PrestoDB

PrestoDB is an open-source distributed SQL query engine designed for running fast analytic queries across large data sets. It enables interactive, ad-hoc analysis by executing SQL queries on data that resides in multiple sources without requiring data movement. Built for a massively parallel processing (MPP) architecture, PrestoDB distributes query planning and execution across a cluster of worker nodes coordinated by a central planner to deliver low-latency results on big data workloads.

Origin and governance

Presto was developed at Facebook (now Meta) and released as open source in 2013. It formed part of the Apache Software Foundation as the Apache Presto project, with PrestoDB commonly used to describe the Apache-licensed implementation. In the late 2010s, the Presto community split into two major lines: the Apache Presto project (often referred to in practice as PrestoDB) and PrestoSQL, a separate community-driven fork that later rebranded as Trino. The two projects have continued to evolve independently, with separate release cycles and feature trajectories.

Architecture and features

PrestoDB follows a master–worker model consisting of a coordinator (the planner) and multiple workers (the executors). It supports ANSI SQL and a range of analytical features, including complex joins, aggregations, and window functions, with performance optimized for large-scale queries. The system uses a catalog and connector framework to access data from diverse sources, enabling queries across data lakes (such as HDFS or object stores like S3 and GCS), relational databases (via connectors), and NoSQL stores.

Data sources and connectors

A core strength of PrestoDB is its pluggable connectors that allow querying data in various storage systems without moving data. Connectors exist for Hive/HDFS, cloud storage, relational databases, and several NoSQL and object stores, enabling federated analytics across heterogeneous data environments.

Usage and ecosystem

PrestoDB has been adopted by numerous organizations to enable fast, interactive analytics over data lake and warehouse ecosystems. It remains a foundational technology in the broader Presto ecosystem, alongside forks and successors such as Trino, which offer alternative features and development paths.