BigQuery Graph became generally available with native Graph Query Language, petabyte-scale traversals, and no extract-transform-load requirement.

Google Cloud made BigQuery Graph generally available after offering it in preview. The service brings graph capabilities directly into the BigQuery data warehouse, with International Organization for Standardization-standard Graph Query Language alongside SQL.

Graph traversals run natively without requiring extract-transform-load, or ETL, into a separate graph database. Google says the service reaches petabyte scale without the memory bottlenecks associated with a scale-up database and operates under existing row- and column-level security.

BigQuery Graph can call BigQuery ML and AI functions in the same query. Google identified uses including fraud and threat detection, supply-chain digital twins, identity resolution, customer profiles, knowledge graphs, Graph Retrieval-Augmented Generation, network topology, service dependencies, and data lineage.

Thales Cybersecurity Products said it uses property-graph modeling for multi-hop traversals and anomaly clustering in its threat-detection pipeline. It also combines the graph output with Gemini models to generate threat narratives, according to the customer statement included by Google.