PrestoTrino

PrestoTrino is not a single software product. It is a way to refer to the family of open-source distributed SQL query engines that originated from the Presto project, most notably the two active main branches: PrestoDB and Trino. The lineage begins with Presto, developed at Facebook to enable fast analytic queries across large, heterogeneous data stores. In 2020, a fork of the project led to the renaming of PrestoSQL to Trino, while the original PrestoDB branch continued under its own governance. Since then, both projects have evolved independently, offering similar capabilities with their own release cycles, features, and connectors.

Architecture and connectors: Both engines operate on a distributed architecture with a coordinator and multiple workers. They execute queries in parallel and rely on a pluggable connector framework to access data in a wide range of sources, including Hadoop/HDFS and Hive metastore, object stores (such as S3 or ADLS), relational databases (MySQL, PostgreSQL, etc.), NoSQL stores, Kafka, and more. They support ANSI SQL, offer built-in functions and optimizers, and allow extendability through plugins and user-defined functions.

Performance and features: Trino (the renamed PrestoSQL) is known for high-performance, low-latency query execution and a focus on broad community contributions. PrestoDB emphasizes reliability and enterprise-oriented deployments. Both projects provide security features (authentication, authorization, encryption), monitoring, and governance-friendly features, alongside extensive documentation and community support.

Use cases: PrestoTrino engines are used for federated analytics and ad hoc querying across data lakes, warehouses, and multiple data sources, enabling analysts to perform cross-source analysis without moving data excessively.

See also: PrestoDB, Trino, data connectors, distributed SQL.