BigQuery Omni initially supports Google Cloud and AWS, with Azure to follow.
A key feature is that it saves customers from moving or copying their datasets between clouds, and therefore avoids egress charges.
BigQuery Omni takes advantage of Google's Anthos hybrid and multi-cloud application platform to connect Google Cloud with the other public clouds.
It allows multi-cloud analytics using the BigQuery user interface and APIs, and SQL. The BigQuery Dremel query engine runs – under Google Cloud management – in the region where the data is stored.
"For our customers, data is no longer one room in the house – it's the foundation. However, moving data across different clouds is both cumbersome and expensive," said Google Cloud general manager and vice president of engineering Debanjan Saha.
"With BigQuery Omni, customers will get a multi-cloud analytics solution that enables them to gain critical data insights, in one unified experience. We continue to invest in multi-cloud in an effort to democratise access to the best technologies for our customers, no matter what cloud provider they're using today."
BigQuery Omni is initially available in private alpha for AWS S3, with Azure support coming soon. Interested organisations can apply here to join the alpha program.