Install prestodb with Homebrew, apt

Project links

• Homepage
• Package manager page

Commands and aliases

• presto
• presto-server

Project history

Presto was created at Facebook in 2012 to provide low-latency interactive analytics over a massive Hadoop data warehouse. It was released publicly under the Apache License in 2013 and developed into a general distributed SQL engine with a pluggable connector architecture.

In 2019, Facebook, Uber, Twitter, and Alibaba helped establish the Presto Foundation at the Linux Foundation. A separate community fork formed around the original creators as PrestoSQL and was renamed Trino at the end of 2020; PrestoDB continued under the Presto Foundation and Linux Foundation governance.

Adoption history

Presto moved from an internal Facebook analytics engine to a widely deployed open-source query layer. Official project materials document use by organizations including Meta, Uber, ByteDance, Adobe, Alibaba, IBM, Intel, Tencent, and Twilio, with workloads spanning interactive analysis, scheduled pipelines, reporting, and lakehouse queries.

The Presto Foundation joined the Linux Foundation in September 2019, giving PrestoDB a vendor-neutral governance structure. Continued investment has included a native C++ execution path based on Velox alongside the established Java engine.

How it is used

Administrators deploy a coordinator and one or more workers, configure each node through files under `etc`, and define data-source connectors through catalog property files. Analysts then submit standard SQL through the CLI, JDBC/ODBC clients, BI tools, or application APIs for federated OLAP workloads over data lakes and other systems.

Why package nerds care

Presto helped normalize the idea that an SQL engine need not own the data it queries. Its connector architecture can federate data lakes, relational databases, NoSQL systems, streaming systems, and other stores behind one distributed SQL surface.

For infrastructure and data-tooling enthusiasts, PrestoDB is also historically important because its community split produced PrestoSQL, later renamed Trino. The two projects share ancestry but have developed independently, making their packaging, connector compatibility, configuration, and operational ecosystems distinct choices.

Timeline

• 2012: Facebook engineers created Presto for low-latency interactive analytics over the company's Hadoop warehouse.
• 2013: Presto was released publicly under the Apache License.
• 2019-09: The Presto Foundation joined the Linux Foundation.
• 2020-12: The separate PrestoSQL fork announced its rename to Trino.
• 2020s: PrestoDB expanded lakehouse connectors and developed a Velox-based native C++ execution path.

Related projects

• Trino descends from the PrestoSQL fork of the original Presto codebase and remains the closest alternative, but its releases, namespaces, connectors, and operational details have diverged.
• Velox underpins PrestoDB's native C++ execution work, providing a reusable execution engine for higher performance and efficiency.

Sources

• https://prestodb.io/
• https://prestodb.io/blog/2025/05/13/fueling-prestos-momentum-and-ibms-growing-role-in-the-open-source-sql-engine/
• https://prestodb.io/docs/current/overview/use-cases.html
• https://prestodb.io/join/
• https://prestodb.io/powered-by-presto/
• https://trino.io/blog/2020/12/27/announcing-trino.html