Dremel Paper (2010)

Back to Key Papers to Read

“Dremel: Interactive Analysis of Web-Scale Datasets” (2010) — Melnik et al.

Introduced columnar storage for nested data with a tree-shaped distributed execution engine. Enabled interactive (seconds-level) SQL queries on petabyte-scale datasets. Became the foundation for BigQuery (GCP). Inspired Apache Drill, Presto/Trino, and Apache Impala.


google-internal papers dremel #2010