BigQuery is [really fast]((https://cloud.google.com/blog/products/bigquery/anatomy-of-a-bigquery-query). In order to achieve this, it relies on powerful technologies as:
- Dremel, Google’s execution engine: Dremel orchestrates a SQL statement into an execution tree and re-assembles the results.
- Colossus, Google’s distributed filesystem: Each Google datacenter has its own colossus cluster and each Colossus cluster has enough disks to give every BigQuery user multiple disks.
- Borg, Google’s cluster management system: Borg cluster runs of dozen of thousands of machines and hundred thousands of cores.
- Jupiter, Google’s networking infrastructure: impressive network allowing to distribute large workloads.
- Capacitor, BigQuery’s columnar storage format which encodes data at column level to optimize queries.
References
BigQuery under the hood: Google’s serverless cloud data warehouse. [Online]. Available at: https://cloud.google.com/blog/products/bigquery/bigquery-under-the-hood [Accessed 4 April 2025].