In a global economy, data is generated and stored all over the world. For a multinational corporation, customer data may reside close to headquarters in the US, while transaction logs are split between Europe and Asia, helping to meet performance, regulatory compliance, and data sovereignty needs. However, this creates a challenge: How do you get a single, unified view of your business if your data is spread across continents?
Historically, the answer has involved complex, time-consuming, and costly ETL pipelines to copy and centralize data before it could be analyzed. This introduces delays, adds complexity, and hinders teams from performing timely, ad-hoc analysis.
Today, we’re changing that with global queries in BigQuery, which lets you query data stored in different geographic locations with a single, standard SQL query – no ETL required. Global queries are available in preview.
What are global queries?
Global queries break down the barriers between datasets distributed across different geographies. Our multinational corporation can now join, union, and analyze all of its data together in a single query, directly from the BigQuery console.
In the background, BigQuery automatically handles the data movement required to execute the query, giving you a seamless, zero-ETL experience for multi-location analytics. BigQuery identifies different parts of the query that must be executed in different regions and runs them accordingly. Next, results of these partial queries are transferred to a selected location when the main query is run (with an optimization attempt to minimize the size of transfer). Finally all parts are combined and the whole query returns the results.
BigQuery user EssilorLuxottica is an early global query adopter, and has seen great results with the new feature:
“At EssilorLuxottica, we place the highest importance on data security and compliance. That’s why information is always stored in the region where it originates. With BigQuery’s global queries, we can seamlessly bring this distributed data together. This allows us to perform cross-region, aggregated analysis without compromising compliance. A powerful way to stay both secure and insight-driven.” – Rubens Ballabio, Customer Data Platform Manager, EssilorLuxottica
At the same time, transferring data across regions brings additional costs and various regulations often prohibit PII data leaving the original location. For these reasons we recognize that transferring data across geographic boundaries requires robust governance. Global queries are designed with security and user control as top priorities. Here’s how:
-
You must explicitly opt in: The global query feature is disabled by default to prevent accidental data transfers or costs. To use it, administrators must explicitly enable global queries for their projects. Additionally, running a global query requires special permission per user or service account.
-
You control the location: You specify the location where a global query is executed, allowing you to control where your data is processed, so you can adhere to data residency and compliance requirements.
-
By respecting governance controls: Global queries respect your existing security posture, including VPC Service Controls, so that data doesn’t move in a way that violates established policies.
Global queries in action
Let’s look at a practical example: a consolidated report of global sales, generated with a single query. Before, this would require a data engineering project. Now, any authorized analyst can get the answer in seconds with a single query:
