natural Language

Grafana Labs has expanded Grafana Assistant to query and correlate data across 30+ different data sources through natural language, addressing a core challenge in modern observability: fragmented operational data spread across disconnected platforms. The update enables operators, developers, and SREs to investigate incidents, generate queries, and troubleshoot distributed systems without switching between monitoring tools.

The expansion now includes cloud platforms, databases, observability backends, issue trackers, and infrastructure monitoring systems—including Snowflake, Oracle, Elasticsearch, Dynatrace, Honeycomb, MongoDB, Zabbix, and Jira. Rather than forcing users to learn PromQL, LogQL, SQL, or TraceQL, Grafana Assistant translates natural language into appropriate queries while respecting existing permissions and role-based access controls.

The move reflects a broader competitive shift in observability: vendors like Datadog (Bits AI), Dynatrace (Davis AI), Splunk, and New Relic are all competing on AI capabilities rather than telemetry collection alone. For architects, Grafana's multi-data-source aggregation via LLM represents the evolving pattern: as systems become more complex and heterogeneous, single-tool observability dies; instead, the advantage goes to platforms that can bind multiple backends into a unified conversational query layer with proper governance.